Version of Record: https://www.sciencedirect.

com/science/article/pii/S0921818115301909
Manuscript_993561bee3dd824a64266b514078b383

1 The influence of global climate change on the environmental fate of

2 persistent organic pollutants: A review with emphasis on the Northern
3 Hemisphere and the Arctic as a receptor
4
5 Authors: Jianmin Ma1,5, Hayley Hung2，Robie W. Macdonald3,4
6 Affiliations:
7 1KeyLaboratory for Environment Pollution Prediction and Control, Gansu Province,
8 Department of Environmental Science and Engineering, Lanzhou University, China
9 2Air Quality Processes Research Section, Science and Technology Branch, Environment
10 and Climate Change Canada, 4905 Dufferin Street, Toronto, Ontario, M3H 5T4 Canada
11 3Department of Environment and Geography, Centre for Earth Observation Science,

12 University of Manitoba, Manitoba R3T 2N2, Canada

4
13 Fisheries and Oceans Canada, Institute of Ocean Sciences, Sidney, British Columbia
14 V8L 4B2, Canada.
15 5CAS Center for Excellence in Tibetan Plateau Earth Sciences, Beijing 100101, China

16
17 Revised and resubmitted to: Global and Planetary Change
18
19 Key words: Climate, toxic chemicals, global distribution, bioaccumulation, Arctic
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

© 2016 published by Elsevier. This manuscript is made available under the Elsevier user license
https://www.elsevier.com/open-access/userlicense/1.0/
44 Abstract
45
46 Following worldwide bans and restrictions on the use of many persistent organic

47 pollutants (POPs) from the late 1970s, their regional and global distributions have

48 become governed increasingly by phase partitioning between environmental reservoirs,

49 such as air, water, soil, vegetation and ice, where POPs accumulated during the original

50 applications. Presently, further transport occurs within the atmospheric and aquatic

51 reservoirs. Increasing temperatures provide thermodynamic forcing to drive these

52 chemicals out of reservoirs, like soil, vegetation, water and ice, and into the atmosphere

53 where they can be transported rapidly by winds and then recycled among environmental

54 media to reach locations where lower temperatures prevail (e.g., polar regions and high

55 elevations). Global climate change, widely considered as global warming, is also

56 manifested by changes in hydrological systems and in the cryosphere; with the latter now

57 exhibiting widespread loss of ice cover on the Arctic Ocean and thawing of permafrost.

58 All of these changes alter the cycling and fate of POPs. There is abundant evidence from

59 observations and modeling showing that climate variation has an effect on POPs levels in

60 biotic and abiotic environments. This article reviews recent progress in research on the

61 effects of climate change on POPs with the intention of promoting awareness of the

62 importance of interactions between climate and POPs in the geophysical and ecological

63 systems.

64 Table of Contents
65 1. Introduction
66 2. How climate change interacts with POPs
67 2.1 Temperature-dependent physicochemical properties of POPs
68 2.2 Terrestrial and fresh water environments
69 2.3 Atmosphere

2
70 2.4 Marine environment
71 2.5 Cryospheric environment
72 3. Searching for climate change signals in POPs monitoring data
73 4. Model simulation and prediction of future POPs trends under projected climate
74 change and emission scenarios
75 5. Food web and wildlife exposure risk to POPs under a changing climate
76 6. Conclusions
77 7. Acknowledgements
78 8. References
79
80 1. Introduction
81
82 Persistent organic pollutants (POPs) are toxic substances produced by industry and

83 released to the environment either deliberately or unintentionally by human activities.

84 Within the environment, POPs are resistant to degradation, and can, therefore,

85 accumulate over long periods of time in solid, liquid or gas-phase reservoirs from which

86 they enter food webs, posing risks to ecosystems and human health. These contaminants

87 include industrial chemicals [such as polychlorinated biphenyls (PCBs)], pesticides [such

88 as dichlorodiphenyltrichloroethane (DDT)], and unintentionally produced chemicals

89 (such as dioxins and furans); among which are compounds with exceptional toxicity to

90 human development [e.g., 2,3,7,8 –tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD)] (Berg et

91 al., 2006). Owing to their persistence, POPs can transport long distances in the

92 atmosphere and oceans, thence to accumulate in water, sediments, vegetation and in the

93 crysosphere (snow, ice, permafrost) (Muir and Nordstrom, 1994; Mackay and Wania,

94 1995; Wania and Mackay, 1996; Macdonald et al., 2005; Bailey et al., 2000; Grannas et

95 al., 2013). Global thermodynamic forcing by temperature gradients has been proposed as

96 a leading mechanism to enhance the accumulation of semivolatile chemicals in cold

97 environments (Wania and Mackay, 1996). Ample evidence has been collected to

3
98 demonstrate that remote, pristine locations, like the Arctic and high mountains, far from

99 the predominant POPs sources, reflect this cold-trapping process and have been widely

100 contaminated by POPs (Goldberg,1975；Ottar, 1981; Meijer et al., 2003; Hung et al.,

101 2010). Recently, the European Union instigated investigations into linkages among

102 climate change, persistent toxic chemicals, and ecosystem risks in the Arctic through the

103 ArcRisk (Arctic Health Risks) project, with the intent of furthering the understanding of

104 POPs cycling in the environment and biota under a warming Arctic

105 (http://www.arcrisk.eu/about-us/). The cycling of POPs is driven primarily by

106 physicochemical properties of the chemicals (volatility, partitioning among phases) and

107 environmental factors, such as temperature, precipitation, winds, and currents (Simonich

108 and Hites, 1995; Dachs et al., 2002; Gobas and Maclean, 2003; Stemmler and Lammel,

109 2009; Meijer et al., 2009). In addition to these physical factors, biogeochemical cycles

110 also play crucial roles because they provide transport pathways (e.g., Blais et al., 2007)

111 and mechanisms to concentrate POPs (Lohmann et al., 2007; Nizzetto et al., 2010;

112 Macdonald et al., 2002). Efforts have been made nationally and internationally to

113 regulate or phase out POPs through domestic chemical control initiatives and

114 international environmental treaties, such as the United Nations Environment Programme

115 (UNEP) Stockholm Convention on Persistent Organic Pollutants and the United Nations

116 Economic Commission for Europe (UNECE) POPs Protocol of the Convention on

117 Long-range Transboundary Air Pollution. The Stockholm Convention on POPs, which

118 came into effect in 2004, initially identified twelve chemicals or chemical groups (the so

119 called Dirty Dozen) for regulation and elimination of production and use (www.pops.int).

120 As a result of these and other control efforts, the environmental levels of legacy POPs

4
121 have been declining worldwide, demonstrating some success of POPs regulations (Kong

122 et al., 2014). On the other hand, the number of chemicals added to the list of POPs under

123 the Stockholm Convention has been increasing, with 26 chemicals or chemical groups

124 presently listed for elimination (see http://chm.pops.int/default.aspx).

125 While long-term POPs monitoring data show generally decreasing patterns during

126 the past two decades for regulated (legacy) POPs, the influence of climate change

127 (referred to as CC hereafter) on temporal and spatial distributions of POPs in biotic and

128 abiotic environments poses a new challenge to governments and scientists. For many

129 industrial POPs and pesticides, significant release to the environment started after World

130 War II and peaked during 1970 - 1990 (Li and Macdonald, 2005). Once released at their

131 sources, POPs were transported by winds and, following atmospheric circulation patterns,

132 arrived within days at remote regions where they had never been used. In remote places

133 like the Arctic, lower temperatures promoted the partitioning of POPs from air to liquid

134 or solid, and thus POPs were deposited in snow, ice, vegetation, and soil where prevailing

135 cold temperatures helped to preserve them and, therefore, extend their residence time in

136 the environment. The distribution of POPs in the global environment can be attributed

137 largely to their physiochemical properties, which are temperature dependent and often

138 parameterized by linear relationships with temperature. With warming, POPs that had

139 been sequestered in water or terrestrial surfaces by low temperatures tend to return to the

140 atmosphere. The thawing of permafrost or melting of sea-ice and glaciers due to warming

141 provides another means for historically archived contaminants to mobilize into aquatic

142 systems (e.g., Blais et al., 2001; Bogdal et al., 2010). Subsequent to entry into aquatic

143 systems, released POPs may then enter the food chain and become biomagnified in top

5
144 predators. There is a growing body of evidence that CC results in the release of POPs

145 locally, regionally and globally from containment in historically-accumulated reservoirs.

146 The CC-induced release of archived POPs, together with direct and indirect alterations in

147 biogeochemical cycles, also due to CC, then provides opportunities for ‘surprises’ that

148 could undermine global efforts to reduce environmental and human exposure to these

149 toxic chemicals. The daunting complexity of the global environment into which POPs are

150 released (Figure 1, top panel, showing the Northern Hemisphere only) has been

151 simplified by viewing the system as a series of arrows signifying fluxes, and boxes

152 signifying reservoirs (Figure 1, bottom panel). These fluxes (release, transport,

153 degradation) and reservoirs (soil, vegetation, water, ice) are all vulnerable to CC, which

154 then leads to the re-distribution of POPs.

155

6
156 Figure 1. A simplified schematic view of the Northern Hemisphere into which POPs are released (top
157 panel). Release points for POPs lie predominantly in temperate regions from which the POPs spread
158 out into the environment on winds and currents. Initially, POPs transport is affected by large-scale
159 atmospheric climate variability patterns, like the NAO or PDO, but as they become widely distributed
160 through environmental and biological pathways, they become subject to many climate variables, in
161 particular temperature, the hydrological cycle and the organic carbon cycle. Depending strongly on
162 their physicochemical properties and ease of degradation, POPs interact variously with the Earth’s
163 surface along the transport routes as shown by flux arrows. A common approach to modeling
164 represents the Earth system as a set of arrows to indicate POPs fluxes (releases, transport, losses) and
165 boxes to indicate POPs reservoirs (air, water, soil, vegetation, ice) (bottom panel). The impact of CC
166 on the fate of POPs can then be projected; for example, by evaluating the effect of increasing
167 temperature on partitioning between phases and on degradation rates, or the effects of loss of
168 compartment size due to loss of vegetation, or the thawing of snow and ice, or the alteration of
169 foodweb structures and foraging. The circled numbers in the bottom panel refer to the sections where
170 this part of the diagram is predominantly discussed. Only selected large-scale climate variability
171 patterns discussed in the text are represented. Terrestrial-related processes (air-vegetation exchange,
172 snow-ice deposition, snow/ice melting and runoff, river input and soil erosion to lakes/oceans) apply
173 to all global regions but for simplicity are represented only for the Pacific Ocean.
174

175 Given the dramatic changes now unfolding in the cryosphere and elsewhere, an

176 increasing number of observational and modeling studies over the past decade have

177 focused on what climate change and variability imply for the transport, deposition and

178 cycling of POPs (and other contaminants) in abiotic and biotic systems (e.g., Lucia et al.,

179 2015; Manciocco et al., 2014; Armitage et al., 2013; McKinney, 2012; Kallenborn et al.,

180 2012; Bidleman et al., 2012; Sumaila et al., 2011; Bustnes et al., 2010; Bogdal et al.,

181 2009, 2010; McKinney et al,, 2009; Becker et al., 2008, 2012; Geisz et al., 2008; Thies et

182 al., 2007; Eckhardt et al., 2007; Garrett and Zhao, 2006; Stern et al., 2005; Ma et al.,

183 2005, 2006, 2010, 2011; MacLeod, 2005; Lamon et al., 2009; Wang et al., 2010a,b;

184 Komprda et al., 2013; Wöhrnschimmel et al., 2013; Rigét et al., 2010, 2013; Octaviani et

185 al., 2015; Zhao et al., 2015). Likewise, reviews have highlighted the wide variety of

186 potential CC-contaminant interactions (Noyes and Lima, 2015; Teran et al., 2012;

187 UNEP/AMAP , 2011; Kallenborn et al., 2011; Rockström et al., 2009a, b; Dauvin , 2007;

7
188 Schindler and Smol, 2006; Watson et al., 2005; Macdonald et al., 2005; Schiedek et al.,

189 2007; Noyes et al., 2009; Armitage, et al., 2011; Kallenborn et al., 2012; Teng et al.,

190 2012; Gouin et al., 2013; Grannas et al., 2013). Climate variability and climate change

191 operate on almost every aspect of a POP’s lifetime in the environment. However, due to

192 knowledge gaps in understanding the complex dynamic exchange and transport,

193 toxicological, and biological processes in POPs contamination to the environment, and

194 the lack of sufficient monitoring data, many of the modeled or predicted effects of CC on

195 POPs have not yet been confirmed (Zhao et al., 2015).

196 In the present article, we aimed to provide an overview of recent progresses in

197 understanding the associations between POPs in biotic and abiotic environments and CC.

198 The article is particularly aimed to inform geophysical and ecological scientists about the

199 importance of climate variability and change to the cycling and fate of POPs in the

200 environment.

201

202 2. How climate change interacts with POPs

203

204 2.1 Temperature-dependent physicochemical properties of POPs

205 Once released into the environment from applications in agriculture, commerce and

206 industry, POPs immediately begin to distribute themselves among available reservoirs,

207 such as air, water, soil, vegetation, snow and ice. The exchange of POPs between these

208 reservoirs depends critically on the physicochemical properties of a given POP and

209 environmental conditions. With respect to the redistribution of POPs, the important

210 intrinsic physicochemical properties, including vapour pressure (PA), Henry’s Law

211 constant (H), partition coefficients for octanol–air (KOA) and octanol-water (KOW), and

8
212 susceptibility to degradation or transformation (e.g., photolysis, metabolism, hydrolysis),

213 provide kinetic controls on how long a chemical can remain in any given compartment.

214 These thermodynamic and kinetic properties, which are functions of temperature (T), are

215 often parameterized using linear regressions or other relationships between the property

216 in question and temperature (Table 1; Wania et al., 1998; Shoeib and Harner, 2002; Cetin

217 et al., 2006; Jantunen and Bidleman, 2006). Negative linear regression relationships

218 between ln PA and 1/T and positive regression relationships between log KOA and 1/T

219 (Table 1) indicate that increasing temperature enhances volatilization by increasing PA

220 and decreasing KOA. An increase in T of 1°C has been estimated to increase the volatility

221 of a typical POP by 10-15% (Lamon et al., 2009). Such an increase in T could result in a

222 7.8% and 8.5% increase, respectively, for the total yearly volatilization fluxes of

223 hexachlorobenzene (HCB) and dichlorodiphenyldichloroethylene (DDE), a degradation

224 product of DDT (Komprda et al., 2013). Further temperature increases by 2 and 3oC

225 could lead, respectively, to increases in volatilization fluxes for HCB of 8.6% and 9.3%,

226 and for DDE of 9.4% to 10.4%. Locally, atmospheric temperatures may increase

227 episodically up to 10°C to 15°C, which would double the PA of PCB-153 with a relatively

228 higher internal energy of vaporization (91.6 kJ mol-1, Lamon et al., 2009)

229 Rising temperatures will increase Henry’s law constants, promoting higher

230 partitioning from water into air for POPs. These properties have also been linked with

231 POPs atmospheric and oceanic transport, global distribution, partitioning between

232 environmental media, secondary emission, and bioaccumulation in food webs

233 (Macdonald et al., 2005; Lamon et al., 2009; Kallenborn et al., 2012; Bustnes et al.,

234 2010). The presence of virtually every widely-used POP in the ambient environment of

9
235 remote areas like the Arctic, demonstrates the efficacy of atmospheric and oceanic

236 transport to these remote areas from the temperate regions where most of these

237 substances have been released (Figure 1). A widely-cited hypothesis to explain why

238 some POPs are found in remote, cold environments at concentrations that cause concern

239 is the effect of low temperatures on physicochemical properties that enhance deposition

240 within such environments (Mackay and Wania, 1995). This process has been termed

241 "cold condensation” (Ottar, 1981), but perhaps a better way describing it would be

242 “global fractionation” (Wania and Mackay, 1993, 1996), or “cold trapping” (Rahn and

243 Heidam, 1981). These terms refer not just to condensation but to a set of properties that

244 favour partitioning from air to water, snow or solids at low temperatures. Ambient

245 atmospheric monitoring in the Arctic and other northern territories has revealed higher

246 proportions of compounds with relatively higher volatility at environmental temperatures

247 (Wania and Mackay, 1993). von Waldow et al (2010) proposed a differential removal of

248 POPs, which better explained the long-range transport (LRT) and fractionation of PCBs

249 in the Northern Hemispheric atmosphere than did global fractionation. With differential

250 removal, the latitudinal fractionation of PCBs is driven by variation in loss rates

251 (deposition and degradation) in the atmosphere along the transport pathway from

252 emission to remote deposition, rather than just air temperature, although temperature

253 clearly affects loss rates in transport. As shown in Table 1, when temperature, T,

254 increases, the first term on the right hand side of logPA equation increases，resulting in an

255 increase of vapour pressure, whereas, logKOA becomes smaller with increasing T; both of

256 these changes favour higher gaseous concentrations in air. On the other hand, chemicals

257 with low PA and high KOA tend to remain in the source regions because the former favours

10
258 low gaseous concentrations in air and the latter results in greater tendency to sorb to

259 air-borne particles, which can be more easily deposited close to source. Abundant

260 observations and modeling studies have revealed that ‘heavier’ chemicals are less prone

261 to be transported to remote pristine areas (Wania and Mackay, 1993, 1996; Wania and Su,

262 2004).

