^eDS7^
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
REGION 7
11201 Renner Boulevard
Lenexa, Kansas 66219
"'i PRO^
? * JUN 2016
MEMORANDUM
SUBJECT: Cave Air Risk Estimates and Levels of Concern
La Jolla Spring Cave Complex
Oak Grove Village Well Site
Franklin County, Missouri
FROM: Kelly Schumacher C"—*-
Toxicologist
ENST/EDAB
TO:
Tonya Howell
Remedial Project Manager
SUPR7MOKS
As requested, we have quantified the potential risks to human health from exposures to trichloroethylene
in air within the La Jolla Spring Cave, based on data collected from 2002 through 2016, and we have
derived levels of health concern for TCE in air protective of all current and potential future human
receptors at the cave. The first objective supports the decision to close the cave and implement control
measures, while the second objective provides the criteria for the re-opening of the cave to workers and
visitors. If you have any questions or need further assistance, please contact me at x7963.
30286191
III
Superfund
Printed on Recycled Paper
-------
Cave Air Risk Estimates and Levels of Concern
La Jolia Spring Cave Complex
Oak Grove Village Well Site
Franklin County, Missouri
1.0 INTRODUCTION
1.1 Background
The Oak Grove Village Well Site, added to the U.S. Environmental Protection Agency's National
Priorities List in September 2002, is located in Oak Grove Village and Sullivan in Franklin County,
Missouri. The primary contaminant of concern is trichloroethylene, which was first detected in the Oak
Grove Village municipal drinking water wells in 1986. Sources include the former TRW/Ramsey
Facility and the closed Sullivan Landfill. Site groundwater contamination is now widespread,
influenced by well pumping and karst geology.
The Site is subdivided into two Operable Units: OU1 incorporates the contamination in the area of the
Oak Grove Village wells, including impacted wells in Sullivan, and OU2 incorporates the Sullivan
Landfill, as well as nearby wells and springs impacted by contamination from the landfill. Both operable
units include the La Jolla Spring Cave Complex, which is located approximately two miles east of the
Sullivan Landfill and approximately four miles northeast of the TRW/Ramsey facility in Sullivan,
Missouri. La Jolla Spring serves as a drainage point for the Sullivan and Oak Grove Village area,
flowing at approximately 4 cubic feet per second for more than one-half mile through a privately-owned
show cave.
Concentrations of trichloroethylene detected in spring water samples collected from the La Jolla Spring
since the early 1990s have ranged from less than 1 to 12.6 pg/L. Between October 2002 and October
2003, three rounds of air sampling were conducted within the cave to examine potential transport of
volatile constituents from the spring water to the cave air. During these sampling events, the
concentrations of TCE detected in cave air ranged from 100 to 1,700 pg/m3. After the October 2003
sampling event, a natural vent was opened in an attempt to lower cave air concentrations. Five additional
rounds of sampling conducted between February 2004 and January 2007 showed a reduction in TCE air
concentrations ranging from non-detect to 180 pg/m3, which were below levels of health concern at that
time. In September 2011, new, more stringent toxicity values for TCE were published (USEPA, 201 la).
Cave air sampling resumed in 2013; the highest concentration of TCE detected in the cave air since
sampling resumed was 660 pg/m3 in October 2015. The cave was closed by the owner in March 2016.
Since then, TCE levels have decreased in portions of the cave, as a result of the recently implemented
control measures such as air scrubbers and airlock doors.
1.2 Objectives
The objectives of this document are to 1) quantify the potential risks to human health from exposures to
trichloroethylene in air within the La Jolla Spring Cave and 2) to derive levels of health concern for TCE
in air protective of all potential future human receptors at the cave. The first objective supports the
decision to close the cave and implement control measures, while the second objective provides the
criteria for the safe re-opening of the cave to workers and visitors.
2 of 21
-------
Table 1. TCE Concentrations in La Jolla Spring Cave Air (ng/m3).
Date
Gift
Shop
Cafeteria
Kitchen
Ball-
room
Theater
Loot
Rock
Lassie
Room
Jungle
Room
Riverboat
Tour
Shelter
Ticket
Counter
Fallout
Shelter
Mud
Alley
Bat Cave
Conference
Room
Site
Bll
Wine
Room
Pendulum
Between
Airlock
Doors
Motel
Amphi-
theater
10/7/2002
n/a
n/a
n/a
100
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
4/30/2003
n/a
n/a
n/a
1,000
n/a
n/a
1,100
1,400
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
10/8/2003
840
n/a
n/a
1,100
340
1,500
900
1,700
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
2/3/2004
5.2
n/a
n/a
4.9
5.1
7.8
18
23
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
1/19/2005
7.8
n/a
n/a
5.2
7.8
9.8
22
24
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
10/2005
15
n/a
n/a
26
25
85
94
87
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
2/2006
39
n/a
n/a
47
13
78
120
180
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
1/25/2007
7.5
n/a
n/a
6.7
U
13
22
30
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
3/13/2013
1.2 U
n/a
n/a
5.16
4.19
7.09
18.2
22.4
1.2 U
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
8/27/2013
97.5
n/a
n/a
138
0.43 U
164
117
177
0.483
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
1/13/2014
5.37
n/a
n/a
20
12.3
40.1
48.8
64.8
0.43 U
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
3/24/2014
1.13
n/a
n/a
4.51
4.35
14
21.4
17.4
0.43 U
4.83
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
6/2/2014
125
n/a
n/a
169
1.29
220
196
252
1.18
183
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
8/5/2014
143
n/a
n/a
133
0.698
224
176
240
0.806
193
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
2/24/2015
1.56
n/a
n/a
8
4.83
9.83
11
14.8
0.86 U
n/a
10.4
n/a
n/a
0.43 U
n/a
n/a
n/a
n/a
n/a
n/a
2/25/2015
1.67
n/a
n/a
7.68
5.32
14.1
12.4
16.5
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
4/28/2015
n/a
n/a
n/a
44.5
31.8
79.1
79.7
111
0.43 U
n/a
76.8
117
0.43 U
7.57
n/a
n/a
n/a
n/a
n/a
n/a
6/22/2015
100
n/a
n/a
127
0.43 U
114
120
147
0.43 U
n/a
n/a
n/a
0.43 U
125
130
n/a
n/a
n/a
n/a
n/a
6/24/2015
9.78
n/a
n/a
24
n/a
n/a
n/a
137
n/a
n/a
n/a
n/a
n/a
142
n/a
n/a
n/a
n/a
n/a
n/a
7/28/2015
110
n/a
n/a
170
3.6
230
220
250
1.1 U
n/a
230
18
n/a
220
210
28
170
220
n/a
n/a
8/25/2015
14
n/a
n/a
43
13
300
260
300
1.1 U
n/a
n/a
45
n/a
250
260
n/a
n/a
270
n/a
n/a
9/17/2015
19.4
n/a
n/a
56.2
4.46
158
n/a
207
n/a
n/a
n/a
n/a
n/a
n/a
119
n/a
n/a
n/a
n/a
168
9/22/2015
15.3
n/a
n/a
55
17.9
181
n/a
241
0.43 U
n/a
n/a
n/a
n/a
n/a
129
n/a
n/a
n/a
n/a
159
10/21/2015
51
n/a
n/a
150
45
590
n/a
660
1.1 U
n/a
n/a
97
n/a
410
170
n/a
n/a
540
n/a
n/a
11/30/2015
6.5
n/a
n/a
43.3
33.4
36.7
n/a
51.8
n/a
n/a
n/a
n/a
n/a
1.02
52.9
n/a
n/a
n/a
n/a
n/a
2/17/2016
3.38
n/a
n/a
20
14.6
n/a
n/a
19.6
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
2/22/2016
5.69
n/a
n/a
36.3
28.4
22.8
n/a
29.2
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
0.43 U
0.43 U
3/8/2016
64
n/a
n/a
210
59
420
n/a
490
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
3/21/2016
4.2
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
3/23/2016
5.6
6.9
n/a
n/a
n/a
n/a
n/a
460
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
3/27/2016
1.88
1.29
0.913
n/a
n/a
n/a
n/a
38.4
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
4/4/2016
3.2
3
n/a
5.4
4.4
n/a
n/a
390
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
4/11/2016
2.5
2.6
n/a
n/a
4.3
n/a
n/a
100
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
Bold: Compound was detected. U: Compound was not detected above the reporting limit, n/a: Sample was not collected.
