Available via license: CC BY 4.0
Content may be subject to copyright.
Application of a weight of evidence approach to evaluating risks
associated with subsistence caribou consumption near a lead/zinc mine
Michael R. Garry
a,
⁎, Scott S. Shock
b
, Johanna Salatas
c
, Jim Dau
d
a
Exponent, Center for Health Sciences, 15375 SE 30th Place, Suite 250, Bellevue, WA, USA
b
Exponent, Environmental Group, 15375 SE 30th Place, Suite 250, Bellevue, WA, USA
c
Teck Alaska Incorporated, Anchorage, AK, USA
d
Alaska Department of Fish and Game (retired), Box 689, Kotzebue, AK 99752, USA
HIGHLIGHTS
•Multiple lines of evidenceused to evalu-
ate risks associated with subsistence
caribou consumption near a mine in
NW Alaska.
•The study supplements a multi-pathway
humanhealthriskassessmentoffugitive
dust emissions from mine operations.
•Data from satellite-collared caribou show
that caribou utilize the area near the
mine at a lower rate than previously as-
sumed.
•Ongoing monitoring indicates a lack of
consistent trends for either increasing or
decreasing metals concentrations in cari-
bou.
•Evidence indicates fugitive dust emissions
from operations at this mine are not a
significant source of metals in caribou.
GRAPHICAL ABSTRACT
abstractarticle info
Article history:
Received 28 August 2017
Received in revised form 7 November 2017
Accepted 14 November 2017
Available online 29 November 2017
Editor: E. Capri
Overlandtransport of ore concentrate from the RedDog lead/zinc mine in northwest Alaskato its seaport has his-
torically raised concerns amonglocal subsistence users regarding the potential impacts of fugitive dust fromthe
operation,including the potential uptake of metalsinto caribou meat.Caribou are an integral part of lifefor north-
ern Alaska Natives for both subsistence and cultural reasons. The Western Arctic caribou herd, whose range in-
cludes the Red Dog mine, transportation corridor, and port site, sometimes overwinter in the vicinity of mine
operations. A weight of evidence approach using multiple lines of evidence was used to evaluate potential
risks associated with subsistence consumption of caribou harvested near the road and mine. Data from a long-
term caribou monitoring program indicate a lack of consistent trends for either increasing or decreasing metals
concentrations in caribou muscle, liver, and kidney tissue. Lead, cadmium, and zinc from all tissues were within
the range of reference concentrations reported for caribou elsewhere in Northern Alaska. In addition, a site use
study based on data from satellite-collared caribou from the Western Arctic Herd showed that caribou utilize
the area near the road, port, and mine approximately 1/20
th
to 1/90
th
of the time assumed in a human health
risk assessment conducted for the site, implying that risks were significantly overestimated in the risk
Keywords:
Fugitive dust
Western Arctic herd
Native northern Alaskans
Metals
Science of the Total Environment 619–620 (2018) 1340–1348
Abbreviations: ADPH, Alaska Department of Public He alth; ALM, adult lead model; CAH, central Arctic herd; DMTS, DeLo ng Mountain Regional Transportation Syst em; EPA,
Environmental Protection Agency; FI, fractional intake; TCH, Teshekpuk herd; WAH, Western Arctic herd.
⁎Corresponding author.
E-mail addresses: mgarry@exponent.com (M.R. Garry), ssshock@exponent.com (S.S. Shock), johanna.salatas@teck.com (J. Salatas), jimdau@otz.net (J. Dau).
https://doi.org/10.1016/j.scitotenv.2017.11.149
0048-9697/© 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Contents lists available at ScienceDirect
Science of the Total Environment
journal homepage: www.elsevier.com/locate/scitotenv
assessment. The results from multiple lines of evidence consistently indicate that fugitive dust emissions from
Red Dog Operations are not a significant source of metals in caribou, and that caribou remain safe for human
consumption.
© 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
1. Introduction
The Red Dog mine area in northwest Alaska is highly mineralized
and overland transport of lead and zinc ore concentrate from the mine
to the seaport occurs throughout the year along the DeLong Mountain
Regional Transportation System (DMTS). Sampling results from 1999
indicated the presence of lead in road soil and on moss growing near
the road (NPS, 2001). Caribou (Rangifer tarandus) are an integral part
of life for northern Alaska Natives for both subsistence and cultural rea-
sons. Multiple studies of caribou were conducted to evaluate potential
risks associated with subsistence consumption of caribou harvested
near the RedDog mine and DMTS road. These studies, along withan ad-
ditionalsite use study based on data from satellite-collared caribou from
the Western Arctic Herd, are described in this paper. We evaluated mul-
tiple lines of evidence from these studies to determine whether fugitive
dust emissions from Red Dog Operations are a significant source of
metals in caribou, and to assess whether caribou remain safe for
human consumption.