263 Primary emission of POPs continues to contribute to atmospheric trends through

264 volatilization from poorly-contained contaminated materials and stockpiles. These

265 sources, which account for the majority of ongoing primary emissions (Breivik et al.,

266 2007), are also increased by rising temperatures (Lamon et al., 2009; Gouin et al., 2013;

267 Wöhrnschimmel et al., 2013), which led Lamon et al. to assume that the strength of

268 primary emissions for any given PCB congener would be proportional to PA, expressed

269 by the relationship ET = ET − ref exp [∆U A (1 Tref − 1 T )/ R ] , where Tref is the reference

270 temperature (Co), T is the environmental temperature, ΔUA is the internal energy of

271 vaporization, R is the ideal gas constant and ET and ET-ref are the volatilization emission

272 rates at the respective temperatures. For PCB-153, a relatively hydrophobic congener, a

273 temperature increase of 4 °C above the reference temperature would lead to an

274 approximate 1.7-fold increase in emission rate (Gouin et al., 2013). This increase alone

275 would dominate all of the other processes known to have the potential to contribute to air

276 concentration trends under climate warming.

277 2.2 Terrestrial and fresh water environments

278 Since POPs tend to partition from gas and liquid phases to organic phases (Nizzetto et al.,

279 2010), the spatial distribution of atmospheric POPs is also driven in part by gradients in

11
280 organic carbon content in soils (Simonich and Hites 1994; Gobas and McLean, 2003;

281 Macdonald et al. 2005; Cousins et al., 1999; Zheng et al., 2014). Field experiment

282 (Moeckel et al., 2008) has suggested that in the year of 2000, about 50% of the total

283 annual anthropogenic emission of PCB-153 was archived in storage compartments (soil

284 or sediments) subject to organic carbon (OC) distributions. Soils and plants act as

285 important reservoirs because POPs tend to be preferentially adsorbed or partitioned onto

286 OC or carbon-rich particles contained within terrigenous systems. Compared with fluids

287 like the atmosphere and water bodies, which transport POPs that have entered the

288 environment, organic carbons in soils and sediment provide stationary “end points”

289 where POPs may reside for longer periods of time. For most legacy POPs whose use and

290 production have been banned worldwide, the mass of legacy chemical stored locally or

291 globally is what now forces exchange among different environmental compartments.

292 Deposition of POPs out of the atmosphere and relatively faster degradation within the

293 atmosphere often lead to faster reduction in atmospheric concentrations, which produces

294 a thermodynamic imbalance between soil and air. This imbalance will force a net

295 exchange of POPs from soil to air thus producing a secondary source (Lohmann et al.,

296 2007). Determination of regional and global mass balances of POPs has provided

297 quantitative estimates of the importance of secondary sources to POPs environmental

298 cycling, particularly for the legacy POPs (Nizzetto et al., 2010). Re-emissions of

299 relatively volatile agricultural pesticides (e.g., α-HCH and HCB) and industrial chemicals

300 (e.g., PCBs) accumulated in organic soils and vegetation during high-use periods in the

301 past have now become major environmental factors explaining the response of these

302 chemicals to interannual and longer time-scale CC, characterized primarily by increasing

12
303 temperature anomalies (Ma et al., 2004a, b; Ma and Li, 2005; Gao et al., 2010; Ma and

304 Cao, 2010). The volatilization of many pesticides from either agricultural or background

305 soils presently exceeds dry deposition (including gas absorption and particle deposition),

306 leading to upward fluxes from soil to air (Komprda et al., 2013). On the other hand, the

307 strong association of POPs with OC pools in soils demands that we also consider change

308 in soil biogeochemistry (quality and quantity of OC), which is affected by processes such

309 as loss or gain of plant material, humus accumulation, and loss of soil carbon by

310 metabolism. Each of these processes has the potential to alter the thermodynamic stability

311 of POPs in the surface reservoir (Moeckel et al. 2008, 2009; Teng et al., 2012). Thus,

312 POPs cycling between surface and atmosphere in terrestrial environments is linked both

313 to physical changes (e.g., temperature and moisture) and biogeochemical changes (e.g.,

314 carbon cycle) with the latter also providing feedback in the coupling between terrestrial

315 ecosystems and the climate system (Cao and Woodward, 1998a and b; Falkowski et al.

316 2000; Xu and Chen 2006; Chen and Xu, 2010;Prentice, 2001; McGuire et al., 2009). The

317 storage and spatial distribution of soil organic carbon, and their susceptibility to change

318 with altered temperature or moisture conditions, magnifies or mitigates the feedback

319 between global change and soil carbon inventory (Karhu et al., 2014). Increasing

320 temperature and moisture and decreasing snow cover can enhance metabolism of soil

321 organic carbon. Terrestrial ecosystem POPs can be altered by soil-air exchange and

322 partitioning processes, also affected by temperature (Cabrerizo et al. 2009, 2011), while

323 soil organic carbon can be altered through fixation and metabolism, which are affected by

324 temperature, moisture and nutrients, changing the storage capacity for POPs (Cabrerizo et

325 al., 2012; Macdonald et al., 2003). For some water soluble POPs such as perfluorooctane

13
326 sulfonate (PFOS), warmer temperatures may increase plant uptake by transpiration,

327 although these changes may be offset by the effects of increased carbon dioxide in carbon

328 cycling that could reduce the activity of plant stomata and reduce plant transpiration. This

329 would in turn affect the uptake and deposition of POPs to terrestrial surfaces,

330 sequestering them and moving them into vegetation-covered soil C pools (Nizzetto et al.

331 2008; Teng et al., 2012). Bioavailability of POPs in soils (i.e., a higher proportion may be

332 in the dissolved phase in soil-water system) may increase with the predicted decline in

333 soil organic carbon content (Bellamy et al., 2005). Soil respiration induced by microbe

334 and plant metabolism, which is the second largest terrestrial C flux from the soil surface

335 to the atmosphere (Bond-Lamberty and Thomson 2010), can affect POPs partition and

336 transformation as well (Teng et al., 2012). However, POPs transfers and partitioning in

337 organic and inorganic soil matrixes at the microscale is, as yet, unclear (Doick et al.,

338 2005) because of the knowledge gaps in the understanding of the response of soil

339 respiration to temperature and soil humidity and how this will affect the capacity of soils

340 for biodegradation of POPs and OC turnover (Semple et al., 2007).

341 Overall, terrestrial carbon and biogeochemical processes are recognized for their

342 potential importance to POPs mass balances in soils. It is also clear that biogeochemical

343 cycling will change with increasing temperatures, thawing of permafrost, change in soil

344 moisture, and alteration in nutrient supply in terrestrial ecosystems.

345

14
346 Table 1. Major physicochemical properties of POPs as functions of temperature (T)

Physicochemical Expressions by temperatures*

property
Vapour pressure (PA) ln PA = -m / T + b
Partitioning coefficient  ∆U p  1 
(Kp) K p1 = K p 2 (Tref ) exp−  − 1 
 R  T Tref 

Henry’s law constant (H)   a 
H = exp 2.303 − h + bh 
  T 
Octanol–air partition log K OA = a0 /T - bo
coefficient (KOA)
First order rate constant  ∆E  1 
for degradation in air (k) k = k (Tref ) exp  − 1 
 
 R  T Tref 
347 *where T and Tref are temperature and reference temperature (°K); m and b are linear regression
348 coefficients for vapour pressure; ah and bh are linear regression coefficients for Henry’s law constant,
349 a0 and b0 are linear regression coefficients for octanol–air partition coefficient; ∆E is the activation
350 energy (J.mol-1), and R is ideal gas constant.
351
352 For freshwater systems, understanding the sources and sinks for POPs in any given

353 lake is essential to assess bioaccumulation in the lakes’ food web. Toxic substances have

354 been routinely discharged to the Great Lakes basin through waste sites, river runoff, and

355 atmospheric deposition. During the last century, these pollutants accumulated in large

356 enough quantities to warrant concern regarding the effects on human health and wildlife

357 in the Great Lakes. The adverse effects of POPs on the Great Lakes have instigated

358 efforts to clean up and reduce these chemicals through international agreements (Great

359 Lakes Water Quality Agreement (GLWQA) and the Great Lakes Binational Toxics

360 Strategy (GLBTS)). Mackay et al. (1994) and Gobas et al. (1995) have investigated major

361 pathways of POPs entering and leaving the Great Lakes by developing models that

362 incorporate dry and wet atmospheric deposition, diffusive gas-phase exchange between

363 the air and water (or absorption by lake water and volatilization from lakes),

364 water-sediment exchange, inputs from land-based sources, inflow and outflow from

15
365 interlake rivers, changes in POPs inventories in lake water and sediment, transformations

366 in water and sediment, and burial. All of these pathways have plausible links with CC.

367 Since warming favours POPs partitioning from particle and liquid phase to gas phase, it

368 can be anticipated that under global warming POPs would tend to volatilize from the

369 lakes to the air, especially for legacy POPs whose atmospheric emissions have been

370 terminated and whose life time in the air is much shorter than that in water. Change in

371 air-water exchange will, in turn, alter the balance in water-sediment exchange. Lake

372 sediment, which contains a high organic carbon content, is a major reservoir of POPs.

373 The drawdown of POPs in lake water due to net evasion to the atmosphere then promotes

374 net exchange from the large sediment reservoir back into the water column (Jeremiason

375 et al., 1998). Such feedback may be associated both with changes in the interaction of

376 organic carbon cycling between lake water and sediment, and change in mean

377 temperature, linking POPs fate in fresh water system with global change, including also

378 eutrophication. Figure 2 illustrates long term trend of perturbed water concentrations

379 (Cw’, ng L-1) due to air-water exchange and water-sediment exchange of PCB-153 in

380 Lake Ontario of the Great Lakes from 1970 to 2100 using the multi-model ensemble

381 forecasted surface air temperature (SAT) anomalies from the Intergovernmental Panel on

382 Climate Change (IPCC) 20th century 20C3M and 21st century Special Report on

383 Emissions Scenarios (SRES) - A1B scenarios (Pachauri and Reisinger, 2007). The

384 perturbed concentration is defined as the departure from the mean concentration in a

385 perturbation theory for POPs (Ma and Cao, 2010). In this theory, the concentration of a

386 persistent chemical can be defined by the sum of its mean concentration, c , plus a

387 perturbation concentration, c', produced by changes in temperature and temperature

16
388 anomalies; namely c = c + c ' , or c ' = c − c where c is averaged over a sufficiently long

389 period of time (preferably decadal or longer period of time). From its definition, c' can be

390 either positive or negative and reflects, to some extent, the potential CC signals in the

391 time series of a persistent chemical. Figure 2 shows that the perturbed water

392 concentrations of PCB-153 decline quickly due to the increasing water-air exchange

393 (deep blue line), as a result of increasing temperatures (dashed purple line), which

394 favours liquid to gas phase partitioning. Declining PCB-153 in water then leads to an

395 increasing perturbed water concentration due to enhanced water-sediment exchange

396 (solid red line), which counters the loss of PCB-153 by air-water exchange.

397

398 .
399 Figure 2. Perturbed water concentrations (Cw’, ng L-1) of PCB-153 due to air-water exchange
400 (air/water, solid deep blue line), water-sediment (water/sedi, solid red line), summed perturbed water
401 concentrations (air-water exchange + water-sediment exchange, solid deep brown line), and SAT
402 anomalies (K) from 1970 to 2100 (blue dashed line). Cw' is scaled on the left-hand-side Y axis and T'
403 is scaled on the right-hand-side Y axis. For details of perturbation model, readers are referred to Ma
404 and Cao (2010), Ma et al., (2011), and Zhao et al. (2015).
405
406 2.3. Atmosphere

407 The atmosphere is an important transport route for POPs. How far in the atmosphere a

408 chemical can travel depends on its persistence and degree of mobility (Bennett et al.,

409 1999), the latter of which is affected by many meteorological variables including

17
410 especially temperature, precipitation and particulates. Air concentrations and transport

411 properties are two major factors contributing to the regional and global distribution of

412 these toxic chemicals. Long-range atmospheric transport (LRAT) patterns of POPs are

413 complex due to rapid and often highly variable changes in meteorology, but transport

414 follows certain rules defined by the physics of atmospheric circulation and the

415 physical-chemical properties of the chemicals. Although LRAT of POPs may be difficult

416 to predict over short periods of time (days, weeks), the climatic statistics of transporting

417 events (frequency, intensity) can be used to provide a more robust estimate of LRAT on

418 seasonal, annual, and longer time scales. This is because episodic transport events

419 appearing to occur randomly may, nevertheless, exhibit stable long-term mean states and

420 general statistical properties (e.g., standard deviation, extremes). For example, northward

421 episodic atmospheric transport routes of POPs from the southern US to Canada have been

422 shown to coincide with west-northwest storm tracks approaching from the US (Ma et al.,

423 2005; Yao et al., 2007). Trans-Pacific transport of POPs also follows west to east storm

424 tracks across the North Pacific Ocean (Bailey et al., 2000; Zhang et al., 2008). These

425 storm tracks are climatological winds that repeatedly occur, but may change in position

426 and strength with climate variability. Using atmospheric concentration data for the North

427 American Great Lakes region, Ma et al. (2004a, b) sought relationships between POPs

428 concentration time series and sources of interannual climate variability, notably the El

429 Niño-Southern Oscillation (ENSO), North Atlantic Oscillation (NAO), and tropical

430 Pacific sea surface temperature anomalies (SSTA) (Ma et al., 2004a, b; Ma and Li, 2006).

431 The data show that during positive phases of the NAO, stronger than normal westerly

432 winds over POPs source regions in the Canadian Prairies coincide with enhanced POPs

18
433 air concentrations in the downwind Great Lakes region. Likewise, increase in the eastern

434 tropical Pacific sea surface temperature (SST) correlated positively with increasing POPs

435 air concentrations over the Great Lakes. Poleward atmospheric transport of POPs is also

436 influenced by climate change (Hansen, et al., 2015; Friedman et al., 2014; Friedman and

437 Selin, 2016). Signatures of atmospheric circulation teleconnection patterns, mostly on an

438 interannual scale, have been detected in atmospheric concentrations of HCHs, chlordanes,

439 and PCBs measured at Alert, Nunavut, a high Arctic site in Canada, and at Zeppelin

440 Mountain, an Arctic monitoring site of POPs operated by Norwegian Institute for Air

441 Research (NILU). The three most important atmospheric circulation modes, the ENSO,

442 NAO, and Arctic Oscillation (AO)，have also been linked to the changes in atmospheric

443 POPs concentrations in the Arctic (Ma et al., 2004a; Macdonald et al., 2005; Hung et al.,

444 2005; Becker et al., 2008). Anomalies in these atmospheric circulation modes indicate

445 changes in the intensity and position of major meridional wind streams in the Northern

446 Hemispheric atmosphere, which enhance pole-ward atmospheric transport depending on

447 phase. Detection of an association between deposition of POPs and climate change (as

448 manifested by changes in air temperature and precipitation on decadal or longer

449 time-scales) is subject to large uncertainties. These uncertainties are due primarily to: (1)

450 short and sparse air measurement data of POPs across the globe, which are not long

451 enough to discriminate CC signatures that contain decadal scale variability; (2) Changes

452 in magnitude and direction of large-scale winds as a consequence of CC are more

453 difficult to assess because such changes are not readily measurable (UNEP/AMAP, 2010),

454 and; (3) There are very limited measurements of POPs in multi-media compartments

455 (ocean, soil, ice/snow, permafrost), which makes it difficult to assess exchange of POPs

19
456 between atmosphere and these compartments, and the spatial distribution of POPs

457 globally. To address these difficulties, numerical models, as elaborated in Section 4, have

458 been used to examine the influence of long-term climate change on global distribution

459 and levels of POPs under projected CC and emission scenarios in the 21st century

460 (MacLeod et al., 2005; Lamon et al., 2009; Ma and Cao, 2010; Wöhrnschimmel et al.,

461 2013; Friedman et al., 2013; Octaviani et al., 2015).