Page3 of 21
-------
2.0 DATA EVALUATION
Table 1 presents the concentrations of TC-E detected in air samples collected using 24-hour Summa
canisters deployed within and outside the La Jolla Spring Cave from 2002 through 2016. The samples
were analyzed for a variety of volatile organic compounds, but compounds other than TCE were either
not detected or were detected below levels of health concern. In the few samples where TCE was not
detected, the laboratory reporting limits ranged from 0.43 to 1.2 [ig/m3, which are adequately low to
detect levels of concern for human health.
During the first eight monitoring events, in 2002 - 2007, air samples were collected from the cave gift
shop, Ballroom, Theater, Loot Rock, Lassie Room, and Jungle Room, all of which are located along the
full tour route. Beginning in 2013, additional locations along the cave tour route were sampled (e.g.,
former ticket counter location, Wine Room), as well as points outside the cave (e.g., Riverboat Tour
Shelter, motel) and branches of the cave that are not contacted by workers or visitors (e.g., Fallout
Shelter, Site B11, Bat Cave, Mud Alley, Conference Room, Amphitheater). For completeness, all
available cave air data through April 11, 2016, are included in Table 1, but only those sampling
locations along the cave tour route were used to estimate potential health risks, as described in Section 5.
The data show significant seasonal and spatial variability. Seasonally, TCE concentrations inside the
cave have been lowest during the winter, when outside air temperatures are low, and higher during the
summer, when outside temperatures are high. However, some of the highest TCE levels have been
detected in the spring and fall, when the air temperatures inside and outside the cave are similar. The
TCE concentrations within the cave appear to be associated with the cave air flow, which is influenced
by outside air temperature and humidity levels. Spatially, the highest TCE levels have been detected at
the sampling locations deepest within the cave (e.g., Loot Rock, Lassie Room, Jungle Room), and the
lowest TCE levels have been detected closest to the cave entrance (e.g., gift shop). Outdoor ambient
TCE concentrations, measured at the Riverboat Shelter and motel, have generally been non-detect or
less than 1 jag/m3, regardless of the season.
3.0 EXPOSURE
3.1 Exposure Pathways
The EPA evaluates complete exposures to contaminants in soil, water, air, and/or other environmental
media via ingestion, inhalation, and/or dermal contact. The only complete exposure pathway evaluated
in this document is inhalation of trichloroethylene in cave air. Since the water from the La Jolla Spring
is not used as a potable source, the direct ingestion pathway is incomplete even though TCE is present in
the spring. Also, because contact with the La Jolla Spring itself is limited, any incidental ingestion or
dermal absorption of the spring water or sediment is considered de minimus.
3.2 Exposure Scenarios
The two main types of human receptors at the La Jolla Spring Cave are visitors and employees. Visitors
include adults and children who attend guide-led walking tours of the cave, and/or who spend time in the
gift shop and cafeteria. Cave employees include full- and part-time tour guides, as well as workers
stationed in the gift shop and cafeteria. Prior to March 2016, tour guides and visitors followed the tour
schedule shown in Table 2.
Page 4 of 21
-------
Table 2. La Jolla Spring Cave Tour Schedule, Prior to March 2016.
Description
Time (minutes)
Intro/Log Cabin
5 Minutes
Travel Time
1 Minute
Ballroom
3 Minutes
Travel Time
1 Minute
Gun Powder
4 Minutes
Travel Time
1 Minute
First Division
4 Minutes
Travel Time
2 Minutes
Loot Rock
4 Minutes
Travel Time
5 Minutes
Lassie Room
4 Minutes
Travel Time
3 Minutes
Jungle Room
3 Minutes
Travel Time
3 Minutes
Mirror River
2 Minutes
Travel Time
6 Minutes
Onyx Mountain
3 Minutes
Travel Time
2 Minutes
Echo Room
3 Minutes
Travel Time
2 Minutes
Wine Room
3 Minutes
Travel Time
2 Minutes
Theater Room
10 Minutes
Travel Time
4 Minutes
Total
80 Minutes
Note: The shaded portion represents the tour time spent past the airlock doors.
3.2.1 Cave Visitors
It was assumed that adult and children visitors participate in one cave tour during any 24 hour period.
Prior to closure of the cave in March 2016, each tour was scheduled to last 80 minutes. However, it is
possible that tours could have lasted longer than scheduled, for instance, if visitors had many questions.
Therefore, under a reasonable maximum exposure scenario, which is the highest exposure that is
reasonable expected to occur at a site (USEPA, 1989), tours were assumed to last 90 minutes. Of this
time, approximately 30 minutes were assumed to be spent in the Ballroom area closer to the front of the
cave near the gift shop, 30 minutes were assumed in the Loot Rock/Lassie Room/Jungle Room area
deeper within the cave past the airlock doors, and 30 minutes were assumed in the Theater/Wine Room
area located at the top of the stairs from the Ballroom area, closer to the ground surface. These
allocations are based on the amount of time scheduled for each area of the cave on the tour, as shown in
Table 2. Note that air samples (Table 1) were not collected from all of these areas. Visitors were also
assumed to spend 30 minutes in the gift shop/cafeteria, which is located within the cave entrance,
spending a total of 2 hours at the facility.
3.2.2 Tour Guides
The workday for a full-time tour guide is 8 hours per day, plus a 30 minute lunch break. Prior to March
2016, guides tended to lead two or three tours per day, following the schedule shown in Table 2.
5 of 21
-------
However, during the busiest parts of the year, such as the 4th of July holiday weekend, guides led up to
four tours per day, which was considered a reasonable maximum exposure scenario.
3.2.3 Gift Shop/Cafeteria Employees
The workday for full-time gift shop and cafeteria employees is 8 hours per day, plus a 30 minute lunch
break, all of which was assumed to be spent in the gift shop and/or cafeteria.
4.0 TOXICITY
Human health toxicity values for trichloroethylene were published in 2011 by the United States
Environmental Protection Agency's Integrated Risk Information System program (USEPA, 201 la). The
EPA's final toxicological review incorporates comments by the U.S. National Academy of Sciences
(NRC, 2006), two U.S. EPA Science Advisory Boards (USEPA, 2002 and 201 lb), the Executive Office
of the President (OMB, 2009 and 2011), the U.S. Department of Defense (DOD, 2009a, 2009b and
2011), the National Aeronautics and Space Administration (NASA, 2009 and 2011), internal Agency
reviewers, and the public, among others. The Halogenated Solvents Industry Alliance, Inc., which
represents the interests of TCE manufacturers and producers, submitted a Request for Correction of the
TCE IRIS assessment (HSIA, 2013), which was denied by the EPA's Acting Assistant Administrator
Lek Kadeli (USEPA, 2015). The HSIA then submitted a Request for Reconsideration (HSIA, 2015),
which was also denied by the EPA (USEPA, 2016). The EPA found the Requests "directly contrary to
the SAB's conclusions and recommendations, such that to accept HSIA's RFC/RFR would require EPA
to reject SAB's advice" (USEPA, 2016).
The EPA's Office of Land and Emergency Management recognizes an IRIS assessment as the official
Agency scientific position regarding the toxicity of a chemical based on the data available at the time of
the review (USEPA, 2003). As such, IRIS is the preferred source of human health toxicity values used to
evaluate risks at Superfund and RCRA hazardous waste sites. In accordance with Directive 9285.7-53
(USEPA, 2003), the 201 1 IRIS TCE toxicity values will be used to evaluate risks and derive levels of
health concern at the Oak Grove Village Well Site until the 2011 values are either revised or rescinded
by the IRIS program.
4.1 Non-Carcinogenic Health Effects
In general, the EPA assumes that a dose or exposure level exists below which adverse non-carcinogenic
health effects will not occur (USEPA, 1989). Below this threshold, it is believed that exposure to a
chemical is tolerated without adverse effects. Adverse health effects occur only when physiologic
protective mechanisms are overcome by exposure to doses or concentrations above the threshold. For
chronic toxicity values, the first adverse effect (or its known precursor) that occurs to the most sensitive
species as the dose rate of an agent increases, regardless of the exposure duration, is designated the
critical endpoint. The dose or exposure at which the critical endpoint is observed is the point of
departure. Uncertainty factors reflecting limitations of the data used are applied to the point of departure
to derive the inhalation reference concentration, which is an estimate (with uncertainty spanning perhaps
an order of magnitude) of a continuous inhalation exposure to the human population (including sensitive
subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime.