1.1. Caribou herds in Northern Alaska
Four distinct caribou herds are found in northern Alaska. The area
near the Red Dog mine and the DMTS is included in the normal annual
range of two of those herds, the Western Arctic (WAH) and the
Teshekpuk (TCH) herds (Fig. 1). The Western Arctic herd (WAH) ranges
over approximately 140,000 mile
2
(Dau, 2015). This herd numbered ap-
proximately 242,000 in 1970, declined to approximately75,000 in 1976,
and peaked at approximately 490,000 in 2003. In 2013, the herd num-
bered approximately 235,000 (Dau, 2015). The smaller TCH numbered
approximately 39,000 in 2013, the last year for which photocensus
data were available (Parrett, 2015). The TCH range also extends to the
area around the DMTS, particularly in the winter (Person et al., 2007).
In addition, caribou from these two herds regularly mix (Dau, 2015;
Parrett, 2015). Person et al. (2007) estimated an apparent emigration
rate from the Teshekpuk herd of approximately 7% in 1990–2005, even-
ly split between the WAH and the Central Arctic herd (CAH). The WAH
ranges widely throughout theyear, and in some years caribou from this
and the TCH winter near the road and mine thus raising the possibility
these animals might be exposed to mine-generated metals in dust,
soil, water, or plants.
1.2. Previous caribou studies
In 2001, the Alaska Department of Public Health (ADPH) conducted
a health study evaluating exposure of residents from Kivalina and
Noatak, the two villages nearest Red Dog mine, to metals present in sub-
sistence foods (ADPH, 2001). The ADPH study included a review of prior
studies conducted by ADFG and others, as well as an analysis of metals
levels in subsistence foods, including in caribou collected from near
the Red Dog mine as part of a study conducted by ADFG in 1996.
ADPH concluded that “…average concentrations of metals found in car-
ibou harvested near the Red Dog Mine and DMTS road were low. Eating
caribou from the Western Arctic caribou herd does not pose a public
health threat.”
An additional site-specific, multi-pathway human health risk assess-
ment (HHRA) was conducted using data collected from 2000 to 2004
that evaluated exposure to DMTS-related metals through incidental
soil ingestion, water ingestion, and subsistence food consumption
under three scenarios: 1) child subsistence use, 2) adult subsistence
use, and 3) combined worker/subsistence use (Exponent, 2007). Metals
exposure associated with caribou consumption was estimated in the
HHRA based on measured metals concentrations in muscle, liver, and
kidney from caribou harvested in 2002, along with consumption data
from surveys administered by the Alaska Department of Fish and
Game, Division of Subsistence in the two villages in closest proximity
to Red Dog, Kivalina and Noatak. Estimated risks from each of the sce-
narios were well within acceptable public health limits. Because of
this finding, along with additional supporting data, the HHRA concluded
that caribou remain safe for subsistence consumption. However, expo-
sure to metals through caribou consumption accounted for approxi-
mately 70% of the total estimated risk in all three scenarios,
highlighting the importance of additional understanding of this poten-
tial exposure pathway.
In risk assessment practice, the term “subsistence level consump-
tion”describes consumption of most or all of a particular food type or
group of foods from a locally harvested source and/or consumption of
those foods at significantly higher rates than the general public. Two
key assumptions were included in the HHRA to address uncertainty in
the contribution of subsistence foods to the overall diet and the contri-
bution of site-related metals to human health risk. First, although data
from the region indicate that subsistence foods compose approximately
20% of the total diet (Johnson et al., 2009), the HHRA assumed that sub-
sistence contribute 100% of the diet of local subsistence users. Second,
fractional intake (FI, defined as the fraction of metals in a food type de-
rived from the site relative to total metals present in the food source)
was estimated to be 0.09 (i.e., 9%) based on the ratio of the total land
area covered by the study area relative to the total harvest area used
by neighboring communities. The latter assumption overestimates the
contribution of metals from the site both because the caribou home
range is far greater than the harvest area for local communities and be-
cause individual caribou spend a relatively small amount of time in the
area near the mine and DMTS road.
Although the ADPH analysis and the HHRA indicated that potential
impacts to caribou and subsistence users were minimal to negligible,
additional monitoring and study of caribou and other subsistence
foods are included as part of ongoing fugitive dust risk management ac-
tivities for the site. Results from that monitoring are included in the
analyses described herein.
2. Methods
One aspect of ongoing risk management for fugitive dust at Red Dog
Operations includes periodic collection and necropsy of caribou to as-
sess their general health and determine tissue metal concentrations.
Additionally, a caribou site use study was conducted to reduce uncer-
tainty related to risk assessment assumptions. In this paper, the results
of these studies are evaluated in the context of potential risks associated
with subsistence consumption of caribou harvested near the road and
mine. The methods used in these studies are described below.