462 Friedman et al. (2014) conducted a numerical assessment to show that the effect of

463 CC on polycyclic aromatic hydrocarbons (PAHs) transport into and out of the Arctic via

464 the atmosphere was insignificant. Due to the temperature dependence of physicochemical

465 properties (Table 1), alterations in air temperatures induced by CC may play a more

466 significant role in the LRAT of POPs compared to winds. This is demonstrated by the

467 stronger response of volatile chemicals to CC, manifested by air concentrations that

468 respond to increasing temperature, which promotes volatilization from surface reservoirs

469 to air, and speeds up degradation. Persistence and mobility of persistent chemicals in

470 mobile environmental compartments (air, ocean) are often expressed as the long-range

471 transport (LRT) potential (Bennett et al., 1999, 2001; Scheringer, 1997，2001;

472 Müller-Herold et al., 1997; van Pul et al., 1998; Beyer et al., 2000, 2003), which may be

473 formulated by a characteristic travel distance (CTD) as a function of speed (wind, current)

474 and physicochemical properties of POPs (e.g., KOA, Henry’s law constant, and the rates

475 of degradation in a given environmental medium, see Table 1). Given relationships

476 between physicochemical properties and T, the effect of CC on CTDs of POPs may be

477 estimated from projected or actual temperature increases. Figure 3 illustrates the change

478 in CTDs in air at the 3500 m height for 6 PCB congeners (from PCB-28 to PCB-180)

20
479 between 2000 and 2100 using mean T averaged over the Northern Hemisphere (NH) as

480 projected by the IPCC SRES-A1B emission scenario, estimated using 2D CanMETOP

481 model [The Canadian Model for Environmental Transport of Organochlorine Pesticides,

482 (Ma, 2010)]. CTD is defined as the product of wind speed and an overall instantaneous

483 characteristic time (Bennett et al., 1999). A constant mean wind speed has been used here

484 because the projected average annual wind speed changes little over this time period.

485 Among these congeners, the more volatile PCB-28 and -52 are primarily in the gas phase

486 and therefore have higher CTDs, but all 6 PCB congeners show the increase in CTDs

487 with the projected increase in temperature towards the end of the 21st century.

488 Depending on their physicochemical properties, POPs transport in the atmosphere as

489 gases or bound to aerosol particles, with the latter more subject to deposition and washout

490 with precipitation. PCBs, PAHs and PCDDs/PCDFs have high retention to carbon

491 particles (black carbon or soot), which might be transported long distances as fine

492 aerosols (Pekar et al., 1999). Friedman et al. (2014) evaluated the influence of secondary

493 organic aerosols (SOA) and black carbon (BC) on the LRAT of PAHs by examining

494 physicochemical characteristics of various aerosol types and gas-particle partitioning.

495 Their model simulations, which evaluated a range of conditions, showed that PAHs

496 trapped in BC upon emission yielded best agreement with observed PAHs concentration

497 data. BC, which is frequently rich in PAH, is a short-lived atmospheric climate forcer.

498 The deposition of BC onto ice and snow darkens the surface, leading to warming of the

499 lower atmosphere and melting of snow and ice (Flanner et al., 2007), which then clearly

500 has the potential to alter the LRAT, air-surface exchange of PAHs carried by the BC, and

501 subsequent movement when snow or ice melts.

21
502 Precipitation and clouds play critical roles in global climate change through either

503 climate forcing or feedback. Given that, generally, the fractional cloud cover increases

504 towards the Arctic (Ma, 2010, Ma et al., 2013, Mao et al., 2014), we may project that

505 increased partitioning and absorption of hydrophilic POPs will take place as clouds

506 transport northward to the Arctic. The understanding of the interaction between clouds

507 and POPs is not yet sufficient to predict to what extent change in fraction of cloud cover

508 due to sea-ice melting associated with CC would alter POPs environmental cycling in the

509 Arctic. Current modeling exercises suggest that CC-associated precipitation would not

510 alter the environmental fate of POPs significantly compared with temperature-mediated

511 changes (Gouin et al., 2013).

512

513
514 Figure 3. Characteristic travel distance (CTD = wind speed × overall instantaneous characteristic
515 time, Bennett et al., 1999) in air for 6 PCB congeners (from PCB-28 to PCB-180) between 2000 and
516 2100 using mean temperatures averaged over the NH as projected by the IPCC SRES-A1B emission
517 scenario and a constant mean wind speed (~14 m s-1) at 3500 m height, estimated using 2D
518 CanMETOP model (Ma., 2010)
519
520 2.4 Marine environment

22
521
522 The oceans differ significantly from the atmosphere in their response to climate

523 change and how that response might affect POPs cycling. First, the transport rates in the

524 ocean are far slower: ocean currents are typically in the 0-25 cm s-1 range compared to

525 winds, which are typically in the 0-10 m s-1 range. The Fukushima accident in 2011

526 provides an exceptionally clear example of the contrast in rates of transport: the

527 atmospheric radioactive signal (137Cs and other nuclides) from Fukushima was detected

528 in Europe within 7 days (Masson et al., 2011) and likely reached the North American

529 coast within a couple of days (Stohl et al., 2012). In contrast, the 137Cs tracer signal in the

530 ocean took a little over two years to reach the Canadian continental shelf in the Pacific

531 Ocean (Smith et al., 2015). An important consequence, therefore, of the slow transport by

532 ocean currents is that climate-mediated changes in processes (e.g., degradation,

533 photolysis, sedimentation) have a much longer time to operate over comparable transport

534 distances within the two media. Second, the ocean has an important carbon cycle that

535 includes primary production, a food web, organic carbon metabolism, a large organic

536 carbon reservoir and vertical flux of particulate matter within the ocean. The carbon

537 reservoir is especially important for POPs because many of the compounds are

538 hydrophobic and tend to partition strongly onto organic carbon-rich particles and into

539 lipids. This then leads directly to bio-uptake, bioaccumulation and biomagnification in

540 marine foodweb creating, in some cases, surprising exposures in top predators

541 (Macdonald et al., 2005). Third, with respect to polar oceans, seawater can freeze

542 producing a solid, floating cover. Sea ice, which has many internal processes that affect

543 POPs cycles (e.g., Pućko et al., 2011; 2013a; 2015), also may directly affect air-sea

544 exchange, dry and wet deposition, distribution of solar radiation, and biological habitat

23
545 and foraging. Fourth, warming of terrestrial environments leads to melting of glacial ice

546 which produces sea-level rise (SLR); we have already seen a global mean SLR of about

547 0.2 m since 1900 and a further rise by as much as 1 m as projected by the end of the

548 century (IPCC, 2013). Sea level rise together with increasing frequency or intensity of

549 storm surges would enhance erosion of contaminated soils in urban and industrial areas,

550 thereby transferring legacy POPs to coastal oceans. Secondarily, the loss of perennial

551 snow and glacial mass would also release POPs archived during earlier periods of high

552 use.

553 There are also features of global change common to both atmosphere and ocean.

554 For example, both fluid media of air and ocean are becoming loaded with anthropogenic

555 CO2 (IPCC, 2005; 2013). In the case of the ocean, the CO2 leads to acidification, which

556 has important consequences for inorganic equilibria within the ocean and for biological

557 populations (RSC, 2005; AMAP (The Arctic Monitoring and Assessment Programme),

558 2013), which may then interact with organic carbon and POPs pathways. Virtually

559 nothing is known about potential interactions between POPs and acidification, either

560 directly or indirectly, but it seems likely that the largest effects would accompany

561 changes in foodweb structure caused by acidification. The atmosphere and oceans are

562 both warming (IPCC, 2005; 2013). The heat capacity of the ocean is far larger than that

563 of the atmosphere such that the modest rate of heat increase in the upper ocean

564 (~0.11°C/decade, 1971-2010; IPCC, 2013); ~ 0.005°C yr-1 between 2006-13 for the upper

565 500 m, ~0.002°C below this down to 2000m (Roemmich et al., 2015) dominates the

566 energy imbalance in the climate system. Chemical kinetics would predict that this small

567 rate of warming of the upper ocean will generally lead to more rapid degradation and

24
568 evasion of POPs, but the increase would be hardly detectable. However, as noted above,

569 the cryosphere is special in its response to warming due to the possibility of phase change

570 between solid and liquid. In particular, the Arctic Ocean has witnessed a remarkable loss

571 of summer sea ice cover, especially in the western (Pacific) regions (e.g., Stroeve et al.,

572 2012; Wood et al., 2015). Ice buffers surface water temperatures, maintaining them near

573 the freezing point of water. Melting of ice contributes to surface stratification, which

574 prevents mixing, and loss of ice cover permits solar radiation to be absorbed into the

575 low-albedo surface layer where it goes entirely to heating. Although solar radiation is the

576 dominant source of heat to the surface mixed layer, advected heat in water entering from

577 the Pacific and Atlantic Oceans has also increased during the past two decades (Carmack

578 et al., 2015). The oceans initially obtain their POPs through atmospheric transport

579 followed by dry and wet (snow and rain) deposition, diffusive gas exchange between air

580 and surface ocean, and riverine input (Gioia et al., 2008). During the initial period of

581 POPs usage, when emissions and air concentrations were increasing, the ocean lagged the

582 atmosphere in contamination such that net deposition was from air to surface waters.

583 Once the use of a particular POP was restricted or banned, atmospheric levels began to

584 decline, and when these dropped below a certain threshold, net exchange reversed and the

585 reservoir of the contaminant accumulated in the surface ocean then became a source back

586 into the atmosphere. The first environmental evidence of this reversal of exchange was

587 demonstrated for HCH (Jantunen and Bidleman, 1995). The continuing importance of the

588 surface ocean reservoir as a secondary source of POPs contamination to the atmosphere

589 has long been followed with interest (Li et al., 2004; Jantunen et al., 2008; Bidleman et

590 al., 2007; Shen et al., 2004; Wong et al., 2010), with a recent paper suggesting that

25
591 eventually the various degradation processes in the water will remove the bulk of the

592 HCH from its last reservoir – the Arctic Ocean – by about 2040 (Pućko et al., 2013a).

593 Although the Arctic Ocean is special due to loss of ice cover and subsequent warming,

594 the reversal witnessed there has likely also taken place elsewhere and for other

595 compounds like DDT and in other oceans, such as the South Pacific (Stemmler and

596 Lammel, 2009; Zhang and Lohmann, 2010; Wong et al., 2010). On the other hand,

597 measured PCBs data in air and water across the Atlantic and Arctic in 2004 showed that,

598 on average, deposition still dominated over volatilization for PCBs in the Arctic region

599 (Gioia et al., 2008). Nevertheless, the important point to be made here is that once

600 primary sources are removed, the secondary sources to the atmosphere begin to dominate,

601 and emissions from these secondary sources are strongly affected by environmental

602 conditions, which are subject to climate change and variability (e.g., Wang et al., 2010).

603 The fate and cycling of POPs within the oceans also depend on a number of

604 time-dependent chemical, physical, and biological processes, which again emphasizes the

605 importance of the transport time set against the residence time of a POP within the

606 surface ocean. Arguably the most important processes affecting the dynamics of POPs in

607 the ocean are associated directly or indirectly with the carbon cycle. The hydrophobicity

608 of POPs can be represented by KOW, which for most POPs ranges from 104 to 108,

609 suggesting that the concentration of POPs in organic matter is 4–8 orders of magnitude

610 higher than in seawater. Hence, most of the POPs become strongly associated with the

611 various organic pools (dissolved, colloidal, particulate) in the water column. Accordingly,

612 organic carbon pathways end up becoming the dominant pathways for POPs, especially

613 when the toxic effects of these chemicals are under consideration. What makes it so

26
614 challenging to project the effects of climate change on POPs in the oceans, are the

615 numerous possibilities for bifurcations in pathway (e.g., primary production to pelagic,

616 benthic or ice foodwebs, foraging patterns, vertical flux or migration to deep ocean), and

617 for POPs concentrating processes (e.g., partitioning, biomagnification, respiration)

618 (Macdonald et al., 2002). Climate change can operate on both the bifurcations and on the

619 concentrating processes, thus producing a collateral effect on the POPs pathway.

620 While the potential significance of CC for POPs cycling in marine environments

621 has been highlighted in several review articles (e.g., Macdonald et al., 2005; Schiedek et

622 al., 2007; Noyes et al., 2009; Noyes and Lima, 2015; Kallenborn et al., 2012;

623 UNEP/AMAP expert group, 2011; O’Driscoll et al., 2014; Teran et al., 2012; Rockstrom

624 et al., 2009; Schindler and Smol, 2006), there have been no extensive field investigations

625 in the marine environment designed to produce a quantitative understanding of the effect

626 of CC on POPs. Among the processes highlighted above, air-water exchange probably

627 enjoys the strongest basis to forge linkages because there are a number of reasonably long

628 time series for atmospheric POPs concentrations. In the cryosphere, air-water exchange is

629 also facilitated by decreasing sea ice cover, which permits atmospheric deposition to

630 occur directly on the water surface and enhances the rate of progress toward equilibration

631 between air and water no matter which direction the net exchange favours (e.g., see Ma et

632 al., 2011). The aquatic carbon cycle, which begins with the production of phytoplankton

633 biomass, also plays the important role in influencing the net air–water exchange by

634 controlling the dissolved water concentration. Uptake of POPs by phytoplankton and

635 zooplankton occurs rapidly in the upper ocean (e.g., Pućko et al., 2013b). These lower

636 trophic levels then support a vertical flux of organic matter out of the ocean’s surface

27
637 mixed layer (Hwang et al., 2015; Bishop, 1989). Given the strong partitioning of many

638 contaminants to organic matter, the process of vertical flux ‘draws down’ dissolved POPs

639 concentrations in the surface layer much in the same way it draws down CO2 and,

640 likewise, forces net exchange of POPs from atmosphere to ocean to re-attain equilibration

641 (Dachs et al., 1999). The process of transfer out of the upper ocean can be passive in the

642 form of biological and inorganic detritus (e.g., Timothy et al., 2013), or larger dead

643 animals (Bidleman et al., 2013), or the vertical migration of living zooplankton (Pućko et

644 al. 2013b). The general process of removing biological matter (C, N, P, Si) from the

645 upper ocean in this way is termed the "biological pump”. In proportion to how strongly

646 any particular contaminant associates with C and lipids, or absorbs onto particulate

647 surfaces (that is, exhibits Redfield behaviour), the biological pump then becomes also a

648 contaminant pump (Dachs et al., 2002; Nizzetto et al., 2012; Åkerblom et al., 2015).

649 Using a coupled global 3-D model of oceanic mercury (Hg) and plankton dynamics and

650 carbon respiration, Zhang et al. (2015) assessed riverine input of Hg to global oceans.

651 Their results confirmed that atmospherically deposited Hg to the open ocean in the

652 middle and lower latitudes was more accessible for biological uptake than the input from

653 rivers. What is of exceptional interest with respect to CC is that that biological pump may

654 be strongly affected by climate variability and change, driven either by light or nutrient

655 supply. Evidence has been provided not only for shifts in phytoplankton productivity (e.g.,

656 Henson et al., 2010; Arrigo et al., 2015; Lowry et al., 2014; Palmer et al., 2014) but also

657 for significant change in vertical flux (Boetius et al. 2013).