The 2011 Scientific Advisory Board panel recommended that, "The two endpoints for immune effects
from Keil et al. (2009) and the cardiac malformations from Johnson et al. (2003) should be considered
the principal studies supporting the RfC" (USEPA, 201 lb). The panel considered the immune effects
6 of 21
-------
and cardiac malformations co-critical endpoints (USEPA, 201 lb). In accordance with the SAB panel
recommendations, the IRIS program based the TCE chronic reference concentration of 2 ng/m3 on these
two co-critical endpoints, each of which can support the RfC independently: autoimmune disease
following chronic exposure in adults (0.00033 ppm, or 1.8 |ag/m3) and heart defects following exposure
during early pregnancy (0.00037 ppm, or 2.0 |ig/m3). The RfC is also supported by nephrotoxicity
(kidney effects) following chronic exposure in adults (0.00056 ppm, or 3.0 |ig/m3). Following
publication of these values, the IRIS program further addressed the developmental cardiac effects in
"TCE Developmental Cardiac Toxicity Assessment Update1' (USEPA, 2014a).
Chronic exposure to TCE poses a potential human health hazard to the central nervous system, kidneys,
liver, immune system, and male reproductive system. As mentioned above, adult immunotoxicity is
considered a co-critical endpoint, at a slightly lower concentration than that associated with cardiac
defects. Overall, the IRIS program concluded that "the human and animal studies of TCE and immune-
related effects provide strong evidence for a role of TCE in autoimmune disease and in a specific type of
generalized hypersensitivity syndrome" (USEPA, 201 la). Kidney toxicity was considered a supporting
endpoint, with high confidence found in multiple lines of evidence in both human and animal studies.
Exposures to TCE during pregnancy are associated with many forms of developmental toxicity,
including spontaneous abortions, decreased growth, developmental neurotoxicity, developmental
immunotoxicity, and birth defects. However, the critical developmental endpoint is cardiac
malformations. The primary types of heart defects observed with TCE exposures include atrial and
ventricular septal defects, which are holes in the wall (septa) between the top two chambers (atria) or
bottom two chambers (ventricles) of the heart, and pulmonary and aortic valve stenoses, which are
thickened or fused heart valves that do not properly open and/or close and may leak blood. The critical
window of susceptibility for these types of defects is an approximate three week period ( i.e.,
valvuloseptal morphogenesis, or the period in which major cardiac morphogenic events such as heart
valve formation occur) approximately four to seven weeks after conception, early in the first trimester of
human pregnancy (Dhanantwari et cil., 2009). The type and severity of the resulting cardiac
malformation or malformations depends on the timing and level of exposure to TCE within this
approximate three week period. Exposures that clear the body before this period do not impact the heart
valves and septa, because they have not yet begun to form. In humans, TCE and most of its metabolites
are eliminated within a week of exposure (USEPA, 201 la).
4.2 Carcinogenic Health Effects
When evaluating the potential carcinogenicity of a chemical, the EPA generally assumes that any
exposure to a chemical will increase an individual's risk of developing cancer. In other words, there is
no threshold below which the probability of developing cancer is zero. The EPA evaluates
carcinogenicity in two parts. First, the Agency evaluates all available scientific information and assigns
a weight-of-evidence classification based on a compound's potential to cause cancer in humans. Second,
a toxicity value is derived to define the quantitative relationship between dose or concentration and
carcinogenic response. For inhalation exposures, this value is known as the inhalation unit risk. The IUR
is a generally plausible upper-bound estimate of the increased probability of developing cancer
following a lifetime of exposure. This value is used to estimate the increased risk of developing cancer
from inhalation of potentially carcinogenic chemicals.
Following the EPA's Guidelines for Carcinogen Risk Assessment (USEPA, 2005), the IRIS program
has evaluated the carcinogenic potential of TCE and has classified it as "carcinogenic to humans" by all
routes of exposure. This conclusion is based on convincing evidence of a causal association between
7 of 21
-------
TCE exposure in humans and kidney cancer, strong evidence of non-Hodgkin's lymphoma, and more
limited evidence of liver and biliary tract cancer. The inhalation unit risk for TCE, based on these cancer
types, is 4.1E-06 (fig/m3)"1. The available human data also suggest an association between TCE
exposure and bladder, esophageal, prostate, cervical, and breast cancer, as well as childhood leukemia.
Sufficient evidence supports a mutagenic mode of action for TCE-induced kidney tumors in humans, but
modes of actions have not been established for the other TCE-induced cancer types.
5.0 RISK CHARACTERIZATION
5.1 Equations
The EPA's Superfund program characterizes potential human health risks using standardized equations
that combine toxicity values with exposure parameters because risk is a function of both hazard and
exposure. Typically, the EPA's standard default exposure parameters for chronic scenarios, published in
OSWER Directive 9200.1-120 (USEPA, 2014b), are used. However, exposure assessments must take
into account the time scale related to the specific biological response (NRC, 1991). This means that
exposure parameters selected to evaluate risks and/or develop levels of concern for a given chemical and
scenario should correspond as closely as possible with the exposure period used to develop the toxicity
value. For example, time-weighted average exposures over a lifetime have little relevance for a
developmental toxin if the adverse effects could only occur following exposure during a particular stage
of development (USEPA, 1992). Below, the equations and parameters used to evaluate non-cancer
hazard quotients and cancer risks associated with exposures to TCE in the cave air are presented.
5.1.1 De velopm en tal Hazard Quotien ts
The toxicity values considered protective for a lifetime of exposure to TCE are partly based on non-
cancer health effects resulting from less-than-lifetime exposures. As discussed in Section 4.1, one of the
two co-critical endpoints that serves as the basis for the TCE RfC is heart defects. This effect can only
occur when the fetus is exposed during the period of heart development. Therefore, the EPA's standard
default exposure parameters for chronic exposures are invalid for estimating hazard quotients
representing the potential for cardiac defects associated with TCE exposures and for deriving TCE levels
of concern that are protective of developmental endpoints. To select appropriate less-than-lifetime
exposure parameters that may be used to characterize these hazards and derive levels of concern, the
critical exposure period of concern for TCE-related heart malformations must first be identified.
"[F]or developmental toxic effects, a primary assumption is that a single exposure at a critical time in
development may produce an adverse developmental effect, i.e., repeated exposure is not a necessary
prerequisite for developmental toxicity to be manifested" (USEPA, 1991). The EPA's Risk Assessment
Guidance for Superfund Part A (USEPA, 1989) directs the use of a day or a single exposure incident to
assess the potential risks of adverse developmental effects. Following this guidance, it is assumed that a
single exposure to TCE at any time during the approximate three week period of valvuloseptal
morphogenesis could result in one or more of the types of heart malformations described in Section 4.1.
Thus, the critical exposure period of concern used to evaluate the potential for heart defects is one day.
This 24-hour exposure period has previously been used by the EPA to evaluate acute hazards associated
with TCE in the final, peer-reviewed TSCA Work Plan Chemical Risk Assessment (USEPA, 2014c).
The EPA's Risk Assessment Guidance for Superfund Part F, Supplemental Guidance for Inhalation Risk
Assessment (USEPA, 2009) indicates the exposure concentration (EC) that should be used to evaluate
8 of 21
-------
risks and derive levels of concern for acute endpoints is equivalent to the concentration detected in air
(CA), as shown in Equation 1.
ecO=ca<£)
However, visitors and employees at the La Jolla Spring Cave are only exposed to TCE in cave air for a
portion of any given 24-hour period. Moreover, TCE concentrations are known to vary throughout the
cave, as shown in Table 1. Thus, Equation 1 must be modified to account for multiple exposures over a
24-hr period, resulting in a time-weighted average exposure concentration. The TWA exposure
concentration can be calculated using Equation 2,
ECj = H"=i(C/4[ ¦ £T,)/ATj (2)
where: EQ (jig/m3) = time-weighted average exposure concentration for exposure period j;
j (hrs) = exposure period of concern (24 hours)
CAj (ng/m3) = TCE concentration in air in microenvironment (ME) i;
ETi (hours/day) = exposure time spent in ME i;
ATj (hours) = averaging time for the period of concern (24 hours)
Based on the available sampling locations and information regarding the cave tours, five
microenvironments with distinct TCE concentrations were defined: MEi) away from the cave, ME2) the
gift shop/cafeteria, ME3) the ballroom area (representative of the log cabin, ballroom, gunpowder, and
first division), ME4) the Loot Rock/Lassie Room/Jungle Room area (representative of Loot Rock, Lassie
Room, Jungle Room, and Mirror River), and ME5) the theater area (representative of Onyx Mountain,
Echo Room, Wine Room, and the Theater Room). The three types of human receptors are assumed to
have spent different amounts of time (ET) in each of these microenvironments, prior to March 2016.