2.1. Caribou monitoring program
In spring 2002 and again in spring 2009, 10 caribou were harvested
from near the Red Dog mine and the DMTS road by hunters from
Kivalinaor Noatak who had been contracted by Teck Alaska for this pur-
pose. All caribou collected had overwintered near the Red Dog Mine and
1341M.R. Garry et al. / Science of the Total Environment 619–620 (2018) 1340–1348
northeastern section of the DMTS. The 2002 caribou ("Red Dog 2002")
were from the WAH and the 2009 caribou ("Red Dog 2009") were
from TCH. Caribou were killed by a gunshot to the head or cervical ver-
tebrae whenever possible, and intact carcasses were transported to a
heated building immediately after being taken. During 2002 necropsies
were conducted by Dr. P. Meyer and J. Dau following standard laborato-
ry procedures. Necropsies conducted during 2009 were performed by
Dr. K. Beckmen, J. Dau and L. Parrett following the CARMA protocol
(Gunn and Nixon, 2008). All necropsies were conducted on the same
day that the animals were harvested, and usually within 1–4hoftime
of death. Muscle, liver, and kidney tissues were dissected at the time
of harvest, frozen immediately, and shipped for metals analysis. No
samples were taken from tissue areas that might have been affected
by bullet fragments. Incisor teeth were collected to estimate animal
age. All edible meat fromcollected caribou was salvaged and distributed
to residents of Noatak and Kivalina.
Metals data from the 2009 collection were compared to metals con-
centrations in the “reference”caribou tissue harvested in 1996 from
other areas in northern Alaska without known anthropogenic point
sources of metals (O'Hara et al., 2003). Four of the reference locations
included WAH caribou (Anaktuvuk, Barrow, Cape Thompson, Point
Hope) and one included TCH caribou (Teshekpuk). Because raw data
from the 1996 data set were not available, statistical differences be-
tween Red Dog and reference caribou were assessed by treating each
Fig. 1. Site Location –Location of Red Dog Mine and DMTS in northwestern Alaska with home ranges of Western Arctic and Teshekpuk Caribou Herds.
1342 M.R. Garry et al. / Science of the Total Environment 619–620 (2018) 1340–1348
location mean in the reference dataset as an independent measurement
and calculating a one-sided 95% upper prediction limit (95%UPL) for the
range of expected measurements. Red Dog caribou tissue concentra-
tions were considered statistically different if the mean exceeded the
95%UPL.
Data were also compared to metals concentrations in Canadian car-
ibou and Scandinavian reindeer from the literature (Borch-Johnsen et
al., 1996; Chan et al., 2001; Crete et al., 1989; Elkin and Bethke, 1995;
Kuhnlein and Soueida, 1992; Larter and Nagy, 2000; Rintala et al.,
1995). None of the populations included in the literature comparisons
was known to have any specific metal exposure other than to levels
that occur naturally in the environment. Raw data were not reported in
these studies so statistical analysis of the comparison was not conducted.
To assess temporal trends, Red Dog caribou metals concentrations
from 2009 were compared to previous years. Although Red Dog caribou
included animals from different herds (WAH and TCH) in different
years, the temporal comparison is meant to characterize trends in
metals concentrations in caribou harvested and eaten by locals in the vi-
cinity of Red Dog, regardless of herd. If metals in fugitive dust from Red
Dog are influencing tissue metals concentrations in caribou during their
time spent in the vicinity of the site, that influence would theoretically
be measurable in caribou harvested near the site irrespective of herd.
Concentrations in 2009 were compared to 2002 using the two-sided
Student's t-test. To assess the effect of animal age on organ metals con-
centrations, data were also evaluated followingage adjustment, where-
by metals concentrations were divided by the age of the animal, with
results expressed in mg/kg/year. Differences were considered statisti-
cally significant at Pb0.05. Metals concentrations from WAH caribou
harvested near the site in 1996 ("Red Dog 1996"; O'Hara et al., 2003)
were also included in the comparison, but because raw data were not
available for individual animals from 1996, they were not included in
the statistical analysis.
2.2. Caribou site use study
The caribou site use study was conducted using satellite collar
location data from WAH caribou in collaboration with the Alaska
Department of Fish and Game (ADF&G). The purpose was to
determine if the assumed fractional intake used in the HHRA was
realistic. The ADFG satellite collar data needed for the analysis are
protected information, in order to protect the herds from over-
hunting. Analysis parameters and GIS shape files were developed
and transmitted to ADF&G, where site use related statistical analyses
were conducted.
The analyses included data collected from satellite-collared caribou
each year from 2000 to 2012, excluding data from the date of initial
collaring (September of each year) until May 6th of the following year
(beginning of spring migration) to allow newly deployed collars to be-
come distributed throughout the WAH range. The analysis included
data from two types of satellite collars: platform terminal transmitter
(PTT) and global positioning system (GPS). Data from both collar
types were combined after statistical analysis demonstrated a lack of
significant differences in frequency of site usage between the PTT and
GPS collars (Randomized Block ANOVA. Area 1: F = 2.30, P=0.13;
Area 2: F = 2.18, P=0.14).
The analysis was conducted for two zones of distance from the road
complex as alternate definitions of the “site”area:
•Area 1 —Delineated by a 5-km zone to the N/NW and a 2-km zone to
the S/SE of the DMTS and mine air permit boundary (Fig. 2). This was
the assumed site area in the HHRA.
•Area 2 —Delineated by a 10-km zone on all sides of the DMTS and the
mine air permit boundary(Fig. 2). This area was included as an upper-
bound estimate of an area that could beaffected by dust from the site
(although most of this area is unlikely to be significantly affected).