658 The measurement of dissolved or particulate POPs in ocean waters, particularly in

659 deep waters, is exceptionally sparse. There is therefore practically no basis to evaluate the

28
660 effects of CC on POPs in the ocean before they enter the food web. Since it is relatively

661 simple to measure, HCH has accumulated a much better marine data base from which

662 budgets have been determined (Macdonald et al., 2000), and we now understand what

663 controls its vertical and horizontal distributions in the ocean (Li et al., 2002; Pućko et al.,

664 2012a, b). However, even for HCH the time-series data have proven sufficient only to

665 validate time stepped box models at a few points (e.g., Li et al., 2004).

666

667 2.5.Cryospheric environment

668 The cryosphere is a large component of the climate system that is particularly sensitive to

669 global warming due to the feedback inherent in altering reflective surfaces (snow, ice) to

670 energy-absorbing surfaces (soil, water, vegetation), and due to the alteration in

671 biogeochemical systems as a consequence of converting ice to water. A recent article has

672 reviewed many aspects of POPs in the cryosphere and discussed potential associations

673 with CC (Grannas et al., 2013). Within the cryosphere POPs tend to break down more

674 slowly and are subject to a number of magnifying processes, some of which are unique to

675 cold environments (Macdonald et al., 2002). A surprising finding during the 1960s was

676 widespread enrichment of some POPs in animals living in the Arctic, a region where

677 these chemicals were never produced and almost never used. In particular, top predators

678 in marine ecosystems were particularly vulnerable due partly to favorable transport and

679 deposition of POPs to cold regions (Section 2.3) followed by strong biomagnification in

680 fat-dominated food webs. Due to the relatively lower temperatures, there is a general

681 thermodynamic forcing that favours a net transfer of POPs from temperate industrial and

682 agricultural regions to cold regions like the poles and high mountains. Furthermore, these

29
683 regions have mechanisms that favour deposition to surfaces due to stronger partitioning

684 onto particulates and snow, and/or stronger partitioning into cold water (Macdonald et al.,

685 2005; Ma et al., 2011). Although the annual precipitation in the Arctic is relatively low

686 (less than 500 mm), much of the precipitation comes in the form of snow, and snow

687 prevails in winter and spring when global wind fields favour rapid air transport of

688 POPs-bearing air from temperate regions into the Arctic (Heintzenberg, 1989). These are

689 the first links in the chain that places industrial and agricultural POP into Arctic foodwebs.

690 It has long been recognized that, given its greater sorption capability, snow is a very

691 efficient scavenger of organic vapours (Goss, 1997; Wania et al., 1998, 1999; Hanot and

692 Donime,1999; Domine et al., 2002; Cabanes et al., 2003). Due to its strong potential for

693 scavenging vapour and particle-bound POPs, and transporting these toward the surface,

694 snow notably influences the transport dynamics of POPs in the NH. Once snow has

695 deposited, processes within the snow become crucial to the subsequent pathway, be it

696 back into the air, or onward into melt water (e.g., Wania, 1997). Declines in the extent of

697 snow cover as part of the changing polar climate (e.g., Comiso et al., 2014; Derksen and

698 Brown, 2012; Callaghan et al., 2011) will, therefore, affect the fate of POPs in polar and

699 alpine regions (Macdonald et al., 2003, Stocker et al., 2007). Vapour scavenging, a

700 function of the air temperature and the specific surface area (SSA) of the snow, is

701 predicted to be most pronounced for snow possessing high SSA (e.g. 1 m2/g) and at cold

702 temperature (<-10oC), resulting in scavenging ratios for semi-volatile chemicals being

703 generally higher than rain (by over an order of magnitude) (Lei and Wania, 2004). The

704 seasonal snowpack therefore serves as a temporary repository for chemicals transported

705 to higher altitudes/latitudes; high concentrations have been widely reported for a number

30
706 of different POPs in both mountain and arctic snow (e.g., Blais et al., 2001, Herbert et al.,

707 2006; Arellano et al., 2011; Enge et al., 1998).

708 The sorption of organic chemicals onto water droplets and ice surfaces has also

709 been investigated (Goss, 1997; Donaldson and Anderson, 1999; Roth et al., 2002, 2004).

710 Using equilibrium partitioning calculations, Lei and Wania (2004) compared the

711 capacities of rain and snow to scavenge gaseous and particle-bound organic chemicals

712 (HCHs, PAHs, PCBs etc.) under cold and warm atmospheric conditions. Their results

713 demonstrated that partitioning of semi-volatile organic chemicals from the vapour phase

714 to liquid water droplets, atmospheric particles and snow surfaces is favoured by lower air

715 temperatures, which then increases the importance of snow and ice melt water as

716 reservoirs of POPs (Finizio et al., 2006; Villa et al., 2006; Grynkiewicz et al., 2001,

717 2002). Ocean transport also plays a significant role to deliver some contaminants to polar

718 region (Section 2.4), particularly for compounds that partition strongly into cold water

719 like HCHs. In particular, the exceptionally strong partitioning of β-HCH into water

720 diverts its pathway from atmospheric winds to predominantly surface ocean currents on

721 the way from sources in Asia to the Arctic Ocean (Li et al., 2002). The processes of

722 particulate scavenging and air-water partitioning offer non-linear mechanisms to switch

723 between transporting mechanisms that are very sensitive to temperature. As pointed out

724 by Wania (2003), the favourable scavenging and deposition of chemicals into cold

725 environments is a ‘Goldilocks’ problem. Very strong scavenging may limit the potential

726 for a chemical to make it to the Arctic via the atmosphere (e.g., β-HCH), slightly weaker

727 scavenging permits the chemical to arrive in the Arctic where the colder temperatures

728 then favour its trapping (e.g., α-HCH), whereas very weak scavenging would permit the

31
729 chemical to pass through the Arctic in the atmosphere without depositing (e.g.,

730 chlorofluorocarbons CFCs). It is therefore the case that the temperature-sensitive

731 partitioning processes, which are obviously affected by global warming and other

732 changes associated with scavenging (rain, snow), provide variable conditions that sort

733 contaminants into environmental reservoirs and transporting systems, which then alter

734 capacities of the various reservoirs, the rates of transport (discussed above) and the sink

735 strength (Macdonald, 2007). For many POPs, the cold, dark regions beneath perennial sea

736 ice, together with very low vertical particle flux (e.g., Honjo et al., 2010; Hwang et al.,

737 2015), permits chemicals like HCH to persist long enough in near-surface waters to

738 transport over decadal time periods to even the most remote locations (Macdonald et al.,

739 2000). The small vertical particle flux and the occasional sinking of larger organisms do,

740 however, permit further transport into basin waters and sediments for chemicals that

741 partition strongly into organic particles or become concentrated in foodwebs (e.g., Sober

742 and Gustafson, 2014; Bidleman et al., 2013) .

743 The Arctic is warming at a rate of almost twice the global average. Perhaps the

744 most visible change to the globe in recent decades has been the rapid decline of arctic sea

745 ice cover toward the end of summer (Steele et al., 2008; Stroeve et al., 2012) and the loss

746 of snow cover (Derksen and Brown, 2012). The Antarctic has also received contaminants

747 through LRT (e.g.,Cabrerizo et al., 2012; Corsolini et al., 2007; Bengtson Nash, 2011).

748 However, since most POPs have been produced and used in the NH, the Arctic has

749 received a greater input of contaminants and hence experienced greater exposures to

750 biotic and abiotic environments (Figure 1). The fact that human populations live within

751 the Arctic has also underscored the risks there from POPs exposures especially for

32
752 circumpolar countries. It is worth noting that, despite the remoteness of Antarctica, POPs

753 have also been detected in this region as early as the 1960s; arriving primarily via

754 atmospheric LRT from the emission sources in the Southern Hemisphere (SH), as the

755 exchange of air masses within hemispheres occur faster than between hemispheres. Given

756 that the Antarctic is also undergoing rapid changes which could impact both chemical

757 behavior and ecosystem function, CC in the Antarctic and the SH is expected to alter the

758 cycling and bioavailability of POPs in the southern polar region (Bengtson Nash, 2011).

759 Although ambient POPs data and corresponding assessments in the Antarctic are

760 extremely scarce, it is expected that knowledge about interactions between CC and arctic

761 environments is also applicable to the endemic ecosystem in the Antaractic.

762 A direct, readily observable impact of CC on POPs cycling in the Arctic is mediated

763 through snow/ice melting, which releases contaminants deposited during earlier, high-use

764 decades, allowing them to re-enter the air (Meyer and Wania, 2008). This remobilization

765 from these old, archived reservoirs may temporarily reverse, or stall, declining trends in

766 the atmosphere due to emission controls (Zhao et al., 2015). Reduced sea ice extent

767 permits greater warming of the surface Arctic Ocean through the absorption of solar

768 radiation [the so-called albedo feedback (Calaghan et al., 2011)] . In addition, more open

769 water across the Arctic will alter pressure fields potentially leading to more unstable

770 climatic conditions, extreme weather (e.g., Serreze et al., 2000) and increase in cloud

771 cover (Overpeck et al., 2005), as well as alteration in the balance of snow and rain; each

772 of which has the potential to affect POPs pathways as discussed above.

773 Arctic sea ice, permanent ice and glaciers undergo ablation under arctic warming. As

774 reservoirs of previously deposited POPs, increased melting of perennial snow cover and

33
775 permanent ice would release POPs to air and water (Bogdal et al., 2010; Blais et al.,

776 2001). For example, Jantunen et al. (2008) reported that an abrupt increase in α-HCH

777 concentration in air during 1999 accompanied ice breakup in the central Archipelago.

778 Increasing in air concentrations of PCB-52 and 101 observed at Stórhöfði, a coastal site

779 on Iceland, could also have been brought on by sea ice retreat and deglaciation of the

780 retreating ice caps of Mýrdalsjökull and Eyjafjallajökull in close proximity to this site

781 (Hung et al., 2016; AMAP, 2013). Melting snow and ice may also release POPs directly

782 into sea water (Gieisz et al., 2008). For all of these complex interactions between glaciers,

783 snow, ice cover, ocean and atmosphere, the only way to understand consequences to

784 POPs distributions would be to collect coherent time series, something that is scarce for

785 every medium except, perhaps, the atmosphere. A recent series of studies has shown that

786 sea-ice brines provide mechanisms to increase POPs concentrations relative to sea ice and

787 to deliver them efficiently to lower trophic levels in the food chain (phytoplankton,

788 zooplankton) (Pućko et al., 2010a,b; 2011; 2012a, b; 2013b; 2015). The importance of

789 ice-brine and the age of the ice for the accumulation of POPs and their fate in the Arctic

790 has also been noted by the ArcRisk program (http://project.arcrisk.eu/). The major point

791 here with respect to recent change in the Arctic is that the conversion of much of the

792 perennial ice pack to seasonal ice (e.g.,Wood et al., 2015), which is projected to continue

793 until the entire Arctic becomes seasonally ice covered, also changes the coupling between

794 POPs deposited on the ice and the food web (Pućko et al., 2015).

795 There are many important changes going on in the Arctic other than the loss of

796 perennial ice, which include freshwater balance, the organic carbon cycle and

797 acidification (Macdonald et al., 2015). Among these changes, ones that affect primary

34
798 production (e.g., upwelling and mixing), connections between ice, water, sediment,

799 foodwebs, and the vertical flux of organic matter will clearly be important for their

800 collateral effects on POPs pathways as discussed in section 2.4. Despite the importance

801 of changes in these processes (ongoing and projected), and the potential for collateral

802 changes to POPs pathways, this topic remains almost entirely speculative.

803 Galbán-Malagón et al. (2012) have shown that PCBs in transport to the Arctic Ocean are

804 intercepted along the way by sequestration into the oceanic biological pump. Increases or

805 decreases in the biological pump mediated by changes in ice climate or nutrient supply,

806 therefore, can affect the fraction of POPs that make it into the Arctic Ocean, and the

807 air-sea exchange rate along the pathway (e.g., Dachs et al., 2002).

808 3. Searching for climate change signals in POPs monitoring data

809 There are presently three approaches that have been used to look for connections between

810 CC and POPs:

811 1) Statistical methods that examine correlations between climate indicators (e.g., NAO,

812 ENSO, AO) and POPs time series. These methods face the enormous difficulty of

813 detecting trends or cycles set against noisy backgrounds. Power of such approaches can

814 only be increased by longer time series and more monitoring points;

815 2) Coupled numerical computer models that may include representations of a few, or

816 many, of the important processes such as transport, partitioning, degradation,

817 concentration, biogeochemical cycling, food-web transfers etc.. These models then may

818 explore the effect of future climate scenarios developed, for example, by the IPCC;

819 3) Statistical inference based on comparison between observation and model simulated

820 signals.

35
821 Atmospheric POPs monitoring data arguably provide the best opportunities for

822 detecting the effects of climate, partly because they contain reliable data for a suite of

823 compounds together with climate variables (e.g., temperatures, pressures, winds), and

824 partly because some atmospheric time series have been collected long enough to

825 incorporate decadal-scale variability. For instance, Canada and the US have collaborated

826 in the assessment of atmospheric loadings of toxic contaminants over the Great Lakes

827 since 1990 (Hoff et al., 1996; Cortes et al., 1998) and Europe and Canada have

828 maintained long-term atmospheric monitoring programs at research stations in the Arctic

829 (AMAP, 2013; Kallenborn et al., 2012; Hung et al., 2005). Using the Great Lakes POPs

830 dataset sampled for 24 h every 12 days (1991 – 2003), Ma et al. (2003) provided perhaps

831 the first field evidence linking interannual climate variation with POPs concentrations in

832 the atmosphere. Specifically, they found strong correlations between the

833 seasonally-averaged atmospheric concentrations of HCB and the Multivariate ENSO

834 Index (MEI) for winter (December–February) and spring (March–May). These

835 correlations were stronger in the western upper Great Lakes than in the eastern portion of

836 the Great Lakes, and corresponded well to the spatial variability of the surface air

837 temperature in North America during El Ninõ years. Subsequent studies extended the

838 associations between POPs and other dominant indicators of interannual climate variation

839 in the NH, including the NAO, the Pacific North American (PNA) anomaly, and SSTA

840 (Ma et al., 2004a; Ma and Li, 2006; Gao et al., 2010). Using a multivariate regression

841 model, Ma and Li (2006) showed that annual global reemission of α-HCH from

842 secondary sources, its atmospheric half-life, and indices for the spring NAO and ENSO

843 could explain 78% of spring mean air concentration variance in the 1990s over the Great

36
844 Lakes. They attributed the strong interannual variability of atmospheric concentrations of

845 selected POPs to reemission from terrestrial surface reservoirs (e.g., soils, vegetation,

846 lakes), accumulated from past usage, forced by increasing temperatures associated with

847 the atmospheric teleconnection patterns. Using extended time series of measured POPs

848 air concentrations over the Great Lakes region from 1992 to 2007, Gao et al. (2010)

849 revisited relationships between seasonal mean air concentration of organochlorine

850 pesticides and major climate variables in the NH. To increase the power to extract

851 cyclical climate signals from the air-concentration time series, Gao et al. (2010) first

852 removed linear trends, which were likely driven predominantly by degradation processes

853 as manifested by atmospheric half-lives in the post-emission era. The detrended time

854 series correlated more strongly with the ENSO and NAO, and reflected the interdecadal

855 variation of the NAO index, which has been increasing since the mid-1990s, probably in

856 response to global warming (Hoerling et al., 2001).

857 Macdonald et al. (2005) have noted potential influence of the AO on the

858 environmental fate of POPs in the Arctic through significant changes in large-scale wind

859 and temperature fields in the Arctic and the NH forced by the Northern-hemisphere

860 Annular Mode (NAM) of climate system (Wallace and Thompson, 2002). Eckhardt et al.

861 (2007) assessed numerically the association of the poleward atmospheric transport of air

862 pollutants with dominant atmospheric circulation teleconnection patterns. They

863 concluded that the NAO exerts a strong control on the pollution transport into the Arctic,

864 particularly in winter and spring. Positive phase of the NAO favoured atmospheric

865 transport of air pollutants, like CO, emitted from Europe to the Arctic. However, their

866 modeling study did not include POPs. Becker et al. (2008) compared seasonal and

37
867 long-term trends of α- and γ-HCH air concentrations, collected at two arctic stations

868 (Zeppelin and Alert) with the AO Index. They found a correlation between AO

869 fluctuation and α-HCH at Zeppelin (Svalbard, Norway), but not at Alert (Ellesmere

870 Island, Canada). Further examination of the records showed that a much larger air

871 concentration difference occurred between summer and winter months when the AO was

872 in the negative phase during the 2000s. During positive AO years in the 1990s, the

873 concentration difference between summer and winter was much less pronounced.

