Subscriber access provided by FLORIDA STATE UNIV
Article
Assessing cancer risk in China from gammahexachlorocyclohexane emitted from Chinese and Indian sources Yue Xu, Chongguo Tian, Jianmin Ma, Xiaoping Wang, Jun Li, Jianhui Tang, Yingjun Chen, Wei Qin, and Gan Zhang Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/es400141e • Publication Date (Web): 28 May 2013 Downloaded from http://pubs.acs.org on June 3, 2013
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 33
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Environmental Science & Technology
1
Assessing cancer risk in China from gamma-hexachlorocyclohexane
2
emitted from Chinese and Indian sources
3 4
Yue Xu,†,‡ Chongguo Tian,†,* Jianmin Ma,§ Xiaoping Wang,‡ Jun Li,‡ Jianhui Tang,†
5
Yingjun Chen,† Wei Qin,† and Gan Zhang‡
6 7
†
Key Laboratory of Coastal Zone Environmental Processes and Ecological
8
Remediation, Yantai Institute of Coastal Zone Research(YIC), Chinese Academy of
9
Sciences(CAS);
10 11
14
Provincial
Key
Laboratory
of
Coastal
Zone
Environmental Processes, YICCAS, Yantai Shandong 264003, P. R. China ‡
12 13
Shandong
State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou, 510640, China
§
Air Quality Research Division, Science and Technology Branch, Environment Canada, 4905 Dufferin Street, Toronto, Ont, M3H 5T4, Canada
15 16 17 18 19 20 21 22 1
ACS Paragon Plus Environment
Environmental Science & Technology
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
23
Abstract
24
Three models, including an atmospheric transport model, a multimedia exposure
25
model, and a risk assessment model were used to assess cancer risk in China caused
26
by γ-HCH (gamma-hexachlorocyclohexane) emitted from Chinese and Indian sources.
27
Extensive model investigations revealed the contribution of different sources to the
28
cancer risk in China. Cancer risk in Eastern China was primarily attributable to
29
γ-HCH contamination from Chinese sources, whereas cancer risk in Western China
30
was caused mostly by Indian emissions. The contribution of fresh use of lindane in
31
India to the cancer risk in China was almost one order of magnitude higher than that
32
of the reemission of γ-HCH from Indian soils. Of total population, 58% (about 0.79
33
billion) residents in China were found to live in the environment with high levels of
34
cancer risk exceeding the acceptable cancer risk of 10-6, recommended by the United
35
States Environmental Protection Agency (U. S. EPA). The cancer risk in China was
36
mostly induced by the local contamination of γ-HCH emitted from Chinese sources,
37
whereas fresh use of lindane in India will become a significant source of the cancer
38
risk in China if Indian emissions maintain its current level.
39 40
Key words: cancer risk, China, gamma-hexachlorocyclohexane, simulation,
41
long-range atmospheric transport, Indian sources, Chinese sources
42 43 2
ACS Paragon Plus Environment
Page 2 of 33
Page 3 of 33
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Environmental Science & Technology
44
TOC ART
45
46 47 48
1. INTRODUCTION
49
Environmental exposures to toxic contaminants have received considerable
50
attention in recent decades owing to their significant risk factors in the etiology of
51
several chronic diseases, for example, cancer,1 cardiovascular disease,2 chronic
52
respiratory disease,3 and diabetes.4 These chronic diseases account for 60% of the
53
deaths worldwide, and 80% of these deaths occur in low- or middle-income
54
countries.5 In China, the largest developing country in the world, chronic diseases
55
were responsible for 70% of total mortality and 60% of total disease burden in 2002.6
56
The Chinese mortality from cancer and cardiovascular disease in 2009 was reported to
57
have a 2-fold increase compared with that in 1990.7 This inspired a number of studies
58
on the connections between human health risk and exposure levels to toxic
59
contaminants in China.1, 8 Recently, the health benefits from air pollution control have
60
generated considerable research interests.1,
9
However, previous reports only
3
ACS Paragon Plus Environment
Environmental Science & Technology
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 33
61
considered the effect of trans-provincial transport of air contaminants on human
62
health in China.9 Little attention has been paid to the influence of the trans-boundary
63
transport of contaminants on human health.
64
Persistent organic pollutants (POPs) are hardly broken down in the environment
65
and undergo long-range atmospheric transport (LRAT).10 This leads to the presence of
66
POPs not only in their source regions but also in pristine areas far from the sources.