Fig. 2. Site Area–The analysis was conducted for two zonesof distance from the roadcomplex. Area 1 is defined as thatarea within 5 mi NW and 2 mi SE of the DMTS. Area 2 is definedas
that area within 10 mi of the DMTS.
1343M.R. Garry et al. / Science of the Total Environment 619–620 (2018) 1340–1348
Site use was analyzed in three ways
1. Fractional Site Use —The fraction of time spent by WAH caribou in
the site area was estimated by calculating the ratio of total number
of satellite collar locations within the site area relative to the total
number of satellite collar data signals recorded outside of those
areas in that year. Because this metric considers the amount of time
all satellite-collared caribou spent within versus outside the study
areas, it probably better represents intensity of use at the herd level
than the two alternative approaches described below.
2. Site Entry —The fraction of animals from the WAH that entered the
site was estimated by the total number of satellite-collared caribou
that entered the site at least once each yearrelative to the total num-
ber of satellite-collared caribou in the herd. This measurement does
not consider how long the animals spent in the site area, only wheth-
er they entered the site at all. Site entry is, thus, only a crude estimate
of FI.
3. Fractional Site Use by Individual Caribou —This approach considered
only those satellite-collared caribou that entered one of the two
study areas at least once during some year. For these individuals,
the fraction of time spent in the site area was estimated as the total
number of satellite-collar locations within the site area relative to
the number of satellite-collar locations recorded outside of that
area for that year. As with the first approach described above, this
metric considers the amount of time each caribou spent near vs. far
from the development complex. However, it overestimates intensity
of use because it ignores those collared individuals that never en-
tered one of the 2 study areas during a particular year, which was
usually the majority or vast majority of the WAH.
Data distributions in the site use study were determined to be log-
normal; therefore, non-parametric methods of statistical analysis were
used for the analysis. Randomized block analysis of variance (ANOVA)
was used to evaluate differences between PTT and GPS collar data. Dif-
ferences in site use between time periods were evaluated using
Kruskall-Wallis one-way ANOVA. Separate analyses comparing 2011–
2012 with 2000–2010 were conducted because one collared animal
overwintered near the site in 2011–2012 and was observed to have en-
tered the site frequently.
3. Results
3.1. Caribou monitoring program
The 2009 Red Dog tissue sample results werecompared with metals
concentrations in caribou collected in 1996 from North Slope locations
monitored by the North Slope Bureau. There were no consistent differ-
ences between tissue metals concentrations in the 2009 Red Dog data
and the 1996 Northern Alaska data (Fig. 3). In most cases, Red Dog
metals were below the 95%UPL for the reference samples. Lead in Red
Dog caribou slightly exceeded the reference 95%UPL for liver (2.5 versus
2.2 mg/kg) and kidney (1.6 versus 1.4 mg/kg). Mean caribou muscle
Fig. 3. Site Comparison to Alaska Reference Locations –Comparison of metals concentrations in caribou harvested from the Red Dog area with caribou from other locations in northern
Alaska (from O'Hara et al., 2003). *Indicates site sample concentrations are statistically significantly different than combined reference location samples (pb0.05).
1344 M.R. Garry et al. / Science of the Total Environment 619–620 (2018) 1340–1348
lead (0.026 mg/kg) was less than the mean concentrationin each refer-
ence group and similar to the typical lead concentrations in meat, fish,
and poultry in the United States and Canada, where mean lead concen-
trations of approximately 0.02 mg/kg have been reported (ATSDR,
2007). Cadmium concentrations in caribou harvested in 2009 near
Red Dog were similar to the northern Alaska reference concentrations
(O'Hara et al., 2003) for all tissues. Mean muscle cadmium
(0.0135 mg/kg) was less than themean concentration in each reference
group (Fig. 3). Zinc concentrations were also similar to the 1996 North-
ern Alaska reference animals.
There were no consistent trends in differences between the 2009
Red Dog tissue data and metals concentrations compared to Canadian
caribou and Scandinavian reindeer reference populations reported in
the literature (Borch-Johnsen et al., 1996; Chan et al., 2001; Crete et
al., 1989; Elkin and Bethke, 1995; Kuhnlein and Soueida, 1992; Larter
and Nagy, 2000; Rintala et al., 1995;Fig. 4). In most cases, values were
reported both above and below the concentrations in the 2009 Red
Dog samples, with a large degree of intra-group variability. Statistical
analysis was not conducted for these comparisons because of the lack
of raw data and differences in reporting methods. Similar to the com-
parison to the northern Alaska reference data, mean liver and kidney
lead concentrations in the 2009 Red Dog data slightly exceeded the
mean concentrations for each of the reference samples.
Metals concentrations from cariboucollected near theRed Dog mine
and DMTS in 1996, 2002, and 2009 are presented in Fig. 5.Metals
concentrations were not consistently higher or lower in 2009 compared
to 2002. Muscle lead (P= 0.02) and cadmium (P=0.006)concentra-
tions from 2009 caribou were statistically significantly lower than con-
centrations from 2002. Muscle (Pb0.0001) and kidney zinc (P=0.003)
were statistically significantly higher in 2009 relative to 2002. All other
tissue metals concentrations were not significantly different.