874 Composite analysis of the influence of AO, NAO and ENSO on pole-ward atmospheric

875 transport (NCP, 2013) showed warmer conditions from mid to high latitudes in the NH

876 during the positive phase of NAO and AO. Strong warming in the Arctic is also

877 associated with the positive NAO. Stronger than normal southerly winds dominate

878 relatively higher latitudes extending from 45 oN to the Arctic. This wind pattern is

879 particularly favorable for pole-ward LRAT from Eurasia where major emission sources of

880 POPs lie mostly north of 40 oN (e.g, Breivik et al., 2007).

881 To discern decadal or longer variability in climate cycles (e.g., global warming), a

882 time-series must be longer than 30 years [the classical climate change period (Le Treut et

883 al., 2007)]. This questions the capacity to extract decadal-scale climate signals in, for

884 example, atmospheric POPs data from currently available observational datasets, which

885 extend to about 20 years. Given that the Arctic is warming at a rate of almost twice the

886 global average, which has resulted in extensive sea ice melt toward the end of summer

887 since the 2000s (Steele et al., 2008), the POPs atmospheric time series in the Arctic might

888 provide best-available datasets to detect the influence of climate change and variability

889 on the environmental cycling of POPs. Overall, long-term trends of POPs atmospheric

38
890 concentrations have tended to decrease (Hung et al., 2010, 2016; Ma et al., 2011; Kong et

891 al., 2014). However, during the 2000s the declines for many of the toxic chemicals have

892 lessened and there have been intervals where some concentrations have increased (Hung

893 et al., 2016; AMAP, 2013). Given the opposing processes at play, it is not straightforward

894 to detect the effect of CC on POPs concentrations in the Arctic and globally. Declines of

895 chemicals in air are caused primarily by decreasing emissions due to worldwide

896 regulations or restrictions, and continuing losses of the chemicals due to atmospheric

897 reaction/degradation processes. The general decline in atmospheric concentrations is

898 produced by an imbalance between supply of POPs to the atmosphere, now dominated by

899 re-emission and emissions from waste streams/ stockpiles/ old equipment containing

900 POPs, e.g. PCB transformers, and loss of POPs from the atmosphere as indicated by

901 residence time. This imbalance can be affected by CC, which may enhance re-emission

902 and emissions from landfills/stockpiles etc. (higher temperatures), but also increase rates

903 of loss (degradation, scavenging by precipitation, etc.). To detect the effect of warming

904 on POPs cycling in the Arctic, therefore, requires first accounting for, and mathematically

905 removing, the other factors. Gao et al. (2010) and Ma et al. (2011) applied a detrending

906 method to remove the linear trend in monitored POPs time series collected from the Great

907 Lakes and two high arctic monitoring sites (Zeppelin and Alert) to reveal correlations

908 between POPs and increasing air temperatures also an inverse correlation with declining

909 arctic sea ice extent (Gao et al., 2010; Ma et al., 2011). These authors also developed a

910 POPs perturbation model to elucidate the detrended data (Ma et al., 2010, 2011). The

911 model (Section 2.2) assumed that the observed POPs concentrations are the sum of the

912 mean and perturbed concentrations. Ma et al. (2011) found good agreement between

39
913 detrended ambient concentrations and perturbed (modeled) concentrations of POPs at the

914 arctic monitoring sites, implying that POPs have been remobilized into the Arctic

915 atmosphere over the past two decades as a function of CC in the Arctic.

916 There has been ongoing debate on abrupt climate change and potential "tipping

917 points" (Lenton, 2011; Duarte et al., 2012; Livina and Lenton, 2013; Holland et al., 2006).

918 One obvious tipping point would be sea ice in the Arctic Ocean, which may be signaled

919 by an abrupt decline in Arctic sea ice area to recorded low value in 2007, which has

920 persisted over time (Livina and Lenton, 2013). Sea ice cover has been monitored for far

921 longer than atmospheric POPs concentrations, and with greater accuracy since the 1970s

922 when satellite data became available (e.g., Serreze and Stroeve, 2015). The large decline

923 in September sea-ice cover, and especially after 2007, provides an obvious system change

924 that could manifest itself in a large effect on air-sea exchange of POPs. Modeled

925 perturbations of air concentration and water-air exchange flux of PCB-28 set against the

926 rapid decline of sea ice cover in 2007 (Figure 4) shows that both the perturbed

927 concentration anomaly Ca' and the air-water exchange flux increased considerably in

928 2007, corresponding well with the large negative summer sea ice anomaly in the same

929 year. Using various statistical techniques, Zhao et al. (2015) examined years exhibiting

930 step changes in the time series of ambient POPs atmospheric concentrations measured at

931 four arctic POPs monitoring sites. Most step changes, particularly for PCBs, were found

932 in 2001-2002 and 2007-2008, with the latter corresponding to the step decrease in arctic

933 sea ice concentration up to that time. Their finding suggests that, although over the long

934 term the atmospheric POPs concentrations will decline due to decreasing primary

935 emissions and general degradation in the environment, the sea-ice melt and strong

40
936 warming in the Arctic, especially from 2007 onward, will render the effect of arctic

937 climate change on POPs environmental fate more detectable due to more rapid air-sea

938 exchange.

939
940 Figure 4. A schematic diagram the influence of rapid decline of arctic sea ice (left panel) in 2007 on
941 modeled perturbations of air concentration and water-air exchange flux of PCB-28 (right panel).
942 Upper right panel shows summer ice extent anomaly averaged over the Arctic and perturbed air
943 concentration (pg m-3) of PCB-28, and lower right panel illustrates summer ice extent anomaly and
944 perturbed air-water exchange flux of PCB-28.
945

946 Ice cores and sediment cores are often used to reconstruct contaminant records over

947 many years and the history of deposition at the core site from which the contaminant use,

948 emission, atmospheric transport, and other processes may be inferred. Ice and sediment

949 core analysis therefore becomes an important option to investigate decadal or longer time

950 scale effects of climate on POPs environmental cycling. Bogdal et al. (2009) analyzed

951 POPs records in dated sediment cores collected from a hydroelectric proglacial reservoir

952 lake fed by melt water from the Oberaar Glacier in the Alps. They found that, since the

953 late 1990s, the input of many POPs into this high-alpine lake increased sharply such that

954 by the mid 2000s, input fluxes of some of POPs were similar to or even higher than they

955 were in the 1960s-1970s when the chemicals were heavily used. The authors proposed

956 that the sharp increase in POPs level associated well with the accelerated melting of local

41
957 alpine glaciers due to climate warming. Likewise, Wang et al. (2010) developed DDT

958 and HCH profiles in ice cores collected from Mt. Everest (the Tibetan Plateau), Mt.

959 Muztagata (the eastern Pamirs), and the Rocky Mountains from which they were able to

960 draw connections between the Mt. Everest record and the ENSO index, between the

961 Muztagata record and the Siberian High, and between the Rocky Mountain HCH record

962 and the NAO.

963 4. Model simulation and prediction of future POPs trends under projected climate
964 change and emission scenarios

965 Detection of an association between POPs environmental cycling and CC is subject

966 to large uncertainties. These uncertainties are due primarily to (1) sparse air measurement

967 data of POPs across the globe that are, as yet, too short to discriminate CC signatures at

968 the decadal scale; (2) in the context of decadal or longer time-scale CC, changes in

969 magnitude and direction of large-scale winds as a consequence of CC are more difficult

970 to assess because such changes are not readily measurable (UNEP/AMAP, 2011); (3)

971 there are very limited measurements of POPs in multi-media compartments (ocean, soil,

972 ice/snow, permafrost), which makes it difficult to assess exchange of POPs between the

973 atmosphere and these compartments, and the spatial distribution of POPs across the globe.

974 To address these difficulties, modeling tools have been developed to examine the

975 influence of long-term climate change on global distribution and levels of POPs under

976 projected CC and emission scenarios in the 21st century (MacLeod et al., 2005; Lamon et

977 al., 2009; Ma and Cao, 2010; Wöhrnschimmel et al., 2013; Friedman et al., 2014;

978 Octaviani et al., 2015).

979 To run POPs projections, these models usually prescribe CC and POPs emission

980 scenarios. Perhaps the simplest type of model views the environment as a set of boxes

42
 981 among which POPs exchange according to their physical-chemical properties (i.e., the

982 multimedia fugacity model, Mackay, 2001; Wania and Mackay, 1999). Fugacity is a

983 mathematical expression of the thermodynamic potential of a given chemical to “escape”

984 from a certain medium, which permits an evaluation of the spontaneous direction of

985 processes like diffusion and partitioning. With some assumptions, fugacity can also be

986 applied to the rate at which such processes proceed between various media. McKone et al.

987 (1996) made perhaps the first attempt to apply a fugacity model to the question of risks

988 posed by chemicals in a warming climate. For a chemical with high volatility, HCB, they

989 found that a temperature increase of up to 5 oC from the current mean temperature would

990 have little effect on risks to human health due to exposure to HCB in air and water. Using

991 a global-scale multimedia fugacity model, MacLeod et al. (2005) found that the NAO

992 would most likely have an effect on atmospheric PCB concentrations in winter and spring

993 in Northern Europe and the Arctic. This model was subsequently extended to simulate

994 changes in the global distribution of two PCB congeners, PCB-28 and PCB-153 under

995 contrasting climate scenarios (Lamon et al., 2009). Using two scenarios, one representing

996 the last twenty years of the 20th century [20CE scenario based on the third phase of the

997 Coupled Model Intercomparison Project (CMIP3) “Climate of the 20th Century

998 Experiment” (20C3M)] and the other representing the global climate under the

999 assumption of strong future greenhouse gas emissions [A2 scenario based on the 21st

1000 century climate change scenario (SRES-A2)], they found that the higher temperatures

1001 projected by the A2 scenario led primarily to increased primary and secondary emissions

1002 through volatilization and, therefore, increased atmospheric concentrations. Lamon et

1003 al.'s (2009) modeling results also suggested that climate warming was one of the most

43
1004 significant factors determining the response of POPs cycling in the environment. Using

1005 the same model, Wöhrnschimmel et al. (2013) examined the difference in emissions and

1006 transport of POPs-like chemicals to the Arctic under the 20th century base-case climate

1007 change scenario (20C3M) and the SRES-A2 future climate scenario. For the period

1008 2020-2050, they projected a 5% yr-1 increase in primary emissions of PCB-153 north of

1009 60°N due to increased industrial activity under ice-free conditions in the Arctic, and

1010 increased emissions of pesticides (e,g, α-HCH) as arable land expanded northward.

1011 Furthermore, this increase in primary emission resulted in as much as a factor of 2

1012 increase in concentrations in air and water. After primary emission ceased, concentration

1013 levels could increase up to a factor of 2 in air and 4 in water, due mostly to changes in

1014 transport and fate of chemicals under the climate change scenario.

1015 A more sophisticated modeling approach, e.g., the GEOS-Chem model, starts with

1016 a general circulation model and incorporates meteorology and its effect on chemical

1017 processes including partitioning and degradation. This model has been widely used for

1018 problems in global atmospheric transport of air pollutants (Bey et al., 2001). Using this

1019 model, Friedman et al. (2014) investigated the effects of projected 2000−2050 emissions

1020 and climate changes on the atmospheric transport of three polycyclic aromatic

1021 hydrocarbons (PAHs). By inputting projected declines in PAH emissions in 2050, they

1022 found a minor increase in modeled air concentrations of more volatile PAHs and a

1023 decrease in heavier PAHs, which tend to be particle bound in mid latitudes of the NH,

1024 suggesting a “climate penalty” for volatile (mostly in gas-phase) PAHs and “climate

1025 benefit” for less volatile, particle-bound PAHs. Their modeling results also suggested that

1026 CC signals would be most observable in the Arctic, confirming strong response of

44
1027 observed ambient POPs concentrations to arctic warming (Section 3). Using the same

1028 model, Friedman and Selin (2016) compared the impact of local sources with

1029 long-distance transport and dynamics on atmospheric PCB concentrations in the Arctic.

1030 They found that processes outside the Arctic, such as volatilization and atmospheric LRT

1031 rather than secondary volatilization from the Arctic Ocean dominated long-term changes

1032 in PCBs within the Arctic. Hansen et al. (2015) also used an atmospheric transport model

1033 to simulate hemispheric distribution, poleward atmospheric transport and fate of POPs in

1034 the Arctic under the SRES A1B climate scenario. Their results revealed increased and

1035 decreased total mass of POPs in the NH and the Arctic depending on the physicochemical

1036 properties of targeted compounds in their model simulation. Using a coupled atmospheric

1037 general circulation model (GCM) and a chemical transport model, Octaviani et al. (2015)

1038 simulated the input into and output of DDT and PCBs from the Arctic under present day

1039 (1970-1999) and future (2070-2099) climates. Their model results indicated that the input

1040 of DDT to the Arctic would increase in the future whereas the output of PCB-153 from

1041 the Arctic would increase, both due to a more frequent occurrence of the positive phase

1042 of the AO and NAO, which enhances meridional air mass transfer between the Arctic and

1043 southern latitudes.

1044 When primary emissions and degradation half-lives dominate the dynamic balance

1045 of POPs in the atmosphere (Armitage et al., 2011), variance due to the more minor effects

1046 of climate change on other factors, such as atmospheric LRT, volatilization, and air-sea

1047 exchange is not readily detectable. Using the perturbation model (Section 2.2), Ma and

1048 Cao (2010) quantified changes in concentration of POPs in various media subject to

1049 fluctuating conditions of air temperature, snow and sea ice, winds, and precipitation (see

45
1050 section 2.2). For HCHs, PCBs and HCB, the model predicted that an increase of 0.05 -

1051 0.1 K yr-1 in the air temperature would lead to 4 – 50 % increases in the perturbed air

1052 concentrations (Ma and Cao, 2010). The model was also applied to predict the

1053 perturbation in air concentrations of PCBs and HCHs likely to occur between 1991 to

1054 2100 driven by air temperature and precipitation anomalies under the IPCC SRES-A1B

1055 emission scenario. For example, the relatively volatile PCB-52 increases in the arctic

1056 atmosphere and remains at higher values till about 2040 and decreases thereafter due to

1057 degradation (Figure 5). Perturbed concentrations of α-HCH and several other lighter

1058 PCBs display similar temporal patterns. In contrast, more hydrophobic PCB-153 shows

1059 increasing trends throughout the 21st century, potentially due to its tendency to remain in

1060 environmental sinks (e.g. sediment/soil and water) for longer periods of time.

1061 Inference of the effects of CC on POPs in oceans has relied almost completely on

1062 models. Using a coupled hydrodynamic, fate and transport ocean model, O’Driscoll et al.

1063 (2014) simulated the response to CC of PCB-153, a hydrophobic POP, and γ-HCH, a

1064 soluble POP, in the North Sea. Accounting for contributions from atmospheric deposition,

1065 riverine input and exchange with adjacent seas, and using the IPCC SRES-A1B scenario,

1066 the model showed that dry gas deposition and volatilization of γ-HCH will increase in the

1067 future, and volatilization in 2090–2099 will be even greater than 2046–2055. Overall, the

1068 model simulations indicated that CC as represented by the chosen IPCC scenario will

1069 have a negligible influence on the fate and transport of the two POPs in the North Sea.

46
1070
1071 Figure 5. Perturbed atmospheric concentration (Ca’, pg m-3) of PCB-52 averaged over the Arctic from
1072 1990 to 2100，scaled on the left Y-axis (blue solid line), and mean temperature anomaly (T’, K)
1073 averaged over the Arctic for the same period of time, scaled on the right Y-axis (deep red dashed line).
1074 Shading lines indicate standard deviations of perturbed air concentrations.
1075
1076 5. Food web and wildlife exposure risk to POPs under a changing climate
1077
1078 The major environmental concern with POPs has to do with toxicity. At high enough

1079 exposures, POPs have been shown variously to cause endocrine disruption, reproductive

1080 and immune dysfunction, neurobehavioral and developmental disorders, and cancer

1081 (World Health Organization, 2010). While adverse health effects associated with

1082 exposure to POPs have been observed in high trophic level wildlife and humans,

1083 evidence specifically connecting CC and adverse effects from POPs is lacking. The

1084 problem is partly one of complexity and partly one of inadequate data (Pacyna et al.,

1085 2015). As already shown, CC can affect transport of, and exposure to, chemicals.