67
Even low environmental levels of these pollutants can pose a considerable threat to
68
human health and ecosystem because they tend to be accumulated in lipid-rich tissues
69
of biota and biomagnified through terrestrial and aquatic food chains.11,
70
human health risk caused by POPs was generally studied on food, human blood or
71
other tissues at a local range, separately, an integrated regional assessment is still
72
limited.
12
While
73
Indian monsoon provides a route of favorable atmospheric transport from India
74
to China in spring and summer.13 Higher atmospheric concentration levels of many
75
POPs in India14-17 than in China18 would yield significant deposition of these
76
contaminants to Chinese environment through LRAT.19 γ-hexachlorocyclohexane
77
(HCH), listed as a new POP by the Stockholm Convention in 2009, is one of the toxic
78
chemicals extensively used in India14-17 and China18. Two formulations of HCH,
79
technical HCH and lindane, have been used widely to control agricultural and public
80
health pests, and hence formed most γ-HCH in the environment.20 Generally,
81
technical HCH contains 8–15% of γ-HCH and lindane is almost pure γ-HCH. 4
ACS Paragon Plus Environment
Page 5 of 33
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Environmental Science & Technology
82
Technical HCH has been banned globally before 2000, and China and India were the
83
largest producers and users of this pesticide in the world.20 The restricted use of
84
lindane is still permitted in India due to its low cost on disease control17, 21 although
85
the use in China has been stopped since 200322. This enables us to assess the health
86
risk of γ-HCH in China emitted from China and India, the two major sources of the
87
chemical in the world20. In general, there were limited regional atmospheric
88
monitoring programs to assess temporal and spatial patterns of γ-HCH in China and
89
India. Relatively abundant data on regional atmospheric concentrations of γ-HCH in
90
China and India were reported during 2005–2007 from several field campaigns.14, 16-18
91
To better evaluate model results using this measured data, the model was run from
92
2005 to 2007 in this investigation. The present study aims at assessing cancer risk in
93
China induced by γ-HCH emitted from different sources during 2005–2007 using
94
three models which will be described below. The major objectives are (1) to establish
95
a spatial pattern of the cancer risk level across China, (2) to distinguish the
96
contribution of γ-HCH sources in China and India to the cancer risk in China, and (3)
97
to predict the temporal trend of the cancer risk in China emitted from Chinese and
98
Indian sources.
99
2. MATERIALS AND METHODS
100
In the present study, three models, including an atmospheric transport and fate
101
model, a multimedia exposure model, and a risk assessment model, have been used to
102
assess γ-HCH concentrations in the air, soil, water, food as well as average daily doses 5
ACS Paragon Plus Environment
Environmental Science & Technology
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
103
(ADDs), and the cancer risk. A horizontal resolution of ¼° latitude by ¼° longitude
104
was adopted in the three models. The atmospheric transport model domain spans 0 to
105
60 °N and 50 to 150 °E, covering the entire China and India as shown in Supporting
106
Information (SI) Figure S1. The multimedia exposure model and risk assessment
107
model cover only China.
108
2.1 Atmospheric Transport Model
109
γ-HCH concentration levels in the air and soil as well as its atmospheric
110
transport and deposition during 2005–2007 were predicted by a modified version of
111
Canadian Model for Environmental Transport of Organochlorine Pesticides
112
(CanMETOP). The model was originally developed as a regional scale atmospheric
113
transport model covering the North American continent23 and has been extended to
114
global scale to investigate intercontinental long-range transport of lindane (considered
115
as a product containing pure γ-HCH)10. The model framework used in this study is the
116
same as that adopted in the global version of CanMETOP10 but the model domain has
117
been reduced to China/India region (see SI Figure S1) with a grid resolution of ¼°
118
latitude by ¼° longitude. Briefly, the CanMETOP is a three-dimensional atmospheric
119
dispersion model coupled with a dynamic, three soil layers, fugacity-based mass
120
balance soil-air exchange model, and a two-film model to estimate water-air gas
121
exchange. In the vertical, a terrain-following coordinate is used and 14 vertical model
122
levels from the surface to 11 km are adopted. The environmental processes included
123
in the model are transport and turbulent diffusion in the atmosphere, dry and wet 6
ACS Paragon Plus Environment
Page 6 of 33
Page 7 of 33
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Environmental Science & Technology
124
deposition, and exchange at the interfacial boundaries of air-water and soil-air, and
125
removal processes in air and water via degradation as well as from the soil via
126
diffusion, leaching, and degradation.