As described in Section 3.2.1, visitors are assumed to have spent 22 hours away from the cave, 0.5 hours
in the gift shop/cafeteria, 0.5 hours in the ballroom area, 0.5 hours in the Loot Rock/Lassie Room/Jungle
Room area, and 0.5 hours in the theater area, over a 24-hour period. Thus, the 24-hour TWA exposure
concentration for visitors can be calculated using Equation 3.
... . (CA^-22 hrs) + (CA2-0-5 hrs) + (CA3-0.5 hrs) + (CA4-0.5 hrs) + (CA5-0.5 hrs)
Visitor LL7a = (3)
24 hrs v '
As described in Section 3.2.2, tour guides are assumed to have led four 1.5-hour tours per day, with the
remainder of their 8-hour workday and 30-minute lunch break spent in the gift shop/cafeteria. Thus, they
are assumed to have spent 15.5 hours away from the cave, 2.5 hours in the gift shop/cafeteria, 2 hours in
the ballroom area, 2 hours in the Loot Rock/Lassie Room/Jungle Room area, and 2 hours in the theater
area, over a 24-hour period. The 24-hour TWA exposure concentration for tour guides can be calculated
using Equation 4.
_ . , __ (Ci4i-15.5 hrs) + (CA-.-2.S hrs) + (CA3-2 hrs) + (CA4-2 hrs) + (CA5-2 hrs) ..
Tour Guide EC24 — 1 (4)
24 hrs
Gift shop and cafeteria employees are assumed to have spent their entire 8-hour work day and 30-minute
lunch break in the gift shop/cafeteria, as discussed in Section 3.3.3. Thus, they are assumed to have
spent 15.5 hours away from the cave, 8.5 hours in the gift shop/cafeteria, and 0 hours in
microenvironments 3, 4, and 5. The gift shop/cafeteria employee 24-hr TWA exposure concentration
can be calculated using Equation 5.
9 of 21
-------
„.r r ¦ (CA^ IS.S hrs) + (CA2'8.5 hrs) + (CA3-0 hrs) + (CAi-0 hrs) + (CAr-0 hrs)
Gift Shop/Cafeteria EC24 - —1 24^ ^
Non-cancer hazard quotients for heart defects can then be derived using Equation 6, where HQ24 is the
developmental hazard quotient; EC24 is the 24-hr time-weighted average exposure concentration
calculated using Equations 3, 4, or 5; and the RfC is 2 |ig/m3, as described in Section 4.1. As shown in
Equation 6, a hazard quotient is the ratio of the exposure to the non-cancer toxicity value. Thus, an HQ
greater than 1 means that the exposure is greater than the RfC, which is unacceptable for the non-cancer
health effect of concern.
(6)
Equation 6 can be combined with Equation 3, 4, or 5 to calculate the developmental hazard quotients'
(HQ24) for each receptor, as follows.
t/¦ ¦*- - un (CA-i-22 hrs) + (CA2-0.5 hrs) + (CA3-0.5 hrs) + (CA4-0.5 hrs) + (CA5-0.5 hrs)
V ISltOT H Q24 — (ua\ (')
24 hrs-Rfc(^)
T- /- ¦ j rm (CA^IS.5 hrs) + (CA2'2.5 hrs) + (CA3-2 hrs) + (CA4-2 hrs) + (CAs-2 hrs)
I our Guide HU2a = ,ua\ (8)
24 hrs-Rfc(^)
„.r, c . . ,,(Ci4i-15.5 /J7"s) + (C/42-8.5 h7\s) + (Ci43-0 /irs) + (Ci44-0 ftrs) + (Cj4c-0 hrs)
Gift Shop/Cafeteria HQ24 = -—1 —— —— —5 ——5 (9)
' Vm3/
5.1.2 Non-Cancer Hazard Quotients for Chronic Exposure
Autoimmune disease, a co-critical endpoint upon which the TCE RfC is based, and kidney toxicity, the
supporting endpoint, are both health effects associated with chronic or long-term exposures. The
standardized equation used to evaluate chronic non-cancer hazard quotients is shown in Equation 10,
and the parameters are discussed in Section 5.2 and defined in Table 3.
CA(m.ET(^){l^).EF(^).ED(years)
Ij V771 / \dayJ \24 hrs) \yearj ^ _ /1 n\
"^chronic ~~ . n--/*u7\ U
ATnc.chronic(-dQ-ys)'RfC\^m3 j
The above equation must be modified to account for multiple exposures to different TCE concentrations
in various parts of the cave. This can be accomplished by replacing the single concentration and
exposure time terms with the 24-hour time-weighted average exposure concentration, as shown below in
Equations 11 and 12. Note that it is only appropriate to calculate chronic non-cancer hazard quotients for
those receptors with long-term exposures, which includes the tour guides and gift shop/cafeteria
workers, not the visitors.
Tour Guide EC2J^)-EF(^^XED(years)
Tour Guide HQchronic = - yea' , . (11)
ATncxhronic(days)Rfc(^)
Gift Shop/Cafeteria EC2J1±^)-EF(^:^-)-ED(years)
Gift Shop/Cafeteria HQchronic = — 7,\ (12)
ATnc,chronic(.days) RfC\^^3 j
10 of 21
-------
In reality, hazard quotients for chronic endpoints should be calculated using the average daily dose for
the entire exposure duration, the default of which is 25 years for a worker scenario. However, most of
the available cave air samples have been collected during the last three years, with limited data collected
from 2002 through 2007 and no data collected prior to 2002. To address this database uncertainty, non-
cancer hazard quotients for chronic exposure were estimated using the individual 24-hour time-weighted
average exposure concentrations calculated for each sampling date, as well as the arithmetic mean of all
the 24-hour TWA exposure concentrations calculated for sampling dates between March 13, 2013, and
March 8, 2016, prior to cave closure. Although an average value based on the last three years of data
could underestimate true exposures and associated long-term health hazards based on fluctuations in the
cave air TCE concentrations over the last 25 years, the more recent data provides the best estimate of
seasonal variability.
5.1.3 Cancer Risks
TCE is classified "carcinogenic to humans,1' based on kidney cancer, non-Hodgkin's lymphoma, and
liver and biliary tract cancer, as discussed in Section 4.2. The standardized equation used to evaluate
excess individual lifetime cancer risks is shown in Equation 13, and the parameters are discussed in
Section 5.2 and defined in Table 3.
ATcancer(rfays)
Like Equations 1 and 10, the above equation must be modified to account for multiple exposures to
different TCE concentrations in various parts of the cave. This can be accomplished by replacing the
single concentration and exposure time terms with the 24-hour time-weighted average exposure
concentration, as shown below in Equations 14 and 15. Note that it is only appropriate to calculate
excess individual lifetime cancer risks for those receptors with long-term exposures, which includes the
tour guides and gift shop/cafeteria workers, not the visitors.
Tour Guide EC2J^)-EF(^1)-ED(vears)-IUR(^) *
Tour Guide CR = ——Ky, ' (14)
ATCancer(.day s)
Gift Shop/Cafeteria EC2M}EF(^\ED(years)-IUR(£§) *
Gift Shop/Cafeteria CR — (15)
' J ATcancer{days) V '
Consistent with how non-cancer hazard quotients for chronic exposure were evaluated, excess individual
lifetime cancer risks were estimated using the individual 24-hour time-weighted average exposure
concentrations calculated for each sampling date, as well as the arithmetic mean of all the 24-hour TWA
exposure concentrations calculated for sampling dates between March 13, 2013 and March 8, 2016,
prior to cave closure. An average value based on the last three years of data could underestimate true
exposures and associated excess cancer risks based on fluctuations in the cave air TCE concentrations
over the last 25 years.