3.2. Caribou site use study
3.2.1. Fractional site use
The percentage of time WAH caribou spent in the site area was con-
sistently low throughout thestudy period (Table 1). The median site use
was 0.1% for Area 1 and 0.4% for Area 2 for the entire study period,
2000–2012. Site use differed significantly between the 2000–2010 peri-
od (0.1%) and the 2011–2012 period (0.6%) for Area 1 (p=0.03),and
the difference between those time periods approached significance for
Area 2 (0.3% vs. 1.2%; p= 0.08), although it was still low in the highest
use year, 2012 (0.8% in Area 1, 1.5% in Area 2), relative to the FI assumed
in the HHRA (9%).
3.2.2. Site entry
The percentage of the WAH entering the site was relatively consis-
tent during the 2000–2010 period (median 4% and 9% for Areas 1 and
2, respectively) but increased significantly to 36% and 43% for those
areas in the following 2-year period, 2011–2012 (p = 0.03 for both
Fig. 4. Site Comparison to Global Reference Concentrations –Comparison of metals concentrations in caribouharvested from the Red Dog area with metals concentrations in caribou and
reindeerfrom Canada and Scandinavia (Borch-Johnsenet al., 1996; Chan et al., 2001;Crete et al., 1989; Elkinand Bethke, 1995; Kuhnlein and Soueida,1992; Larter and Nagy,2000; Rintala
et al., 1995).
1345M.R. Garry et al. / Science of the Total Environment 619–620 (2018) 1340–1348
Area 1 and Area 2) (Table 1). This finding is consistent with the in-
creased fractional site use recorded for those years.
3.2.3. Individual fractional site use
The percentage of time spent by individual WAH caribou in the site
area was consistently low throughout the study period (Table 1). From
2000 to 2010, the median site use was1.1% for Area 1 and 1.5% for Area
2. During the 2011–2012 period, despite increased numbers of animals
entering the site (Table 1), the median percentage of time that these an-
imals spent in Area 1 decreased significantly to 0.6% (p= 0.05). Theme-
dian time spent in Area 2 in 2011–2012 (1.1%) did not differ statistically
from the 2000–2010 time period. Combining all years from 2000 to
2012, the median time spent in Areas 1 and 2 by animals that entered
the site at least once was 1.0% and 1.4%, respectively.
4. Discussion
The caribou monitoring program is an integral component of a com-
prehensive fugitive dust risk management plan developed by Teck Alas-
ka that continues to track caribou health and metals concentrations
(Exponent, 2014). The risk management monitoring plan stipulates an
ongoing monitoring frequency of once every 6 years. The analysis pre-
sented in this study shows that metals concentrations in caribou har-
vested near Red Dog remain low and there are no consistent trends in
caribou tissue metals, confirming that the caribou monitoring frequency
identified in the monitoring plan together with other more frequent
monitoring programs also included in that plan are adequate to verify
the continued safety of subsistence use of caribou in the area.
The results of both the ADPH study and the site-specificHHRAindi-
cated human healthrisks associated with subsistence food consumption
and harvest activities are within acceptable limits, including risks asso-
ciated with caribou consumption. As withany risk assessment, assump-
tions incorporated into the evaluation have inherent uncertainty. The
studies described here were conducted to address areas of uncertainty
related to the potential contribution of Red Dog fugitive dusts to metals
concentrations in caribou, and in response to ongoing community con-
cerns for the safety of caribou as a food source.
Results from previous and current studies described herein indicate
that metals concentrations in caribou harvested near the mine and
DMTS were similar to reference caribou from elsewhere in Northern
Alaska and to caribou and reindeer from other parts of the world.
Liver and kidney lead from Red Dog caribou harvested in 2009 were sta-
tistically significantly higher than in reference caribou, but concentra-
tions of lead, as well as cadmium and zinc, were low in all tissues
sampled (Fig. 3). In addition, the 2009 sampling did not find a consistent
trend in metals concentrations compared to 2002. Thus, fugitive dust
emissions from Red Dog Operations do not appear to cause a meaning-
ful increase in metal concentrations in caribou. Although a formal trend
analysis was not yet possible given the few available sampling events,
these data will form the basis for future analyses, which willbe conduct-
ed following the next monitoring period.
Evidence inthe literature suggests that organ metals concentrations
in caribou and other animals are age related and consequently, metals
Fig. 5. Temporal Analysis of Site Caribou –Comparison of metals concentrations in caribou harvested from the Red Dog area in 1996, 2002, and 2009. * indicates sample concentrations
from caribou harvested in 2009 are statistically significantly different than sample concentrations from caribou harvested in 2002 (p b0.05).