127
The meteorological data (including wind, pressure, temperature, precipitation,
128
etc.) were the 6-hourly objectively analyzed data from the National Centers for
129
Environmental
130
http://www.esrl.noaa.gov/psd/data/gridded/data.ncep.reanalysis.html. The data were
131
then interpolated to the model grids and time intervals (¼°×¼° latitude/longitude and
132
20-min time step). Soil residue inventory of γ-HCH with ¼° ×¼° latitude/longitude in
133
2005 (see SI Figure S2) was used as initial conditions for the model integration. Since
134
technical HCH and lindane were mainly used as pesticides in Asia, the inventory was
135
created by splitting a global soil inventory with 1° × 1° latitude/longitude using the
136
area ratios of cropland from the gridded land use data.25 The global inventory was
137
obtained from the historical use of technical HCH and lindane and updated by
138
measured soil data of γ-HCH in China.10 The global inventory was used to examine
139
intercontinental LRAT of γ-HCH.10 Since China and India have been identified as
140
major sources of γ-HCH globally, and their external sources only exert a weak
141
influence on the γ-HCH budget in the environment of the model domain, the external
142
sources are not considered here.20 Initial γ-HCH concentrations in lake and sea were
143
not taken into account due to insignificant γ-HCH reemission from these water bodies
144
across China.26,
Prediction
27
(NCEP)
reanalysis.24
Details
can
be
found
at
The numerical simulations were performed for the period of 7
ACS Paragon Plus Environment
Environmental Science & Technology
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
145
December 1, 2004 to December 31, 2007 with the first month as the model spin up
146
period. The model spin up period is taken in each simulation to eliminate any effects
147
from the initial conditions.10 SI Table S1 lists the physicochemical properties of
148
γ-HCH used in the modeling.
149
2.2 Multimedia Exposure Model
150
The exposure model took into account 14 possible routes that lead to the
151
accumulation of γ-HCH in human body. These were ingestion of cereal, vegetable,
152
edible oil, fruit, fish, meat, milk, egg and drinking water, inhalation through air and
153
airborne particles, dermal contact with airborne particles, soil and bath water. Though
154
there are certain amounts of food circulation for resident consumption among regions
155
in China or from foreign import, the exact data are unavailable. Hence, we assumed
156
that the cancer risk at each model grid point resulted from local contamination of
157
γ-HCH in the atmosphere, water, soil, and food. γ-HCH concentrations in the water,
158
plant food, and fish were calculated using a level III fugacity method. Beef, pork and
159
chicken, which are prevailing animal food for Chinese, were considered in the model
160
as meat. Concentrations of γ-HCH in the meat, milk, and egg were estimated by
161
assuming the accumulation of γ-HCH in related animals (livestock) in the model. The
162
ADD in receptors through the 14 exposure routes was also estimated in the model.
163
Details are described in section SI2 of SI.
164
2.3 Risk Assessment Model
165
The cancer risk is defined to be equal to the product of the ADD and the cancer 8
ACS Paragon Plus Environment
Page 8 of 33
Page 9 of 33
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Environmental Science & Technology
166
slope factors, which represents the number of incidences of cancer.28 For instance, a
167
cancer risk of 10-6, the acceptable cancer risk recommended by the U.S. EPA,
168
corresponds to one incident of cancer per million people due to the agent being
169
assessed. For every exposure route, the same cancer slope factor of 1.3 (per mg kg-1
170
day-1) for γ-HCH, recommended by the U.S. EPA (www.epa.gov/iris/), was adopted.
171
The total cancer risk arising from the 14 exposure routes to environmental
172
contamination of γ-HCH was used to elucidate and assess the health risk of Chinese
173
population.
174
2.4 Model Evaluation
175
Increasing γ-HCH air concentrations in China occurred in autumn,18 whereas in
176
India γ-HCH level in air increased during the autumn-winter period,15, 17 respectively.
177
Here we proposed that the seasonal increase of the air concentrations could be a result
178
of the enhanced reemission from soil due to cropland cultivation during autumn
179
harvest and planting of winter-wheat, or the direct emission owing to lindane use.16 To
180
address the increase of γ-HCH air concentrations in both countries in cold season,
181
several numerical experiments were performed. The modeled atmospheric levels of
182
γ-HCH were compared with measured data to test our hypothesis. The spearman rank
183
correlation method was applied for the assessment since this nonparametric statistical
184
technique is applicable to data with linear or non-linear trends, and is insensitive to
185
outliers.