5.2 Parameters
The definitions, values, and references for the exposure parameters and toxicity values used in this
document are provided in Table 3. For the chronic scenarios, the EPA's standard default exposure
parameters (USEPA, 2014b) were used to best represent reasonable maximum exposure scenarios,
11 of 21
-------
which are the highest exposures reasonably expected to occur at a site (USEPA, 1989). These values
were based on the 2011 Exposure Factors Handbook (USEPA, 201 lc). Site-specific exposure times
were used for the number of hours spent per day in each section of the cave by the cave visitors, tour
guides, and gift shop/cafeteria employees; these exposure times are provided in Section 5.1.1.
Table 3. La Jolla Spring Cave Parameters.
Parameter
Definition
Units
Value
Reference
AT 24
Averaging time over a 24-hour period - developmental
hours
24
-
ATcnriccr
Averaging time over a lifetime - cancer
days
25,550
USEPA, 2014b
AT nc, chronic
Averaging time over chronic exposure - non-cancer
days
9,125
USEPA, 2014b
CA,
TCE concentration away from the cave
Hg/nv1
0
-
ca2
TCE concentration in gift shop
(ig/m3
See Table 4
Calculated
CAj
TCE concentration in ballroom
(ig/m3
See Table 4
Calculated
CA4
Average TCE concentration in Loot Rock/Lassie
Room/Jungle Room
Hg/m3
See Table 4
Calculated
CA5
TCE concentration in theater
l-ig/m3
See Table 4
Calculated
ED
Exposure duration
vears
25
USEPA, 2014b
EF
Exposure frequency
days/yr
250
USEPA, 2014b
ET
Exposure time (work-day, plus lunch)
hrs/day
8.5
-
IUR
TC'E inhalation unit risk
(Hg/m3)-'
4.IE-06
USEPA, 201 la
RfC
TCE reference concentration
Hg/m3
T
USEPA, 201 la
5.3 Estimates of Non-Cancer Hazard Quotients and Cancer Risks
Estimates of non-cancer hazard quotients and excess individual lifetime cancer risks associated with
exposures to TCE via inhalation of air in the La Jolla Spring Cave for past visitors and employees are
presented in Table 4.
To calculate these estimates, the cave air samples representative of exposures (i.e., those collected in the
gift shop and along the cave tour) were first identified. Next, the TCE concentrations for the various
microenvironments (see Section 5.1.1) were determined. Site-related TCE exposures were assumed to
be zero away from the cave (CAi). The values used to represent CA2, CA3, and CA5 are identical to the
TCE concentrations detected in the gift shop, ballroom, and theater, respectively, for those dates. Due to
the spatial variability throughout the cave, there is uncertainty in using a single sampling point to
represent these various cave areas, as discussed in Section 7, but this approach was necessitated by the
limitations in the available data. The values used to represent CA4 are the arithmetic mean values of the
TCE concentrations detected in the Loot Rock, Lassie Room, and/or Jungle Room areas at each
sampling date. Note that in some instances, data were not collected at one or more of these three
locations. Finally, to address potential uncertainties regarding temporal variability over the long-term,
the average TCE concentration for each microenvironment was estimated as the arithmetic mean of
CA2, CA3, CA4, and CA5 for samples collected between March 13, 2013 and March 8, 2016.
The 24-hour time-weighted average exposure concentrations (EC24) for the cave visitor, tour guide, and
gift shop/cafeteria employee were then calculated using Equations 3, 4, and 5. These TWA
concentrations were next used to estimate developmental hazard quotients (HQ24; Equations 7, 8, and 9),
hazard quotients for chronic exposures (HQchronic; Equations 11 and 12), and excess individual lifetime
cancer risks (CR; Equations 14 and 15). Average long-term TWA exposure concentrations were only
used to estimate chronic non-cancer hazard quotients and excess individual lifetime cancer risks
associated with long-term exposures by tour guides and gift shop/cafeteria employees.
12 of 21
-------
The results were then examined for potentially unacceptable non-cancer hazards or excess cancer risks.
For non-cancer endpoints, a hazard quotient is the ratio of the exposure to the non-cancer toxicity value.
Thus, an HQ greater than 1 means that the exposure estimate is greater than the RfC, unacceptable for
the non-cancer health effect of concern. For cancer, an excess individual lifetime cancer risk greater than
1E-04 exceeds the EPA's target cancer risk range, as established in the National Contingency Plan.
Therefore, a CR greater than 1E-04 means the exposure estimate poses unacceptable excess cancer risks.
In Table 4, HQs greater than 1 and CRs greater than 1 E-04 are identified using red font.
For many years, the HQs reflecting potential cardiac defects exceeded one. The concern for
developmental hazards was greatest for the tour guides, who had the highest exposures to TCE due to
the amount of time spent in the back of the cave, followed by gift shop/cafeteria employees, and finally
cave visitors. Note that cardiac defects are only a concern for pregnant women exposed during an
approximate three-week period of heart development that occurs in the first trimester of pregnancy, as
discussed in Section 4.1.
Non-cancer hazard quotients for clironic exposures were also unacceptable for many years. Historical
HQs, based on data collected prior to opening a natural vent in the cave, exceeded 100 for both tour
guides and gift shop/cafeteria workers. This means that exposures prior to 2003, when the natural vent
was opened, were more than 100 times the level considered protective today, based on the 2011 toxicity
values. More recently, the HQs based on individual sampling dates have ranged from less than 1 to 25
for both tour guides and gift shop/cafeteria employees, depending on the time of the year that the data
was collected. Chronic HQs based on average exposures from 2013 to 2016 also exceeded 1; the HQs
for tour guides and gift shop/cafeteria employees were 9 and 5, respectively. This suggests that the
potential for clironic non-cancer health hazards, including autoimmune disease and kidney effects, has
been unacceptable over the long term.
Cancer risks only exceeded the EPA's target cancer risk range prior to 2003. Neither the risk estimates
for the individual sampling dates nor the mean exposures from 2013 to 2016 exceeded an excess
individual lifetime cancer risk of 1 E-04.
13 of 21
-------
Table 4. Non-Cancer Hazard Quotients and Cancer Risk Estimates for TCE in La Jolla Spring Cave Air.