1346 M.R. Garry et al. / Science of the Total Environment 619–620 (2018) 1340–1348
concentrations are often age adjusted prior to analysis. However, the re-
lationship for a given metal and tissue does not appear to be consistent
from one population to another inthese studies. For example, Larter and
Nagy (2000) analyzed metals concentrations in the kidneys of two car-
ibou populations in northern Canada—the Bluenose herd in the North-
west Territories and an arctic herd on Banks Island. Kidney cadmium
concentrations were positively correlated with age in the Bluenose
herd (r=0.64,P= 0.002). But in the Canadian arctic population stud-
ied by Larter and Nagy (2000), kidney cadmium concentrations were
negatively correlated with age as a result of one outlying data point (r
=−0.88, P= 0.05). Excluding the outlying animal, there was no signif-
icant relationship between age and cadmium. There was also no signif-
icant relationship between kidney lead concentration and age. With
respect to the Red Dog results, by comparison with the 2002 data,
there were four results that differed significantly in the 2009 Red Dog
caribou data: muscle cadmium and lead were lower, and muscle and
kidney zinc were higher. However, after adjusting for age, only two re-
sults remained significant: muscle cadmium was lower (P= 0.03) and
muscle zinc higher (Pb0.0001) in 2009 compared to 2002 (Table 2). In
addition, muscle lead was nearly significantly lower (P= 0.06). Thus,
age-adjustment does not alter interpretation of the results.
The HHRA used an estimated value of 0.09 (i.e., 9%) for FI. This esti-
mate was based on the ratio of the total land area covered by the study
area relative to the total harvest area used by residents of Kivalina and
Noatak. This was assumed to overestimate actualsite use by caribou, be-
cause it did not take into account the full extent of the caribou home
range. The actual site use by caribou was uncertain because of the lack
of quantitative temporal data on site use. The site use study using
ADFG radiocollar satellite data from the WAH provides data for a quan-
titative estimate and demonstrates that actual site use by WAH caribou
is much lower than the estimate used in the HHRA. The best estimate of
site use from the current study is 0.001 (i.e., 0.1%), the median site use
for Area 1. Even considering site use only by caribou that entered the
site at least once in a given year, median site use (0.01 for Area 1; i.e.,
1%) is still nearly an order of magnitude lower than the estimate origi-
nally used in the HHRA (Table 3).
Individual animals can spend substantially more time at the site. For
example, one collared caribou that overwintered near Red Dog in 2011–
2012 and frequently came within the two areas of interest substantially
skewed the data for those years. The fractional site use values for that
animal in 2012, which represent the maximum for the 2000–2012
time period, were 0.094 and 0.189 for Areas 1 and 2, respectively.
Caribou do overwinter near the DMTS in some years, but the data and
field observations suggest that it is usually a small proportion of the
herd and that it happens infrequently.Furthermore, caribou tissue sam-
ples used in the HHRA have historically been collected from animals
that overwintered in the area, and therefore any site-related effect on
tissue metals concentrations would likely have been reflected in the
HHRA results.
Data from the site entry analysis indicate that, in some years, many
more caribou enter the site study area than is typical. In addition to ap-
proaching close to the road during July 2011, many WAH caribou
contacted the road as they moved southeast along the coast during
the fall migration. However, it is important to note that although theap-
proach of caribou to the DMTS during July 2011 increased the number of
animals that entered the study area compared to 2000–2010, these an-
imals spentlittle time within the study area. Despite the increased num-
ber of animals entering the study area, the fraction of time spent within
the study area by individualanimals decreased in 2011–2012. The num-
ber of caribou (or proportion of the overall herd) that comes in proxim-
ity to the Red Dog development complex in any particular year, andthe
amount of time that they spend there, are likely driven more by the
myriad of factors that influence annual caribou movements and distri-
bution rather than specific selection based on factors associated with
the development complex itself. In fact, the fractional site use estimates
based on our analysis are consistent with the fraction of the total WAH
range (157,000 mi
2
;Dau, 2015) covered by our study areas. Area 1
(269 mi
2
) makes up 0.2% and Area 2 (776 mi
2
) 0.5% of the total WAH
Table 2
Statistical comparison of caribou metal concentrations: 2009 vs. 2002 (p-values).
Cadmium Lead Zinc
Unadjusted
Muscle 0.006 ↓0.02 ↓b0.0001 ↑
Liver 0.29 n.s. 0.78 n.s. 0.34 n.s.
Kidney 0.42 n.s. 0.56 n.s. 0.003 ↑
Age-Adjusted
Muscle 0.03 ↓0.06 n.s. b0.0001 ↑
Liver 0.45 n.s. 0.68 n.s. 0.89 n.s.
Kidney 0.17 n.s. 0.62 n.s. 0.35 n.s.
Notes:
↑–statistically significantly higher in 2009 relative to 2002.
↓–statistically significantly lower in 2009 relative to 2002.
n.s.–not significantly different in 2009 relative to 2002.