1086 Models like those described above could clearly be applied to both aspects in the context

1087 of CC. However, toxicity from POPs undergoing LRT in the environment results from

1088 the chronic exposure of individuals or populations to mixtures of chemicals including

1089 POPs and other substances together with other stresses brought on by climate change.

1090 These other climate-related stresses (e.g., changes in temperature extremes, changes in

1091 food webs, transmission of diseases, invasive species, to mention but a few) increase the

1092 vulnerability of any given population to the further stress of POPs (Ross et al., 2013;

47
1093 Couillard et al., 2008).

1094 There are many ways that CC can influence POPs bioaccumulation, including

1095 physical phenomena like sea ice loss or rising temperatures, and biological phenomena

1096 like altered food web structures, nutrition, and foraging behaviour. Given the strong

1097 bioaccumulation and biomagnification of many POPs in foodwebs and the persistence of

1098 POPs in the environment, our main concern is the potential for CC to release large

1099 quantities of legacy POPs presently stored in ‘safe places’ like sediments or glacial ice,

1100 thus permitting them to enter, or re-enter, foodwebs (section 2.4).

1101 The potential for CC to affect POPs bioaccumulation in some way, or to alter

1102 pathways, and thereby have an effect on human health has been discussed in a number of

1103 articles (Macdonald et al., 2003; Jenssen, 2006; Noyes, 2009; Boxall et al., 2008; Borgå

1104 et al., 2010; Carrie et al., 2010; Ng and Gray, 2011; UNEP/AMAP, 2011; Gouin et al,

1105 2013; UNEP/AMAP, 2011), but as yet there has been no quantitative linkage between

1106 POPs, CC and health.

1107 What we worry about in contaminant assessments conducted by AMAP or the

1108 Canadian Northern Contaminants Program (NCP) is the risk of POPs present to the

1109 ecosystem and humans. Risk from a toxic chemical is a product of the hazard (toxicity)

1110 multiplied by the exposure (see for example, Gilbert, 2011, Macdonald and Bewers, 1996)

1111 The toxicities, or hazards, of chemicals are, in principle, measureable through laboratory

1112 studies, although toxicity of mixtures of toxic chemicals is far from well-studied (e.g., see

1113 Ross et al., 2013). Climate change, however, does not operate on hazard – it operates on

1114 the exposure component of the risk equation. If it were not for environmental

1115 concentrating processes, most POPs would present little risk because they would dilute as

48
1116 they mix into the environment, and eventually become removed through degradation and

1117 burial in sediments. In this context, CC as projected by the IPCC would appear generally

1118 to enhance diluting processes like mixing and degradation through warming and a

1119 generally more active hydrological cycle. However, the surprising exposures to POPs of

1120 high-trophic species living far from the points of emission underscores the importance of

1121 environmental concentrating processes. We contend, therefore, that CC is likely to have

1122 its strongest effect on risk by enhancing or diminishing these concentrating processes

1123 (e.g., see Macdonald et al., 2003; Houde et al., 2011). Although there are numerous

1124 environmental processes that concentrate POPs, these may be broadly classified into two

1125 fundamental categories (Macdonald et al., 2002). The first class of concentrating

1126 mechanisms is thermodynamically spontaneous, involves partitioning between media,

1127 and has only a small effect on POPs fugacity. These processes can be described and

1128 modeled using temperature-dependent equilibrium constants as described above for the

1129 multimedia fugacity models (Section 3), and therefore one can explore the consequence

1130 of CC on, for example, atmospheric transport and atmosphere-ocean partitioning by

1131 examining the effect of temperature rise or alteration in precipitation. The second class of

1132 mechanisms requires the input of energy; that is, the processes are not thermodynamically

1133 spontaneous and produce increases in POPs fugacities. In these processes some of the

1134 medium holding the contaminant (e.g., fat tissue) is removed (metabolized) but the

1135 contaminant remains behind at ever increasing concentration. Biomagnification in the

1136 food web is the most widely-recognized example of these processes, but there are many

1137 others associated with the carbon cycle and phase change of water (Macdonald et al.,

1138 2002). These processes are far more difficult to model in the context of CC because one

49
1139 cannot predict with certainty how CC will, for example, affect populations, predation,

1140 species distributions and other details in the carbon cycle.

1141 The major exposure pathway of POPs for wildlife and humans is through diet with

1142 other entry points like drinking water, dermal contact, and inhalation of ambient air and

1143 particulates being generally of minor importance except for small animals like

1144 zooplankton where surface area to volume ratio is large (e.g., Pućko et al., 2013b).

1145 Aquatic food webs are particularly prone to biomagnification of hydrophobic,

1146 lipid-soluble POPs: on passing from air into water into the food web and on up to top

1147 aquatic predators, the cumulative concentration magnification for POPs can be as high as

1148 7 to 9 orders of magnitude! Small changes in the multiplication factor at any point,

1149 therefore, can mean the difference between innocuous contamination and toxic exposure

1150 in the top predators. Based on these considerations, it is likely that the most important

1151 effects of CC on risk will be found in high trophic level aquatic species and in humans

1152 that depend on these for food.

1153 Climate warming will increase vapour pressure and alter thermodynamic equilibria

1154 for partitioning of POPs among various environmental media. In general, one can expect

1155 that CC will cause POPs to be more actively transported in the vapour phase, but CC also

1156 will tend to favour the evasion of POPs out of the water and back into air. Warming also

1157 favours metabolism. Therefore, warming is likely to enhance risk through greater

1158 atmospheric transport, but reduce risk through partitioning and metabolism with the latter

1159 probably weighting the balance ultimately toward less risk.

1160 But what about the non-thermodynamic concentrating processes? In the Arctic

1161 marine foodweb, biomagnification proceeds from, for example, phytoplankton to

50
1162 zooplankton to polar/arctic cod to ringed seal to polar bear (Figure 6), with the higher

1163 trophic levels contributing to the traditional diet of the indigenous peoples thus presenting

1164 risk (Vorkamp et al., 2004). Is there evidence that CC has had an effect on POPs

1165 exposure in the higher trophic levels? In a series of papers on polar bears, McKinney et al.

1166 (2009; 2012; 2013; 2015) have examined how the widespread change in Arctic's sea-ice

1167 climate has impacted foraging and diet for this iconic species. Using stable isotope

1168 analyses and fatty acid fingerprinting along with POPs analyses for bear lipids, these

1169 authors have been able to show that bear diets have altered in response to changes in prey

1170 availability, and that this has altered contaminant exposure. It is, perhaps, not surprising

1171 that the best evidence assembled so far to make the case for CC-POPs interactions comes

1172 from the polar bear given that it is an apex marine predator exceptionally dependent on a

1173 sea-ice ecosystem, which has recently undergone drastic change. However, to make the

1174 case for such interactions also depends critically on having at hand appropriate

1175 time-series or baseline data that include not only POPs, but also evidence of bear diet and

1176 foraging (e.g., stable isotopes, fatty acid fingerprints). Exposures to POPs in a changing

1177 Arctic have been shown to weaken the polar bear’s immune system due to high PCBs

1178 concentrations on their own (Bernhoft et al., 2000; Diez et al., 2013, 2015).

1179 There are very few other examples that we are aware of to make firm connections

1180 between CC and POPs mostly because of the sparseness of time-series data. Recently

1181 Loseto et al. (2015) used a time series of mercury (Hg) burdens in western populations of

1182 beluga to seek connections between exposure to Hg and climate cycles represented by the

1183 AO, the Pacific Decadal Oscillation (PDO) and records of sea-ice minimum. The

1184 database, which was not ideal, consisted of 18 sampling events (females and males)

51
1185 between 1981 and 2011. Among the climate indices, the PDO was found to produce a

1186 significant relationship with an 8-year time lag. Beluga, which feed predominantly on cod

1187 and are an important source of traditional food for northerners, are clearly key species to

1188 monitor and thus a fairly long data base exists. This animal, which forages between the

1189 Bering and Beaufort Seas, is to some extent affected by ice cover for its migrations.

1190 Therefore, like the polar bear, one would expect CC to manifest itself in this animal’s

1191 exposure to POPs. Likewise, Gaden et al. (2009) used a record of Hg muscle

1192 concentration in Beaufort Sea seals collected intermittently at 9 periods between 1973

1193 and 2007 to propose a non-linear relationship between Hg exposure and sea-ice climate.

1194 No similar studies have been made for POPs in belugas; however one might expect POPs

1195 also to manifest effects of CC since they share important characteristics with Hg

1196 (widespread contamination historically from human activities, semi volatility, LRT

1197 facilitated by the atmosphere, and bioaccumulation /magnification in marine foodwebs).

1198 In this context, bowhead whales also have an inter-ocean foraging pathway (Bering and

1199 Beaufort Seas) and manifest the effects of seasonal changes in their foraging locations in

1200 the pattern of POPs carried in their fat (Hoekstra et al., 2002), but we are not aware of

1201 any studies that have examined bowhead whales in the context of POPs-CC interaction.

1202 Perhaps the most serious impediment to detecting the effects of CC on top predators

1203 in the ocean is the general inadequacy of time series. Even though we can obtain long,

1204 unbroken temporal records of climate indices at sub-annual resolution, and long records

1205 for sea-ice cover, the sampling data base for marine predators is, at best, annual but often

1206 sporadic (e.g., 1 – 5 year gaps) and frequently do not include a comprehensive data set

1207 with POPs analyses, dietary composition and other ancillary data needed to interpret the

52
1208 results. A second issue is that, while there are reliable time series for air, and some time

1209 series for selected top predators (bears, beluga, seals, narwhal), there are virtually no time

1210 series for POPs in ocean water (possibly excepting HCHs). Given all of the sources of

1211 CC-mediated variation possible between atmosphere and top aquatic predators, the lack

1212 of information on the central components of the POPs cycle makes it exceptionally hard

1213 to understand the processes involved even if correlations were found between apex

1214 feeders and, say, air trends.

1215 Another way that CC can affect POPs through the carbon cycle is by altering

1216 phytoplankton dynamics and diversity (Henson et al., 2010). As mentioned in Section 2.4,

1217 change at the lowest trophic level can affect drawdown due to POPs uptake in

1218 phytoplankton, and subsequent sinking below the mixed layer. The phytoplankton,

1219 themselves, may exhibit growth dilution and other bioenergetic processes which are

1220 passed upward into zooplankton, predator fish and animals that feed on these. In one

1221 example evaluating exposure to Hg, Foster et al., (2012) showed that even at the

1222 zooplankton trophic level, there can be substantive differences in trophic magnification

1223 factors depending on specific predator–prey linkages; there are no such data for POPs,

1224 but clearly CC can alter these linkages (UNEP/AMAP expert group, 2011), with

1225 consequences that may reduce exposure (Borgå et al., 2010) or increase exposure (Carrie

1226 et al., 2010) in predatory fish depending on the circumstances. In another example,

1227 contaminant concentrations in some top predator fish (salmon) in Lake Ontario were

1228 found to decrease with increasing summer temperatures (French et al., 2006). This link

1229 was attributed to the negative effect of warming temperatures on prey species, causing a

1230 decline in the abundance of the diet species that have a preference for cool to cold waters,

53
1231 thereby associated contamination level in fish with temperature. This finding was based

1232 on seasonal changes in temperature that were stronger than interannual and longer term

1233 temperature variability.

1234 While scientists suspect that, generally, legacy POPs concentrations in food webs

1235 are declining in response to elimination from production and use (Rigét et al., 2010), the

1236 general warming trends and other changes in global ecosystems may at times and in some

1237 places oppose this general trend. For example, Bustnes et al. (2010) compared PCB, HCB,

1238 and oxychlordane data collected from glaucous gulls (Larus hyperboreus) blood/plasma

1239 from 1997 to 2006 at Bear Island (Bjørnøya) in the Barents Sea with the AO index. Using

1240 a regression model, these authors found a negative correlation between POPs and the

1241 winter AO index and a positive relationship between POPs and the previous summer’s or

1242 winter’s AO indices. They proposed that the strongest relationship, which was between

1243 POPs and previous winter’s AO index, was produced by the poleward transport of

1244 industrial chemicals from North America and Europe favoured by positive AO conditions.

1245 Of the three chemicals, HCB had the strongest correlation with the AO index, likely due

1246 to its longer residence time in the atmosphere [with an atmospheric half-life of 306 day

1247 (Beyer et al., 2000)] and greater volatility. The consequence that gulls have relatively

1248 higher POPs concentrations in breeding seasons following years with high air transport

1249 toward the Arctic provides direct evidence that climate variability at the decadal scale can

1250 produce increasing exposure even where trends are generally downwards.

1251 Using statistical approaches similar to those used by Bustnes et al. (2010), Rigét et

1252 al. (2013) found correlations between POPs in blubber of ringed seals and several climate

1253 indices including the AO Index, winter sea-ice coverage (November–May), the number

54
1254 of sea-ice days during winter, summer mean water temperature and salinity in central

1255 western Greenland. Statistically significant correlations were found between PCB-52 and

1256 -153 and the number of winter sea ice days, α-HCH and the preceding winter's AO Index,

1257 and β-HCH and the preceding summer's salinity. Among these relationships, PCB and

1258 p,p'-DDE were positively correlated with the AO Index and α-HCH was negatively

1259 correlated with the AO Index. These results lead to conclusions similar to those of the

1260 Bustnes et al.'s study: conditions favouring poleward atmospheric transport deliver warm

1261 air masses that reduce sea ice cover and alter sea ice distribution. α-HCH is more volatile,

1262 thus readily transported in air, but partitions more strongly into cold water (H=0.082

1263 Pa·m3/mol at 0 °C, Xiao et al. , 2004) than PCB-153 (H=17 Pa· m3/mol at 0 °C, Schenker

1264 et al., 2005) or p,p’-DDE (H= 0.39 Pa·m3/mol, Shen and Wania, 2005). Decreasing sea

1265 ice area enhances air-sea exchange and thus permits α-HCH to escape from water to air,

1266 reducing its concentration in water and, subsequently, its concentration in the marine

1267 food web supporting seals. Rigét et al. (2013) proposed that years with strong ice cover

1268 (high number of sea-ice days) led to lower concentrations of PCB-153 in ringed seals due

1269 to the reduction in availability of fish and amphipods for seals to forage upon. Ocean

1270 currents have also been shown to carry climate signals. For example, positive correlations

1271 found between salinity at Fyllas Banke, west Greenland Sea and β-HCH, PCB-52, and

1272 p,p’-DDE was interpreted by Rigét et al. (2013) as evidence that these chemicals were

1273 associated with the relatively saline Irminger Current, derived from North America,

1274 rather than the East Greenland Current derived from North America. Variability in the

1275 strength of source-water currents, therefore, led to variability in the POPs concentrations.

1276 Likewise, Li et al. (2002) concluded that β-HCH was strongly deposited from the

55
1277 atmosphere into the surface waters of the Bering Sea, thence to transport into the Arctic

1278 Ocean via currents entering through Bering Strait. The strength of deposition of β-HCH

1279 into the Bering Sea depends on rainfall and the subsequent transport to the Arctic Ocean

1280 depends on the inflowing current, which is supplied partly by the Anadyr current and

1281 partly by the Alaska coastal current; these factors all vary at the decadal scale with

1282 climate (Weingartner et al., 1999; Luchin and Paneled, 2014; Woodgate et al., 2006).

1283 Therefore, for the western Arctic Ocean, β-HCH variability (and probably other POPs to

1284 some degree) is directly related to ocean current variability. Once entering the Arctic

1285 Ocean, β-HCH may then produce higher concentrations in arctic air due to outgassing

1286 forced by reduced emissions and warming water, facilitated by loss of summer ice cover.