Date
Gift
Shop
ca2
Ball-
room
ca3
Loot
Rock
Lassie
Room
Jungle
Room
CA4
Theater
ca5
Visitor
ec24
Visitor
HQ24
Tour
Guide
EC24
Tour
Guide
HQ24
Tour
Guide
HQchronic
Tour
Guide
CR
Gift
Shop/
Cafeteria
EC„
Gift
Shop/
Cafeteria
HQ24
Gift
Shop/
Cafeteria
HQchronlc
Gift
Shop/
Cafeteria
CR
10/8/2003
840
840
1,100
1100
1,500
900
1,700
1367
340
340
76.0
38
321.4
161
110
3.2E-04
297.5
149
102
3.0E-04
2/3/2004
5.2
5.2
4.9
4.9
7.8
18
23
16
5.1
5.1
0.7
0.3
2.7
1.4
0.9
2.7E-06
1.8
0.9
0.6
1.8E-06
1/19/2005
7.8
7.8
5.2
5.2
9.8
22
24
19
7.8
7.8
0.8
0.4
3.4
1.7
1.2
3.5E-06
2.8
1.4
0.9
2.8E-06
10/2005
15
15
26
26
85
94
87
89
25
25
3.2
1.6
13.2
6.6
4.5
1.3E-05
5.3
2.7
1.8
5.3E-06
2/2006
39
39
47
47
78
120
180
126
13
13
4.7
2.3
19.6
9.8
6.7
2.0E-05
13.8
6.9
4.7
1.4E-05
3/13/2013
1.2 U
1.2
5.16
5.16
7.09
18.2
22.4
16
4.19
4.19
0.6
0.3
2.2
1.1
0.8
2.2E-06
0.4
0.2
0.1
4.3E-07
8/27/2013
97.5
97.5
138
138
164
117
177
153
0.43 U
0.43
8.1
4.0
34.4
17
12
3.5E-05
34.5
17
12
3.5E-05
1/13/2014
5.37
5.37
20
20
40.1
48.8
64.8
51
12.3
12.3
1.9
0.9
7.5
3.8
2.6
7.5E-06
1.9
1.0
0.7
1.9E-06
3/24/2014
1.13
1.13
4.51
4.51
14
21.4
17.4
18
4.35
4.35
0.6
0.3
2.3
1.2
0.8
2.3E-06
0.4
0.2
0.1
4.0E-07
6/2/2014
125
125
169
169
220
196
252
223
1.29
1.29
10.8
5.4
45.8
23
16
4.6E-05
44.3
22
15
4.4E-05
8/5/2014
143
143
133
133
224
176
240
213
0.698
0.698
10.2
5.1
43.8
22
15
4.4E-05
50.6
25
17
5.1E-05
2/24/2015
1.56
1.56
8
8
9.83
11
14.8
12
4.83
4.83
0.5
0.3
2.2
1.1
0.8
2.2E-06
0.6
0.3
0.2
5.5E-07
2/25/2015
1.67
1.67
7.68
7.68
14.1
12.4
16.5
14
5.32
5.32
0.6
0.3
2.5
1.2
0.8
2.5E-06
0.6
0.3
0.2
5.9E-07
6/22/2015
100
100
127
127
114
120
147
127
0.43 U
0.43
7.4
3.7
31.6
16
11
3.2E-05
35.4
18
12
3.6E-05
7/28/2015
110
110
170
170
230
220
250
233
3.6
3.6
10.8
5.4
45.4
23
16
4.6E-05
39.0
19
13
3.9E-05
8/25/2015
14
14
43
43
300
260
300
287
13
13
7.4
3.7
30.0
15
10
3.0E-05
5.0
2.5
1.7
5.0E-06
9/17/2015
19.4
19.4
56.2
56.2
158
n/a
207
183
4.46
4.46
5.5
2.7
22.3
11
7.6
2.2E-05
6.9
3.4
2.4
6.9E-06
9/22/2015
15.3
15.3
55
55
181
n/a
241
211
17.9
17.9
6.2
3.1
25.3
13
8.6
2.5E-05
5.4
2.7
1.9
5.4E-06
10/21/2015
51
51
150
150
590
n/a
660
625
45
45
18.1
9.1
73.6
37
25
7.4E-05
18.1
9.0
6.2
1.8E-05
11/30/2015
6.5
6.5
43.3
43.3
36.7
n/a
51.8
44
33.4
33.4
2.7
1.3
10.8
5.4
3.7
1.1E-05
2.3
1.2
0.8
2.3E-06
2/17/2016
3.38
3.38
20
20
n/a
n/a
19.6
20
14.6
14.6
1.2
0.6
4.9
2.4
1.7
4.9E-06
1.2
0.6
0.4
1.2E-06
2/22/2016
5.69
5.69
36.3
36.3
22.8
n/a
29.2
26
28.4
28.4
2.0
1.0
8.2
4.1
2.8
8.2E-06
2.0
1.0
0.7
2.0E-06
3/8/2016
64
64
210
210
420
n/a
490
455
59
59
16.4
8.2
67.0
34
23
6.7E-05
22.7
11
7.8
2.3E-05
Average
3/13/2013
- 3/8/2016
43
78
162
14
25.5
8.7
2.6E-05
15.1
5.2
1.5E-05
Red font indicates the risk estimates are unacceptable. For estimates of developmental and chronic hazard quotients (HQ24 and HQchromc, respectively), the hazard quotient > 1.
For estimates of excess cancer risk, CR > 1E-04.
Page 14 of 21
-------
6.0 LEVELS OF HEALTH CONCERN
The second objective of this document is to derive levels of health concern for TCE in air that are
protective of all potential future human receptors at the cave, including workers and visitors. As
potential human health risks based on exposures to TCE throughout the La Jolla Spring Cave were
evaluated, the high degree of spatial variability became apparent. Levels of health concern could be
derived for each area of the cave, using a similar spatial and temporal averaging approach that was
necessary in order to evaluate the existing historical data collected using stationary Summa canisters.
Alternatively, a less uncertain approach is to derive levels of concern strictly based on the number of
hours a person is exposed to the cave air each day. Here, the levels of concern could be compared with
personal monitoring data that more accurately reflects the true average exposure concentrations over the
periods of concern. Thus, levels of concern are needed for employees exposed to TCE in cave air over
an 8.5-hr workday, including lunch, and they are needed for visitors exposed to TCE during a single
tour. The equations used to derive these levels of concern are presented in Sections 6.1 and 6.2, and the
levels are provided in Table 5.
6.1 Levels of Concern for Cave Employees
6.1.1 Developmental LOC
Equations 2 and 6 can be manipulated to solve for the level of concern for developmental health effects,
using a target non-cancer hazard quotient of 1 and the exposure parameters presented in Table 3, as
follows. For an 8.5-hour workday, which includes a 30-minute lunch break, this LOC equals 6 (Jg/m3
(rounded to one significant figure).
6.1.2 Chronic Non-Cancer LOC
Equation 10 can be manipulated to solve for the level of concern for chronic, non-cancer health effects,
using a target non-cancer hazard quotient of 1 and the exposure parameters presented in Table 3, as
follows. For an 8.5-hour workday, which includes a 30-minute lunch break, this LOC equals 8 ng/m3
(rounded to one significant figure).
6.1.3 Cancer LOC
Equation 13 can be manipulated to solve for the level of concern for cancer risks, using a target excess
cancer risk (TR) of 1E-04, which is the upper bound of the EPA's target cancer risk range, and the
exposure parameters presented in Table 3, as follows. For an 8.5-hour workday, which includes a 30-
minute lunch break, this LOC equal 280 |ug/m3 (rounded to two significant figures).
(16)
(17)
TCE LOC
TR'ATcancer(days)
(18)
Page 15 of 21
-------
6.2 Level of Concern for Cave Visitors
Equation 16 can be used to derive a level of concern for developmental health effects posed to cave
visitors. This level of concern is dependent upon the length of time a visitor spends in the cave,
including time in the gift shop, cafeteria, and on tour. For a 2-hour exposure scenario, the LOC equals
24 pg/m3. However, the exact length of future cave tours may differ from what occurred prior to the
cave closure in March 2016. Note that it is unlikely that future tour guides could be exposed to 24 pg/m3
TCE on each of three or four tours per day without exceeding the 8.5-hour level of concern of 6 pg/m3.
Table 5. Levels of Health Concern for Trichloroethylene in La Jolla Spring Cave Air (ng/nv').
Cave Employees (8.5-hr Exposure Scenario)
Developmental Non-Cancer LOC:
6
Chronic Non-Cancer LOC:
8
Cancer LOC:
280
TCE Level of Concern for 8.5-hr Exposures:
6
Cave Visitors (2-hr Exposure Scenario)
Developmental Non-Cancer LOC:
24
TCE Level of Concern for 2-hr Exposures:
24
7.0 UNCERTAINTIES
7.1 Uncertainties in Data Representativeness
There are uncertainties in the temporal and spatial representativeness of the data used to estimate
potential risks to human health associated with exposures to TCE in the La Jolla Spring Cave air.
As discussed in Section 2, a natural vent in the cave was opened in response to the high TCE
concentrations (100 to 1,700 pg/m3) that were detected in air samples collected on three dates in 2002
and 2003. Of the sampling events that provided sufficient data to estimate potential health risks
associated with the cave in this document (see Table 4), only one occurred when the vent was still
closed. The remainder occurred after the vent was opened and the TCE levels dropped, predominantly
between 2013 and 2016, with a few samples collected between 2003 and 2007. It is possible that the
TCE concentrations in the cave had been elevated at the 2002 - 2003 levels for many years, when the
natural vent was still closed. However, since no air samples were collected prior to 2002, long-term
exposures and estimates of chronic non-cancer hazard quotients and cancer risks to tour guides and gift
shop/cafeteria employees may have been underestimated. For example, the non-cancer hazard quotients
for chronic exposures to a tour guide, based on samples collected after 2003, ranged from 1 to 25, with
an average HQ of 9 based on the 2013 - 2016 data. In contrast, the HQchronic for a tour guide based on the
2003 data was much higher, at 110. Similarly, the non-cancer HQs for chronic exposures to gift
shop/cafeteria employees ranged from less than 1 to 17, with an average HQchronic of 5 based on the
2013-2016 data; in contrast, the 2003 HQchronic was 102. The excess individual lifetime cancer risks
posed to both tour guides and gift shop/cafeteria employees, based on data collected after 2003, were all
within the EPA's target cancer range of 1E-06 to 1E-04, but the excess cancer risks based on the 2003
data were greater than 1E-04 for both types of employees. Importantly, while there are uncertainties
regarding the true magnitudes of the non-cancer hazard quotient estimates for tour guides and gift
shop/cafeteria employees, the HQs are well above 1 and clearly indicate unacceptable health risks even
without data on the potentially higher TCE concentrations expected prior to 2002.