186
The production of lindane was stopped by Chinese manufactures in 2003 due to 9
ACS Paragon Plus Environment
Environmental Science & Technology
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
187
its low profit. In that year the stock was estimated to be about 100 t in China.22 This
188
suggests that the increase in γ-HCH air concentrations in China in autumn could be a
189
result of the enhanced reemission from soil due to cropland cultivation during autumn
190
harvest and planting of winter-wheat or the direct emission owing to the illegal use of
191
the pesticide.29 Because of low stock of lindane in China,22 we postulated that
192
γ-HCH reemission due to cropland cultivation contributed primarily to its changes in
193
the atmosphere. This has been verified by two numerical experiments without and
194
with cropland cultivation (see section SI3.1) using the γ-HCH soil residue in 2005.
195
SI Figure S4 displays the correlations between modeled and measured air
196
concentrations of γ-HCH. Significant correlations between modeled and measured
197
concentrations over the model domain suggest that the soil inventory reflects fairly
198
well the spatial distribution of γ-HCH (see SI Figure S4 a and d). Modeled γ-HCH
199
concentrations in the atmosphere from the numerical experiment considering cropland
200
cultivation matched better with the measured data within China compared with that
201
without accounting for cropland cultivation (see SI Figure S4 b and e and Figure S5).
202
Further assessment was carried out by comparing modeled and measured γ-HCH air
203
concentrations at Ocean University of China (36°3’N, 120°20’E) in Qingdao, China.
204
The model simulation with cropland cultivation in autumn matched well with
205
measured γ-HCH air concentrations after October which was not caught by the model
206
scenario without cropland cultivation, as shown in SI Figure S6.
207
The predicated atmospheric concentrations by the numerical experiments without 10
ACS Paragon Plus Environment
Page 10 of 33
Page 11 of 33
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Environmental Science & Technology
208
and with cultivation were inconsistent with measured data in India (see SI Figure S4 c
209
and f). The lack of the agreement can be attributed to a sharp increase of air
210
concentrations during autumn-winter in India,14-17 indicating current widespread and
211
intensive use of lindane in the country. Cropland cultivation intensity and population
212
density were often used as surrogate parameters to build the spatial patterns of the use
213
or emission of POPs. However, we found no significant correlation between spatial
214
patterns of γ-HCH air concentration in India sites (SI Figure S3) and Indian cropland
215
cultivation intensity as well as human population density with ¼° ×¼°
216
latitude/longitude. Thus, we split the time series of the mean γ-HCH air concentration
217
(SI Figure S7) collected from those sampling sites in India (SI Figure S3) into two
218
periods according to the concentration levels and calculated the statistical values for
219
the two periods (SI Table S5). Based on the simulation with the cropland cultivation
220
as aforementioned, three model scenarios which add measured mean, maximum and
221
minimum γ-HCH concentrations (SI Table S5) to the model atmosphere over India
222
were performed in order to simulate primary air emission of lindane in India. Details
223
are presented in SI.
224
We compared only the results of the three model scenarios with measured γ-HCH
225
concentrations within India. The correlation and statistical results of measured and
226
modeled air concentrations of γ-HCH are showed in SI Figure S8 and Table S6,
227
respectively. The improvements to the modeled air concentrations can be seen for all
228
three model scenarios as compared with that accounting for cropland cultivation. The 11
ACS Paragon Plus Environment
Environmental Science & Technology
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
229
results from the scenarios adding measured maximum and minimum concentrations in
230
the respective modeling scenarios yielded a better correlation with the measured
231
atmospheric levels (SI Figure S8). The modeled air concentrations from the scenarios
232
adding measured maximum and minimum concentrations were markedly higher and
233
lower than the measured levels (SI Table S6). We then applied the binary linear
234
regression to gain modeled concentration levels comparable to the measurements in
235
India. It was found that the ensemble results of model scenarios, that are measured
236
maximum × 0.2 + minimum concentrations × 0.8, were better than other model
237
scenario results (see SI Figure S8 and Table S6).
238
To further establish the level of confidence for introducing the hypothesis of
239
increasing γ-HCH air concentration in China and India during the cold season, the air
240
concentrations simulated by the model scenarios adding measured maximum and
241
minimum concentrations and their ensemble were compared with the measurements
242
over the model region. Results are plotted in SI Figure S9. It was found that the
243
modeled concentrations from the three modeling scenarios matched well with the
244
measured ones. The Spearman correlation coefficients are 0.54 (p