Table 1
Site Use by WAH Caribou.
sss Fractional site use Site entry Fractional site by individual caribou
Year Total number
satellite collar
locations in WAH
range
Time
spent in
area 1
(percent)
Time
spent in
area 2
(percent)
Total number
satellite-collared
caribou in WAH
Site entry
area 1
(percent)
Site entry
area 2
(percent)
Number
satellite-collared
caribou that
entered area 1
Median
individual time
spent in area 1
(percent)
Number
satellite-collared
caribou that
entered area 2
Median
individual time
spent in area 2
(percent)
2000 6751 0.1 0.2 25 4 4 1 1.2 1 3.9
2001 7003 0.1 0.4 33 3 15 1 1.5 4 1.4
2002 8701 0.1 0.5 39 13 18 5 1.1 7 1.9
2003 13,736 0.1 0.1 45 7 9 2 1.3 4 1.5
2004 11,251 0.1 0.7 47 9 13 4 0.5 6 1.5
2005 7921 0.2 0.6 45 4 9 2 2.0 4 2.9
2006 6486 0.0 0.0 38 0 3 0 0.0 1 0.2
2007 10,592 0.0 0.1 52 0 8 0 0.0 4 0.6
2008 19,285 0.0 0.4 62 3 13 2 0.7 7 1.4
2009 20,554 0.2 0.3 58 5 7 3 1.0 4 2.0
2010 44,439 0.0 0.1 108 1 7 1 2.7 8 0.4
2011 43,859 0.4 0.9 115 25 30 28 0.5 32 1.2
2012 41,587 0.8 1.5 115 46 56 51 0.6 62 0.9
Time Spent in Area = (# collar locations in area/# collar locations in entire range) × 100.
Site Entry = (# collared caribou entering site at least once/# collared caribou) × 100.
Time Spent in Area = (# individual caribou collar signals in area/# collar signals in entire range for that caribou) × 100.
Area 1 is defined as that area within 5 km NW and 2 km SE of the DMTS.
Area 2 is defined as that area within 10 km of the DMTS.
1347M.R. Garry et al. / Science of the Total Environment 619–620 (2018) 1340–1348
range. These values are similar to fractional site use of 0.1% and 0.4% for
Areas 1 and 2, respectively, and less than fractional site use during rela-
tively high use periods (0.6% inArea 1 and 1.2% in Area 2 in 2011–2012).
Thus, WAH caribou do not appear to be specifically avoiding the area
around Red Dog. Rather, our data indicate actual site use is what
would be expected based on the geographic area of the site relative to
the WAH range. We did not evaluate TCH caribou site use in our study
but would not expect a higher rate of usage by TCH than WAH caribou
based on observations. This remains an area for further study.
O'Hara et al. (2003) reported that the animals harvested near Red
Dog mine in the 1996 study were from the WAH. ADFG identified the
animals collected in 2002 as belonging to the WAH, and animals in
the 2009 study as belonging to the TCH. It is possible that sampling of
animals from different herds from one sampling period to the next
could result in increased data variability and mask herd-specificdiffer-
ences in tissue metals concentrations that might exist. The current
dataset does not allow us to directly assess potential differences be-
tween the herds. In addition, as previously noted, both temporary
mixing and more permanent emigration between herds have been doc-
umented (Person et al., 2007; Parrett, 2015). Individual animals sam-
pled during a given sampling event may be associated with another
herd with a different migratory pattern. However, this is likely to have
no more influence on contact rate with the Red Dog mine and DMTS
than typical year-to-year variability in migratory patterns within one
herd. Additional sampling in future years may help better characterize
inter- and intra-herd variability.
The FI estimate in the HHRA did not take into account the time car-
ibou spend at the site, nor did it reflect the total home range of caribou
that spend time in the vicinity of Red Dog. Quantitative analyses of sat-
ellite collar data for WAH caribou showed that median site use was
0.001 (0.1%) for Area1 and 0.004 (0.4%) for Area 2 over thestudy period.
Thus, the FI of 0.09 (i.e., 9%) used in the HHRA overestimates actual site
use by a factor of approximately 20 to 90. Therefore, the FI value used in
the HHRA was protective of human health and may overestimate actual
risk by one to two orders of magnitude. This result is consistent with the
findings to date of the caribou monitoringprogram, which indicate little
if any influence from Red Dog fugitive dusts on metals concentrations in
caribou.
Along with the ADPH evaluation and the site-specificHHRA,thecar-
ibou monitoring and site use studies described here provide multiple
lines of evidence indicating that fugitive dust emissions from Red Dog
operations are not a significant source of metals in caribou. Metals levels
in caribou harvested near Red Dog remain similar to concentrations of
caribou collected elsewhere. Thus, with respect to metals, caribou re-
main safe for subsistence consumption, consistent with the prior con-
clusions of the HHRA.
Acknowledgments
This study was funded by Teck Alaska Incorporated (Teck). PTT and
GPS collars deployed in the WAH were provided, along with related
funding for caribou management, by the Alaska Department of Fish
and Game, National Park Service, Bureau of Land Management and
U.S. Fish and Wildlife Service. One author (JS) is employed by Teck
and two authors (MRG, SSS) are employed by a company that has done
additional work funded by Teck. The fourth author (JD) has not been af-
filiated with Teck. The authors thank Alaska Department of Fish and
Game for their participation in the caribou monitoring program and
the site use study. Theauthors also thankBetty Dowd, Mary Becker, Jane
Ma and Randy O'Boyle for graphics and mapping assistance.
References
ADPH, 2001. Public health evaluation of exposure of Kivalina and Noatak residents to
heavy metals from Red Dog Mine. Alaska Division of Public Health, Department of
Health and Social Services, Section of Epidemiology,and Environmental Public Health
Program, Anchorage, AK.