Another form of temporal variability noted in the La Jolla Spring Cave system is associated with the air
flow within the cave, which is influenced by outside air temperature and humidity levels. Seasonally,
16 of 21
-------
TCE concentrations inside the cave have been lowest during the winter, when outside air temperatures
are low, and higher during the summer, when outside temperatures are high. However, some of the
highest TCE levels have been detected in the spring and fall, when the air temperatures inside and
outside the cave are similar. The earliest cave air samples were collected yearly, but beginning in 2013,
the samples were collected quarterly, then monthly, then weekly, in an attempt to address uncertainty
associated with seasonal variability. If the available data do not capture the true mean TCE
concentrations, accurately accounting for seasonal fluctuations, long-term health risks may have been
under- or overestimated.
In addition to seasonal fluctuations, the influence of outside air temperatures and humidity levels on air
flow within the cave may result in diurnal variability, with higher TCE levels present during the warmer
portion of the day and lower concentrations at night. All of the data used to estimate potential health
risks were collected using 24-hr Summa canisters. However, if the TCE concentrations are higher during
the day and lower at night, the 24-hour samples may underestimate the true mean exposure
concentrations and associated health risks during the 8.5-hr workday.
Developmental hazard quotients are subject to all of the above forms of variability. Because the risks are
calculated for a single day of exposure, they can only be calculated for those days when samples were
collected. Further, diurnal variation may have caused the developmental hazard quotients derived using
24-hrs samples to underestimate true health hazards, compared with samples collected over the actual
exposure time. Nonetheless, the developmental hazard quotients derived using the available data were
greater than 1 for all receptors at the cave, indicating unacceptable health risks.
Spatially, the highest TCE levels have been detected at the sampling locations deepest within the cave
(e.g., Loot Rock, Lassie Room, Jungle Room), and the lowest TCE levels have been detected closest to
the cave entrance (e.g., gift shop). Based on the available sampling locations, five microenvironments
were defined, including away from the cave, the gift shop/cafeteria, the ballroom, the Loot Rock/Lassie
Room/Jungle Room area, and the theater area. Additional microenvironments could have been defined,
but 1) insufficient data was collected in other areas to define additional areas, and 2) uncertainty exists
regarding how much time cave visitors and workers spend in each portion of the cave. Even within the
defined microenvironments, spatial variability likely exists. However, only one sampling location was
used to represent TCE concentrations in the gift shop/cafeteria, ballroom area, and theater area, and
between one and three samples (depending on the sampling date) were used to represent the back
portion of the cave. Despite the uncertainty due to spatial variability, the resulting non-cancer hazard
quotients for each receptor type consistently exceeded 1 (i.e., HQ >1) by such a large margin, that there
is high confidence that the TCE levels in the cave posed unacceptable health hazards. To address this
uncertainty moving forward, personal monitoring data more accurately reflects true exposure
concentrations over the period of concern, for use in comparison with the levels of health concern
derived in Section 6.
7.2 Uncertainties in Risk Characterization
As discussed in Section 4, representatives of TCE manufacturers and distributors have continued to
petition the EPA to change the IRIS assessment so that the non-cancer toxicity values are not based on
the cardiac defects observed in the Johnson et at. (2003) study, in direct contrast to the recommendations
of the EPA's Science Advisory Board and despite more than a decade of external and internal public
peer review. The EPA denied both the Request for Change and the Request for Reconsideration,
regarding the IRIS TCE assessment. Critically, the TCE IRIS RfC is based upon three non-cancer
endpoints. The level of health concern derived for the La Jolla Spring Cave air based on cardiac defects
17 of 21
-------
is 6 |ug/m3 for an 8.5-hr exposure, while the level of concern based on autoimmune disease and kidney
effects is 8 |_ig/m3. Estimates of hazard quotients for both developmental and chronic non-cancer
endpoints were unacceptable for all receptors at the cave.
Developmental hazard quotients were derived assuming a 24-hour critical exposure period, consistent
with the EPA's Developmental Toxicity Guidelines (USEPA, 1991) and with the approach used by the
EPA's Office of Chemical Safety and Pollution Prevention (USEPA, 2014c). The IRIS RfC for cardiac
defects was derived using benchmark dose modeling based on the combined observances of all of the
various types of heart defects in the critical study (Johnson et al., 2003) to determine what level was
associated with a 1% excess risk of any type of heart malformation. Because the animals were dosed
throughout gestation, with no interim sacrifices, the number and type of malformations that occurred on
any single day are unknown. That is, all of the observed heart defects could have occurred on one day
during valvuloseptal morphogenesis, or they could have occurred throughout the entire period of heart
development, which lasts approximately three weeks in humans. The Agency has made the health-
protective assumption that all of the heart defects can occur on one day, so a 24-hour critical exposure
period was used. Had a three-week critical exposure period been used instead, the resulting level of
concern for developmental endpoints associated with 8.5-hr daily exposures, five days per week, would
have been 8 |ig/m3, which is identical to the level of concern derived for the chronic non-cancer effects.
Exposure parameters used in this document were selected to represent reasonable maximum exposure
scenarios, to the extent possible, within the limitations of the available data. Some of the assumptions
outlined in Section 5 may have resulted in under- or over-estimates of the true health risks, but because
the calculated hazard quotients were well above 1, there is little uncertainty that the risks from exposures
to TCE in cave air were unacceptable.
8.0 CONCLUSIONS
o The concentrations of TCE detected in the La Jolla Spring Cave air between 2002 and 2016
posed significant risks to human health for both short- and long-term exposures,
o Individuals who worked in the cave were exposed to concentrations of TCE in cave air
significantly above levels of health concern for developmental effects (6 fig/m3) and chronic
non-cancer health effects (8 |jg/m3) for many years,
o Potential developmental and chronic non-cancer health hazards to tour guides were greatest, due
to the length of the time they spent in the cave leading several tours each day and their repeated
exposures over many weeks or years as part of their job.
o Potential developmental and chronic non-cancer health hazards to employees in the gift shop and
restaurant were less than those to tour guides because TCE concentrations were lower in those
areas, compared to the back portions of the cave,
o Some cave visitors may have had exposures to TCE above a level of health concern for
developmental endpoints, based on a 2-hour exposure; however, the likelihood that cardiac
defects could have resulted from these exposures are low because a woman would have to be at a
specific, approximate three week window in the first trimester of her pregnancy to be at risk,
o The short, one-time exposure to TCE in the cave by a visitor does not pose a significant risk for
chronic, non-cancer health effects such as effects on the immune system or kidneys, or for
cancer.
18 of 21
-------
RKKERKNCES
Agenc> tor low Substances and Disease Registt\ 2009. Solicited comments on IX T Draft Document.
OITice of Science, National Center lor i-n\ trotiiiicnial Health, <\ug»st 24, 2000,
http; ofmpnkcp;t.t!o\ >'tins eunseomni.tiet lile 'p dou nload id 4lf2n»t
Agency lor foxie Substances and Disease Registiy 201 I. t'otnmenh on P PA A fovieological Review
of tnchloroethane |sic] ( 1 CP'). Jul\ 13, 201 I,
http: (diii|>tib,epa,i!o\.eun> tiaisv'oiiiiH.uerfik id 50*44?*
Dhanantuari. P: 1 ee. 1:': Krishnan. A; Samtani, R: Yaniada, S; Anderson. S; I oekett, P: Doiiolt'io, M;
Shiota, K: I catherburs, I.: Lo. (A 20(10, Human cardiac dc\elopmeiu in the first trimester: a
high-resolution magnetic resonance imaging an4 > 51.