1287 . Marine systems have provided some of the best examples of where CC and POPs

1288 interactions are occurring, POPs exposure in terrigenous ecosystems are also likely

1289 affected by CC. For example, arctic foxes feed out of both marine (ringed-seal and

1290 seabird carcasses) and terrigenous (ptarmigan, reindeer, geese) systems; Anderson et al.

1291 (2015) showed that fox diets have varied over the past two decades due to change in food

1292 availability caused by climate-related factors like increasing mortality of Svalbard

1293 reindeer and loss of access to marine food sources due to sea ice melting. These sorts of

1294 changes alter not only the magnitude of exposure [i.e., POPs are at higher concentrations

1295 in marine food than terrestrial food (Fuglei et al., 2007)], but also the composition of the

1296 POPs [see for example, Christensen et al. (2005)]. The net result of these CC effects on

1297 the two food webs led directly to reduced exposure to POPs for arctic foxes. This

1298 particular example has direct relevance to native populations in the Arctic because they

1299 depend on both terrigenous and marine ecosystems for much of their sustenance, and

56
1300 opportunities to forage on these animals depends strongly on a number of factors affected

1301 by CC including population density, migratory patterns, ice cover, and permafrost.

1302

1303

1304 Figure 6. Schematic view of PCB-153 bioaccumulation and bioamplification in typical arctic marine
1305 food web. The sources of lipid concentrations are collected from Carrizo and Gustafsson (2011, water),
1306 Kelly (2006, arctic cod, ringed seal), Muir et al. (2000, ringed seal), Vorkamp et al. (2011, ringed seal),
1307 and Letcher et al. (2010, polar bear).
1308

1309 A variety of models have addressed interactions between climate change, carbon

1310 cycle, foodwebs and POPs. For example, physical-biological coupled models have been

1311 used to project how CC might affect lower trophic levels [Nutrients, Phytoplankton,

1312 Zooplankton (NPZ)] (e.g., Lavoie et al., 2010), multi-media environmental models have

1313 been used to examine how CC might affect the interaction between POPs and the carbon

1314 cycle [e.g., the GloboPOP model applied by Armitage and Wania(2013)], or how

1315 changing ocean temperature might affect primary production and thereby affect POPs

1316 bioaccumulation (Borgå et al, 2010) and, of course, multi-media models have been

57
1317 frequently applied to the problem of POPs distributions in foodwebs (e.g., Mackay, 2001;

1318 Gouin et al., 2013). However, to our knowledge, no models incorporating

1319 physically-based GCMs to deal with climate change in the physical and lower trophic

1320 level systems, have been coupled with multimedia POPs models of the higher trophic

1321 levels in order to explore questions of climate –POPs interactions (and see UNEP/AMAP

1322 expert group, 2011). Hence, to make progress in prediction and understanding how CC

1323 alters pathways, such coupled models need to be developed. Borgå et al. (2010)

1324 constructed a model based on the hypothesis that increase in water and air temperatures

1325 would alter partitioning of organic compounds thereby affecting uptake. Their model was

1326 based on projected increases in temperature in the Arctic of 2.0°C and 4.0°C, and taking

1327 into consideration such factors as the rates of feeding, ventilation, biotransformation and

1328 growth for invertebrates and fish. In the model, a doubling of primary production over the

1329 Arctic shelves, due to the reduced sea ice cover and enhanced upwelling, led to an

1330 increase in particulate organic carbon (POC). The increase in temperature and POC

1331 together then led to reduced bioaccumulation of PCBs. Coupled bioenergetic and

1332 bioaccumulation models applied to the Great Lakes (Ng and Gray, 2011) found that

1333 bioaccumulation of PCB-77 for species in a typical predator-prey relationship (lake trout-

1334 round goby) may have unexpected responses to contaminant cycling, due to the dynamics

1335 of uptake rates, growth dilution, and interaction between species with different

1336 temperature sensitivities. In the model, metabolic effects (temperature effects on the rate

1337 of chemical uptake and loss) were confounded by trophic interactions among species with

1338 different thermal sensitivities, which would promote invasions of warm-water species.

1339 Changes in PCB-77 concentrations in fish due to projected warming in Lakes Erie and

58
1340 Superior were relatively small, particularly when compared with potential changes in

1341 food availability and food type. These and other models (e.g., see Gouin et al., 2013)

1342 emphasize that CC insofar as carbon and POPs cycles are concerned, is far more than a

1343 simple effect of temperature on thermodynamic partitioning and kinetic rates. Rather, one

1344 needs to address also a number of biological factors like growth rates, feeding

1345 preferences, foraging, and species redistributions based on thermal niches (i.e., invasions,

1346 migrations, extinctions) due to the altered climate. Gouin et al. (2013) concluded that the

1347 effects of projected global climate change on long-term chemical transport, partitioning,

1348 and fate are relatively small for persistent neutral organics, but CC-induced effects on

1349 bioaccumulation potential in aquatic foodwebs may vary substantially depending on

1350 partitioning properties and biotransformation rate constants. They also point out the many

1351 uncertainties in simulating the effect of CC scenarios on POPs cycling in biotic and

1352 abiotic environments, including primary and secondary emissions, physicochemical

1353 properties, environmental cycling, bioaccumulation, and, of course, in the projected CC

1354 scenarios themselves.

1355 6. Conclusions

1356 Persistent organic pollutants have been released to the environment predominantly

1357 since the Second World War. Although the timing of use varies between individual POPs

1358 (e.g., PCBs and DDT were used much earlier than PBDEs), they all tend to exhibit a

1359 similar pattern of use: after introduction, there is a period of rapidly increasing emissions

1360 followed by restrictions or regulations, which sharply curtail these primary emissions.

1361 Since POPs are semi-volatile, they enter the atmosphere where they may be distributed

1362 rapidly and widely. However, the tendency for POPs to partition to particulates, soil,

59
1363 water and vegetation leads to a cumulative loading of these reservoirs during the period

1364 of high use. Once the primary emissions have been put under regulation, these

1365 environmental reservoirs (secondary sources), together with stockpiles and waste streams,

1366 then become far more important in the continued cycling of POPs. Temperature and

1367 organic carbon content, which are important control factors on the secondary sources,

1368 provide the mechanisms whereby climate change may alter the stability of POPs and the

1369 relative abilities of the environmental reservoirs to retain POPs, thus affect re-cycling

1370 between reservoirs.

1371 Global warming tends to favour re-volatilization to air and farther transport of

1372 legacy POPs, including those stored in wastes and other industrial reservoirs from past

1373 use (e.g., old transformers, stockpiles, dumps). Enhanced exchange to air has been

1374 demonstrated by increasing air concentrations of POPs associated with warm phases of El

1375 Nino and NAO. In addition, the loss of frozen phases in the cryosphere (e.g., glaciers,

1376 permanent snow, permafrost, sea ice) releases legacy POPs and alters the rate of

1377 exchange of POPs between ocean and atmosphere. The advection of POPs by winds in

1378 the NH is affected by ENSO and NAO, both of which contribute to interannual variability.

1379 However, a clear linkage between POPs cycling in the atmosphere and decadal or longer

1380 time-scale climate change has not yet been demonstrated, likely due to the natural

1381 variability of the magnitude and direction of global winds. However, we can say with

1382 certainty that projected warming will generally increase the potential for POPs to undergo

1383 LRT, especially for the more volatile chemicals. Change in precipitation (amount,

1384 snow/water) associated with CC has not been shown to affect the environmental fate of

1385 POPs compared with the effect of rising temperature.

60
1386 The association between POPs fate in the ocean and CC has been identified

1387 predominantly from time series of air-water exchange. Climate warming generally

1388 favours the volatilization of POPs from surface ocean, lakes and rivers to air. Due to the

1389 atmospheric reductions in POPs concentrations produced by reduced primary emissions

1390 and degradation, we have witnessed the reversal of air-sea exchange for some POPs in

1391 certain part of oceans, such as the south Pacific Ocean and the Arctic Ocean, suggesting

1392 that these oceans have become sources to the atmosphere for these POPs. This has been

1393 especially true for POPs that have low Henry’s law constants (e.g., the HCHs). Warming

1394 of the surface ocean would tend to enhance the reversal and thereby buffer, or delay, the

1395 decline of atmospheric POPs due to emission controls. The effect of CC on air-sea

1396 exchange occurs most significantly in the Arctic Ocean partly due to the large increases

1397 in surface temperature observed in some locations and partly due to the widespread loss

1398 of seasonal ice cover, which permits more rapid exchange of POPs. Furthermore, the

1399 organic carbon cycle in the ocean likely provides an important process where CC may

1400 alter the dynamic balance of POPs between air and water. The biological pump, which

1401 comprises the vertical flux of organic-rich matter from the surface ocean to deep ocean,

1402 also removes POPs from the sea surface. Increase or decrease in the biological pump due

1403 to CC would produce an increase or decrease in air-to-water net exchange, thus affecting

1404 the relative proportion of POPs that remain in the atmosphere or enter the deep ocean.

1405 Changes in the relative proportion of POPs in air and water induced by CC may pose

1406 unknown risks which vary with locations. Little is known about the levels of tolerance of

1407 different species for different mixture combinations.

1408 Direct linkage between POPs in wildlife and climate variability has been observed

61
1409 mostly as the association between seasonal and interannual variation of atmospheric

1410 circulation and changes in POPs concentration in wildlife, or changes in exposure

1411 associated with changes in seasonal sea-ice cover. Climate induced changes in food-web

1412 structure and function provide indirect connections between CC and POPs contamination

1413 to wildlife. For example, warming and loss of ice cover permits the northward migration

1414 of marine species usually found farther south, which may produce altered exposure to

1415 POPs via food-web pathways, even if the migratory transport of POPs is small relative to

1416 quantities transported by winds or currents. Altered ice climate also presents the

1417 likelihood of altered foraging patterns and feeding opportunities for apex feeders like

1418 seals, whales, birds and bears.

1419 Presently, many of the changes in POPs cycling associated with CC remain

1420 speculative because of knowledge gaps and uncertainties in understanding the

1421 associations between CC and POPs in biotic and abiotic environment, some of which

1422 have been listed previously [e.g., UNEP/AMAP (2011)]. Perhaps the greatest hurdle to

1423 defining the role CC may play in POPs cycles is the lack of coordinated, multi-media,

1424 long-term monitoring programs for POPs, which would provide a robust basis to validate

1425 models. This challenge is compounded by the fact that monitoring data lack geographical

1426 coverage and monitoring duration that are of the same spatial and temporal scales that

1427 climate change acts. Most long-term (>20 years) monitoring data of POPs are from the

1428 Arctic, North America and western Europe while comparable datasets are lacking in large

1429 parts of the southern hemisphere, eastern Europe, Russia, and Asia. Furthermore, huge

1430 uncertainties exist in emission inventories which are essential input parameters that drive

1431 models. PCBs and HCHs are probably the only POPs for which better emission

62
1432 inventories have been developed. Encouragingly, most legacy POPs in the environment

1433 and wildlife have been declining subsequent to worldwide controls on their use. This

1434 general decline plus declines due to environmental degradation makes it difficult to

1435 discriminate the specific role that CC may have in producing the observed trends of POPs

1436 with time. Model predictions suggest that CC plays a moderating, or secondary, role in

1437 the long-term trends of POPs. It should be noted, however, that knowledge gaps and

1438 complex interactions between multiple environmental compartments make it extremely

1439 difficult to isolate the effect of CC on POPs trends as revealed by scarce time series data.

1440 Therefore, large effort is still needed to collect field data and improve model tools, and to

1441 enhance our understanding of POPs behaviors and fate in biotic and abiotic environments

1442 in a changing climate.

1443 To date, the best time series data have been collected for air, and it is these data sets

1444 against climate variables (e.g., ENSO, NAO) that have provided the best evidence for

1445 CC-POPs linkage. To be able to draw clear connections between the atmospheric record

1446 and high trophic species at risk from POPs, we need far better time-series data for

1447 intermediate components of POPs pathways including water, zooplankton and high

1448 trophic levels. Time series for vertical flux (biological pump) and its POP content would

1449 also be an exceptionally useful parameter to monitor the intermediary role played by the

1450 upper ocean in POPs pathways.

1451 The predominant effects of CC on POPs in the atmosphere can be captured by

1452 GCMs using partitioning coefficients and kinetic rates of degradation and transformation.

1453 These can be modeled by incorporating temperature into thermodynamic and kinetic

1454 processes, which has led to some success in projecting the effects of CC. This sharply

63
1455 contrasts aquatic systems in that the physical parameters, while important, are

1456 accompanied by changes in the carbon cycle which offer non-thermodynamic processes

1457 (solvent reduction) like the food-web and microbial degradation, with the result that

1458 modeling must also incorporate biology.

1459 Future progress would require the application of physical GCMs to the

1460 atmosphere-ocean system as a physical basis for exploring IPCC scenarios, coupled to

1461 realistic NPZ (nutrient, phytoplankton, zooplankton) models for the ocean to reflect the

1462 lower end of the trophic system. Additional model components would then be required to

1463 capture the behaviour of higher trophic levels, which migrate, forage widely and are,

1464 themselves, undergoing stress from CC.

1465

1466 7. Acknowledgements

1467 Jianmin Ma wishes to thank the National Science Foundation of China for financial

1468 support under grants 41371478 and 41371453. Hayley Hung would like to acknowledge

1469 the Northern Contaminants Program (NCP) (Indigenous and Northern Affairs Canada)

1470 for financial support. Robie Macdonald gratefully acknowledges the Department of

1471 Fisheries and Oceans for support under its Emeritus program.

1472

1473 8. References

1474 Åkerblom, S., Meili, M.,Bishop, K., 2015. Organic matter in rain: an overlooked influence on mercury
1475 deposition. Environmental Science & Technology Letters, 2, 128-132.
1476 AMAP, 2013. AMAP Assessment 2013: Arctic Ocean Acidification. Arctic Monitoring and
1477 Assessment Programme (AMAP), Oslo, Norway, pp. 99pp
1478 AMAP, 2014. Trends in Stockholm Convention Persistent Organic Pollutants (POPs) in Arctic Air,
1479 Human media and Biota. By: S. Wilson, H. Hung, A. Katsoyiannis, D. Kong, J. van Oostdam,
1480 F. Riget, A. Bignert. AMAP Technical Report to the Stockholm Convention. AMAP

64
1481 Technical Report No. 7 (2014), Arctic Monitoring and Assessment Programme (AMAP),
1482 Oslo, Norway, pp. 54
1483 Andersen, M. S., Fuglei, E., König, M., Lipasti, I., Åshild Ø. Pedersen, Polder, A., Yoccoz, N.G.,
1484 Routti, H., 2015. Levels and temporal trends of persistent organic pollutants (POPs) in arctic
1485 foxes (Vulpes lagopus) from Svalbard in relation to dietary habits and food availability.
1486 Science of the Total Environment, 511, 112–122.
1487 Arellano, L., Fernandez, P., Tatosova, J., Stuchlik, E., Grimalt, J.O., 2011. Long-range transported
1488 atmospheric pollutants in snowpacks accumulated at different altitudes in the Tatra Mountains
1489 (Slovakia). Environmental Science & Technology, 45, 9268-9275.
1490 Armitage, J.M., Quinn, C.L., Wania, F., 2011. Global climate change and contaminants – an overview
1491 of opportunities and priorities for modelling the potential implications for long-term human
1492 exposure to organic compounds in the Arctic. Journal of Environment Monitoring, 13(6),
1493 1432–1546.
1494 Armitage, J.M., Wania, F., 2013. Exploring the potential influence of climate change and particulate
1495 organic carbon scenarios on the fate of neutral organic contaminants in the Arctic
1496 environment. Environmental Science: Processes & Impacts, 15, 2263-2272.
1497 Arrigo, K.R., van Dijken, G., 2015. Continued increases in Arctic Ocean primary production. Progress
1498 in Oceanography, 136, 60-70.
1499 Bailey, R., Barrie, L.A., Halsall C.J., Fellin, P., Muir, D.C.G., 2000. Atmospheric organochlorine
1500 pesticides in the western Canadian Arctic: Evidence of transpacific transport. Journal of
1501 Geophysical Research, 105(D9), 11805-11811.
1502 Becker, S., Hallsall, C.J., Tych, W., Kallenborn, R., Su, Y., Hung, H., 2008. Long-term trends in
1503 atmospheric concentrations of α- and γ-HCH in the Arctic provide insight into the effects of
1504 legislation and climatic fluctuations on contaminants levels. Atmospheric Environment, 42,
1505 8225-8233.
1506 Becker, S., Halsall, C.J., Tych, W., Kallenborn, R., Schlabach, M., Manø, S., 2012. Changing sources
1507 and environmental factors reduce the rates of decline of organochlorine pesticides in the
1508 Arctic Atmosphere. Atmospheric Chemistry and Physics, 12, 4033–4044.
1509 Bellamy, P.H., Loveland, P.J., Bradley, R.I., Lark, R.M., Kirk, G.J.D., 2005. Carbon losses from all
1510 soils across England and Wales 1978-2003. Nature, 437(7056), 245-248.
1511 Bennett, D., Kastenberg, W., McKone, T.E., 1999. General formulation of characteristic time for
1512 persistent chemicals in a multimedia environment. Environmental Science & Technology, 33,
1513 503-509.
1514 Bennett, D., Scheringer, M., McKone, T.E., Hungrbühler, K., 2001. Predicting long-range transport: A
1515 systematic evaluation of two multimedia transport models. Environmental Science &
1516 Technology, 35, 1181-1189.
1517 Berg, M.V., Birnbaum, L.S., Denison, M., Vito, M.D., Farland, W., Feeley, M., 2006. The 2005
1518 World Health Organization reevaluation of human and mammalian toxic equivalency factors
1519 for dioxins and dioxin-like compounds. Toxicological Sciences, 93, 223-241.
1520 Bernhoft, A., Skaare, J.U., Wiig, Ø., Derocher, A.E., Larsen, H.J., 2000. Possible immunotoxic effects
1521 of organochlorines in polar bears (Ursus maritmus) at Svalbard. Journal of Toxicology and
1522 Environmental Health Part A, 59(7), 101-114.
1523 Bey, I., Jacob, D.J., Yantosca, R.M., Logan, J.A., Field, B.D., Fiore, A.M., Li, Q.B., Liu, H.Y.,
1524 Mickley, L.J., Schultz, M.G., 2001. Global modeling of tropospheric chemistry with
1525 assimilated meteorology: Model description and evaluation. Journal of Geophysical
1526 Research-Atmosphere, 106, 23073-23095.