ATSDR, 2007. Toxicological profile for lead. Agency for Toxic Substances and Disease
Registry.
Borch-Johnsen, B., Nilssen, K.J., Norheim, G., 1996. Influence of season and diet on liver
and kidney content of essential elements and heavy metals in Svalbard reindeer.
Biol. Trace Elem. Res. 51, 235–247.
Chan, H.M., Kim, C., Leggee, D., 2001. Cadmium in caribou (Rangifer tarandus) kidneys:
speciation, effects of preparation, and toxicokinetics. Food Addit. Contam. 18 (7),
607–614.
Crete, M., Nault, R., Walsh, P., Benedetti, J., Lefebvre,M.A., Weber, J., Gagnon, J., 1989. Var-
iation in cadmium content of cariboutissues from northern Quebec. Sci.Tot. Environ.
80, 103–112.
Dau, J., 2015. Units 21D, 22A, 22B, 22C, 22D, 22E, 23, 24 and 26A. Chapter 14, pages 14-1
through 14-91. In: Harper, P., McCarthy, Laura A. (Eds.), Caribou Management Report
of Survey and Inventory Activities 1 July 2012–30 June 2014. Alaska Department of
Fish and Game (Species Management Report ADF&G/DWC/SMR-2015-4, Juneau).
Elkin, B.T., Bethke, R.W., 1995. Environmental contaminants in caribou in the Northwest
Territories, Canada. Sci. Tot. Environ. 307–321 160/161.
Exponent, 2007. DMTS Fugitive Dust Risk Assessment. Prepared for Teck Cominco Alaska
Incorporated. Exponent, Bellevue, WA November. Available at:. http://dec.alaska.gov/
spar/csp/docs/reddog/bibliography.htm#2007.
Exponent, 2014. FugitiveDust Risk Management Monitoring Plan. Prepared forTeck Alas-
ka Incorporated. Exponent, Bellevue, WA (May).
Gunn, A., Nixon, W., 2008. Rangifer Health & Body Condition Monitoring Manual.
CircumArctic Ra ngifer Monitor ing and Assessme nt Network, Akureyi, Iceland.
http://carma. caff.is/images/Organized/C ARMA/Resources/FieldProtocols/
RangiferHealthBodyConditionManualforwebe42d.pdf.
Johnson, J.S., Nobmann, E.D., Asay, E., Lanier, A.P., 2009. Dietary INTAKE of Alaska native
people in two regions and implications for health: the Alaska native dietary and sub-
sistence food assessment project. Int J Circumpolar Health 68 (2), 109–122.
Kuhnlein, H.V., Soueida, R., 1992. Use and nutrient composition of traditional BaffinInuit
foods. J. Food Comp. Anal. 5, 112–126.
Larter, N.C., Nagy, J.A., 2000. A comparison of heavy metal levels in the kidneys of high
Arctic and mainland caribo u populations in the Northwest Territories of Canada.
Sci. Tot. Environ. 246, 109–119.
National Park Service, 2001. Heavy metals in mosses and soils on six transects along the
Red Dog Mine haul road Alaska. Western Arctic National Parklands May 2001. NPS/
AR/NRTR-2001/38.
O'Hara, T., George, J.C., Blake, J., Burek, K., Carroll, G., Dau, J., Bennett, L., McCoy , C.P.,
Gerard, P., Woshner, V., 2003. Investigation of heavy metals in a large mortality
event in caribou of Northern Alaska. Arctic 56, 125–135.
Parrett,L.S., 2015. Unit 26A, Teshekpuk caribou herd.Chapter 17, pages 17–1through 17–
28. In: Harper, P., McCarthy, L.A. (Eds.), Caribou Management Report of Survey and
Inventory Activities 1 July 2012–30 June 2014. Alaska Depar tment of Fish and
Game, Species Management Report ADF&G/DWC/SMR-2015-4, Juneau.
Person, B.T., Prichard, A.K., Carroll, G.M., Yokel, D.A., Suydam, R.S., George, J.C., 2007. Dis-
tribution and Movements of the Teshekpuk Caribou Herd 1990–2005: Prior to Oil
and Gas Development 60 (3), 238–250.
Rintala, R., Venalainen, E., Hirvi, T., 1995. Heavy metals in muscle, liver, and kidney from
Finnish reindeer in 1990 −91 and 1991 −92. Bull. En viron. Contam. To xicol. 54,
158–165.
Table 3
Comparison of fractional intake estimates*.
FI used in DMTS risk assessment Satellite collar site use Satellite collar individual site use
Area 1 median Area 2 median Area 1 median Area 2 median
0.09 0.0007 0.0025 0.006 0.01
*Expressed as a fraction rather than a percent, consistent with use in risk assessment calculations.
Area 1 is defined as that area within 5 km NW and 2 km SE of the DMTS.
Area 2 is defined as that area within 10 km of the DMTS.
1348 M.R. Garry et al. / Science of the Total Environment 619–620 (2018) 1340–1348