Halogenated Solvents Induslry Alliance. Inc. 2013. Request lor t 'onection IRIS Wessment for
Triehloroethylene. RFC 14(H) 1. Mo\emher 5. 20] V
Halogenated Solvents Induslry Alliance. Inc. 20! >. Request foi Reconsideration a 1400I. RPR 14001 A.
June P, 2015.
Johnson. P.; Goldberg, S.: Ma\s. M.; Dawson, B. 2003. I hreshoKI oftrichloroetluienc contamination in
matenial drinking waters affecting fetal heart de\elopmenl in the rat. /.iiviroit //<-tilth Pciwpcii
111: 280-292. hup: u wu .u JM.nSot ml; eo\ nnbmed i '"O b-so
Keil, D: Peden-Adams. M; Wallace. S: Rui/. P. (iilkeson. it. 200l>. Assessment of tnchloroetln letie
1:.) exposure in murine strains getiettcalh-prone and non-prone to develop autoimmune
disease../ l-'.nxiron Sci Health .1 '/m Ihiziirtl Stthst l:.r,yivoii l.'tig 44: 443-453
Imp; d\,dot.ore 10 luNO IiM > f520 Comments on Draft I PA 1 CP
IRIS Risk Assessment, August 2o. 2000
hup. oftnjnib.opa.yos eimsvmi\co!nm.,gc!lile.'p dow nioad id 493ttOO
National Aeronautics and Space Administration, 201 1. NASA ("omments on the Final Draft TCP IRIS
Risk Assessment. Juh 20] 1,
hup: ofmpub.epa.gm cm is cu nsconnn,get iile:p duw nioa03452
National Institutes of Health. 20! 1, ('omments otr 'the Draft IRIS I oxieological Re\ lew for
Triehloroctln lene (TCP) for Pinal Agency Interagency Science Discussion. National Toxicology
Program. National Institute of Pnvironmcntai Health Sciences.
b0j">:_..ol'n|pt!!>.cj\i,go\ etms cinwcmnm,£eUlle?j>, download id *>«**-!53
National Research Council 1 lH> I. Human PAposute Assessment for Airborne Pollutants: Ad sauces and
()pportunities. Washington. DC. "1 he National '\eadetm Prevs. 10')] _
National Research Council. 200{>, Assessing the Human Health Risks of Triehloroethylene: Rev
Scientific Issues. Washington. DC: The National Academies Press, 2006,
10 of 21
-------
Office ot" Management ami Budget. 2000, OM11 Staff Comments oil HP A IRIS draft TCP. Toxaeological
Rev iew and Draft Charge (drafts dated June 2000), September, 1. 2000,
hup. ofmpybxpa.gfn efm>comin getfdc?p do unload. id.
Office of Management and Budget. 201 I. OMR Stall* Working Comments on FlkVs Final
Agency interagency Science Discussion draft Toxieologieal Review of Triehloroeth) lene (TCH)
and draft IRIS Summary (dated June 2011), July 21. 201 I.
http: otmpub.epu go\ Vims eimscoinm.uetlile'.'p d>>u nktad id St 1345 I
U.S. Department of Defense. 200%. Department of Defense Comments on the Draft 1 CF Toxieologiea!
Review; Peer Review Charge, June 2000. The Chemical Material Risk Management Directorate.
August 27. 2000, hup:1 ofmpub.epa.em cmi* cmiscoitim.jjotft|cj> download id -102092
I LS. Department of Defense. 2000b, Department of Defense Comments on the 1 \R Draft I IS EPA TCH
Risk Assessment. June 2000. The Chemical Material Risk Management Directorate. September
1. 2000, hup/ ofmpub.epa,go\ cans eitnscomiii.geUilc'.'p ,dow nload id 402n0|
U.S. Department of Defense. 2011, Department of Defense Comments on TXT: IASD final draft "1 ox
Review redline.pdf. The Chemical Material Risk Management Directorate, Jul} 15. 201 1.
htm: ofmpuh.eoa,go\,eints etinNCoin.ni.uctlilc"p._dow nload _ id_ 5u345 1
U.S. Environmental Protection Agency. |0X0. Risk Assessment Guidance for Superfund Volume 1:
Human Health Evaluation Manna! - Part *\. I'P A 540 1-80 002, Office of Fmergencx and
Remedial Response, Washington, D.C
U.S. Fnvironmental Protection Agency. 1001. Guidelines toi De\ elopnseiUal "I oxicitv Risk Assessment.
F PA 000 FR-01 ()01. Risk Assessment Forum, Washington. D.C,
I LS. Fnvironmental Protection Agencx, 1002, (iuidelines for iwposure Assessment. HP A 000 Z-l)2;00L
Risk Assessment Forum, Washington, l),C.
U.S. Fnvironmental Protection Agency. 2002. Rex iew of Draft Tnchloroethyleue I (ealth Risk
Assessment: Synthesis and Characterization: An HPA Science Advisor) Board Report. P.PA-
SAB-FHC-03-002. TCP Review Panel. Fnvironmental Health Committee, US FPA Science
Advisory Board, Washington, D.C. December 2002.
htlp: joseisifte..q\t.go\ >ab0f>5( ^thproduct.nsf D14( 30Hi 54N2I,41N525"! CI 0tioo"yv,n Male
ehen,Nf>(L\pdt
U.S. Fnvironmental Protection Agency. 2003. Human Health 'toxicity Values in Superfund Risk
Assessments. OSW'F.R Directiv e 0285 7-53 Office of Superfund Remediation and Technology
Innovation. Washington. D.C.
U.S. Fan ironmental Protection Agene\. 2005. Guidelines for Carcinogen Risk Assessment. FPA030T'-
03-001 F. Risk Assessment Forum. Washington. D.C.
U.S. Fnvironmental Protection Agency, 2000. Risk Assessment Guidance for Superfund: Volume I
Human Health 1:valuation Manual (Pail h Supplemental Guidance for Inhalation Risk
Assessment). OSWFR Publication 0285.7-82. Office of Superfund Remediation and Technology
Innovation, Washington, D.C.
20 of 21
-------
U.S. F»\ironmental Protection Agency, .201 la. I oxieologiea! Review of PnehIorocth\lone (("AS No,
7lM) I -(>), I "PA'035 R-00,01 I ]:. Washington, l) A September 2*0 i.
U.S. Em ironmental Protection Agene\, 201 Ih Re\ lew of PiPAN Orati Assessment entitled
* 1 oncological Review of Triehloroetln lone" ((ictobci 2000 j. I-'PA-SAB- i 1-002. TUP Review
Panel, US PPA Science Advisory Board, Washington. I).(' January 1 1, 201 I,
hftjK \OM.'_mi;e.cpu.yo\ v;ih"„5(\abproiUict ikI'V-;\s| | x.}(>l )S>.'K7sj Ton.4 \I»)XS si sic
I P VSAIUl 1 -(Hi.1 /anMgncd.pdf
I:.S. Hnsironmental Protection Agency, 20! 1 e. Pxposurc 1^aclnrs I landbook: 201 ) edition. National
Center for P"n\ironmental Assessment, Washington, I'M PPA (>(H) R-fiu 05212
U.S. Fnvironmental Proteetion Agency. 2014a. '1(21' L)e\elopmental (, ardiac loxicity Assessment
Update. rj)A"lIt,U)Pl»r^2i»12"tr2-(i045. Juh 2«>N.
iulpi www reguiaiioHN.ui'\ • [document! Mai!.!) PP \ I iiUOPP !'-20 i .Mi""11 VpiM.S
U.S. Frn ironmental Protection Agency. 2014b. Human I Icalih (valuation Manual. Supplemental
Guidance; I -pdate of Standard Default PAposure Factors, ()SWPR Directive 0200.1 -120, Office
of Solid Waste and Fmergcne\ Response, Washington, I).(2
U.S. Fm ironmental Proteetion Agenc\. 2014c. f'St 'A Work Plan C hemical Risk Assessment.
I richloroethvlene: Decreasing, Spot Uleaning and Arh & (Yaft> I !ses. C ' ASRN: I2PA
Document 740-R1 -4002. (Mike of ( henuca! Safetx and Pollution Pre\ ention, Washington, D.U,
June 2014
U.S. bin ironmental Protection Agency 201 \ P P A Respoi^e to RP(' I40ul, Match I1'. 2.
21 of 21
------- |