65
1527 Beyer, A., Mackay, D., Matthies, M., Wania, F., Webster, E., 2000. Assessing long-range transport
1528 potential of persistent organic pollutants. Environmental Science & Technology, 34, 699-703.
1529 Beyer, A., Wania, F., Gouin, T., Mackay, D., Matthies, M., 2003. Temperature dependent of the
1530 characteristic travel distance. Environmental Science & Technology, 37, 766-771.
1531 Bidleman, T.F., Kylin, H., Jantunen, L.M., Helm, P.A., Macdonald, R.W., 2007.
1532 Hexachlorocyclohexanes (HCHs) in the Canadian Archipelago. 1. Spatial distribution and
1533 pathways of alpha-, beta- and gamma-HCHs in surface water. Environmental Science &
1534 Technology, 41(8), 2688-2695.
1535 Bidleman, T.F., Jantunen, L.M., Kurt-Karakus, P.B., Wong, F., 2012. Chiral persistent organic
1536 pollutants as tracers of atmospheric sources and fate: review and prospects for investigating
1537 climate change influences. Atmospheric Pollution Research, 3(4), 371-382.
1538 Bidleman, T.F., Stern, G.A., Tomy, G.T., Hargrave, B.T., Jantunen, L.M. Macdonald, R.W., 2013.
1539 Scavenging amphipods - sentinels for penetration of mercury and persistent organic
1540 chemicals into food webs of the deep Arctic Ocean. Environmental Science &Technology,
1541 47, 5553-5561.
1542 Bishop, J.K.B., 1989. Regional extremes in particulate matter composition and flux: effects on the
1543 chemistry of the ocean interior. In: Berger, W.H., Smetacek, V.S., Wefer, G. (Eds.),
1544 Productivity of the Ocean: Present and Past. J. Wiley & Sons Limited, Dahlem, pp. 117-137
1545 Blais, J.M., Schindler, D.W., Muir, D.C.G., Sharp, M., Donald, D., Lafreniere, M., Braekevelt, E.,
1546 Strachan, W.M.J., 2001. Melting glaciers: A major source of persistent organochlorines to
1547 subalpine Bow Lake in Banff National Park, Canada. AMBIO: A Journal of the Human
1548 Environment, 30(7), 410-415.
1549 Boetius, A., Albrecht, S., Bakker, K., Bienhold, C., Felden, J., Fernández-Méndez, M., Hendricks, S.,
1550 Katlein, C., Lalande, C., Krumpen, T., Nicolaus, M., Peeken, I., Rabe, B., Rogacheva, A.,
1551 Rybakova, E., Somavilla, R., Wenzhöfer, F., 2013. Export of algal biomass from the melting
1552 Arctic sea ice. Science, 339, 1430-1432.
1553 Bogdal, C., Schmid, P., Zennegg, M., Anselmetti, F.S., Scheringer, M., Hungerbuhler, L.K., 2009.
1554 Blast from the past: melting glaciers as a relevant source for persistent organic pollutants,
1555 Environment Science & Technology, 43, 8173-8177.
1556 Bogdal, C., Divna, N., Luthi, M.P., Schenker, U., Scheringer, M., Hungerbuhler, K., 2010. Release of
1557 legacy pollutants from melting glaciers: Model evidence and conceptual understanding.
1558 Environmental Science & Technology, 44, 4063-4069.
1559 Bond-Lamberty, B., Thomson, A., 2010. Temperature-associated increases in the global soil
1560 respiration record. Nature, 464, 579-582. DOI: 10.1038/nature08930.
1561 Borgå, K., Saloranta, T.M., Ruus, A., 2010. Simulating climate change induced alterations in
1562 bioaccumulation of organic contaminants in an Arctic marine food web. Environmental
1563 Toxicology and Chemistry, 29, 1349-1357.
1564 Boxall, A.B.A., Hardy, A., Beulke, S., Boucard, T., Burgin, L., Falloon, P.D., Haygarth, P.M.,
1565 Hutchinson, T., Kovats, R.S., Leonardi, G., Levy, L.S., Nichols, G., Parsons, S.A., Potts, L.,
1566 Stone, D., Topp, E., Turley, D.B., Walsh, K., Wellington, E.M.H., Williams, R.J., 2008.
1567 Impacts of climate change on indirect human exposure to pathogens and chemicals from
1568 agriculture. Environmental Health Perspective, 117, 508-514.
1569 Breivik, K., Sweetman, A., Pacyna, J.M., Jones, K.C., 2007. Towards a global historical emission
1570 inventory for selected PCB congeners—A mass balance approach-3. An update. Science of
1571 the Total Environment, 377, 296-307.

66
1572 Bustnes, J.O., Gabrielsen, G.W. Verreault, J., 2010. Climate variability and temporal trends of
1573 persistent organic pollutants in the arctic: a study of glaucous gulls. Environmental Science &
1574 Technology, 44, 3155-3161.
1575 Cabanes, A., Legagneux, L., Domine, F., 2003. Rate of evolution of the specific surface area of
1576 surface snow layers. Environmental Science & Technology, 37, 661-666.
1577 Cabrerizo, A., Dachs, J., Barcel, D., 2009. Development of a soil fugacity sampler for determination
1578 of air-soil partitioning of persistent organic pollutants under field controlled conditions.
1579 Environmental Science & Technology, 43, 8257–8263
1580 Cabrerizo, A., Dachs, J., Moeckel, C., Ojeda, M.J., Caballero, G., Barcelo, D., Jones, K.C., 2011.
1581 Factors influencing the soil–air partitioning and the strength of soils as a secondary source of
1582 polychlorinated biphenyls to the atmosphere. Environmental Science & Technology, 45,
1583 4785–4792.
1584 Cabrerizo, A., Dachs, J., Barcelo, D., Jones, K.C., 2012. Influence of organic matter content and
1585 human activities on the occurrence of organic pollutants in antarctic soils, lichens, grass, and
1586 mosses. Environmental Science & Technology, 46, 1396-1405.
1587 Callaghan, T.V., Johansson, M., Brown, R.D., Groisman, P.Y., Labba, N., Radionov, V., Bradley,
1588 R.S., Blangy, S., Bulygina, O.N., Christensen, T.R., Colman, J.E., Essery, R.L.H., Forbes,
1589 B.C., Forchhammer, M.C., Golubev, V.N., Honrath, R.E., Juday, G.P., Mescherskava, A.V.,
1590 Pheonix, G.K., Pomeroy, J., Rautio, A., Robinson, D.A., Schmidy, N.M., Serreze, M.C.,
1591 Shevchenko, V.P., Shiklomanov, A.I., Shmakin, A.B., Sköld, P., Sturm, M., Woo, M.-K.,
1592 Wood, E.F., 2011. Multiple effects of changes in Arctic snow cover. AMBIO: A Journal of
1593 the Human Environment, 40, 32-45.
1594 Cao, M.K., Woodward, F.I., 1998a. Dynamic responses of terrestrial ecosystem carbon cycling to
1595 global climate change. Nature, 393, 249-252.
1596 Cao, M.K., Woodward, F.I., 1998b. Net primary and ecosystem productions and carbon stocks of
1597 terrestrial ecosystems and their response to climate change. Global Change Biology, 4,
1598 185–198.
1599 Carmack, E., Polyakov, I., Padman, L., Fer, E., Hunke, E., Hutchings, J., Jackson, J., Kelley, D.,
1600 Kwok, R., Layton, C., Melling, H., Perovich, D., Persson, O., Ruddick, B., Timmermans,
1601 M.-L., Tolle, J., Ross, T., Vavrus, S., Winsor, P., 2015. Towards quantifying the increasing
1602 role of ocean heat in sea ice loss in the new Arctic. Bulletin of the American Meteorological
1603 Society, 96 (12), 2079-2105. DOI:10.1175/BAMS-D-13-00177.1.
1604 Carrie, J., Wang, F., Sanei, H., Macdonald, R.W., Outridge, P.M., Stern, G.A., 2010. Increasing
1605 contaminant burdens in an Arctic fish, Burbot (Lota lota), in a warming climate.
1606 Environmental Science & Technology, 44, 316-322.
1607 Carrizo, D., Gustafsson, O., 2011. Distribution and inventories of polychlorinated biphenyls in the
1608 polar mixed layer of seven pan-Arctic shelf seas and the interior basins. Environmental
1609 Science & Technology, 45, 1420–1427.
1610 Cetin, B., Ozerb, S., Sofuoglub, A., Odabasi, M., 2006. Determination of Henry's law constants of
1611 organochlorine pesticides in deionized and saline water as a function of temperature.
1612 Atmospheric Environment, 40, 4538-4546.
1613 Chen, C.R., Xu, Z.H., 2010. Forest ecosystem responses to environmental changes: the key regulatory
1614 role of biogeochemical cycling. Journal of Soils and Sediments, 10, 210–214.
1615 Christensen, J.R., MacDuffee, M., Macdonald, R.W., Whiticar, M., Ross, P.S., 2005. Pacific salmon
1616 deliver persistent organic pollutants to British Columbia grizzly bears. Environmental Science
1617 & Technology, 39, 6952-6960.

67
1618 Comiso, J.C., Hall, D.K., 2014. Climate trends in the Arctic as observed from space. WIREs Climate
1619 Change, 5, 389-409.
1620 Corsolini, S., Borghesi, N., Schiamone, A., Focardi, S., 2007. Polybrominated diphenyl ethers,
1621 polychlorinated dibenzo-dioxins, -furans, and -biphenyls in three species of Antarctic
1622 penguins. Environmental Science and Pollution Research, 14, 421-429.
1623 Cortes, D.R., Basu, I., Sweet, C.W., Brice, K.A., Hoff, R.M., Hites, R.A., 1998. Temporal trends in
1624 gas-phase concentrations of chlorinated pesticides measured at the shores of the Great Lakes.
1625 Environmental Science & Technology, 32, 1920-1927.
1626 Couillard, C.M., Courtenay, S.C., Macdonald, R.W., 2008. Chemical-environment interactions
1627 affecting the risk of impacts on aquatic organisms: A review with a Canadian perspective -
1628 interactions affecting vulnerability. Environmental Reviews, 16, 19-44.
1629 Cousins, I.T., Gevao, B., Jones, K.C., 1999. Measuring and modelling the vertical distribution of
1630 semivolatile organic compounds in soils. I: PCB and PAH soil core data. Chemosphere, 39
1631 (14), 2507-2518.
1632 Dachs, J., Eisenreich, S.J., Baker, J.E., Ko, F.C., Jeremiason, J.D., 1999. Coupling of phytoplankton
1633 uptake and air-water exchange of persistent organic pollutants. Environmental Science &
1634 Technology, 33(20), 3653-3660.
1635 Dachs, J., Lohmann, R., Ockenden, W.A., Mejanelle, L., Eisenreich, S.J., Jones, K.C., 2002. Oceanic
1636 biogeochemical controls on global dynamics of persistent organic pollutants. Environmental
1637 Science & Technology, 36 (20), 4229-4237.
1638 Dauvin, J.C., 2007. Fifty years of change. Marine Pollution Bulletin, 54, 1673-1676.
1639 Derksen, C., Brown, R., 2012. Spring snow cover extent reductions in the 2008–2012 period
1640 exceeding climate model projections. Geophysical Research Letters, 39, L19504.
1641 doi:10.1029/2012GL053387.
1642 Diez, R., Rigét, F.F., Sonne, C., Born, E.W., Bechshøft, T., McKinney, M.A., Drimmie, R.J., Muir,
1643 D.C.G., Letcher, R.J., 2013. Three decades (1983–2010) of contaminant trends in East
1644 Greenland polar bears (Ursus maritimus). Part 1: Legacy organochlorine contaminants.
1645 Environment International, 59, 485-493.
1646 Diez, R., Gustavson, K., Sonne, C., Desforges, J.P., Riget, F.F., Pavlova, V., McKinney, M.A.,
1647 Letcher, R.J., 2015. Physiologically-based pharmacokinetic modelling of immune,
1648 reproductive and carcinogenic effects from contaminant exposure in polar bears (Ursus
1649 maritimus) across the Arctic. Environmental Research, 140, 45-55.
1650 Doick, K.J., Burauel, P., Jones, K.C., Semple, K.T., 2005. Distribution of aged C-14-PCB and
1651 C-14-PAH residues in particle-size and humic fractions of an agricultural soil. Environmental
1652 Science & Technology, 39, 6575–6583.
1653 Domine, F., Cabanes, A., Legagneux, L., 2002. Structure, microphysics and surface area of the Arctic
1654 snow pack near Alert during ALERT 2000 campaign. Atmospheric Environment, 36,
1655 2753-2765.
1656 Donaldson, D.J., Anderson, D., 1999. Adsorption of atmospheric gases at the air-water interface, 2.
1657 C1eC4 alcohols, acids, and acetone. Journal of Physical Chemistry, A103, 871-876.
1658 Duarte, C. M., Lenton, T. M., Wadhams, P. Wassmann, P., 2012. Abrupt climate change in the Arctic.
1659 Nature Climate Change, 2(2), 60-64.
1660 Eckhardt, S., Breivik, K., Manø, S., Stohl, A., 2007. Record high peaks in PCB concentrations in the
1661 Arctic atmosphere due to long-range transport of biomass burning emissions. Atmospheric
1662 Chemistry and Physics,7, 4527-4536.

68
1663 Enge, E.K., Heimstad, E.S., Kallenborn, R., 1998. Distribution of polychlorinated biphenyls (PCBs) in
1664 snow samples in northern Norway. Organochlorine Compounds, 39, 435-439.
1665 Falkowski, P., Scholes, R.J., Boyle, E., Canadell, J., Canfield, D., Elser, J., Gruber, N., Hibbard, K.,
1666 Hogberg, P., Linder, S., Mackenzie, F.T., Moore III, B., Pedersen, T., Rosenthal, Y.,
1667 Seitzinger, S., Smetacker, V., Steffen, W., 2000. The global carbon cycles: a test of our
1668 knowledge of earth as a system. Science, 290, 291–296.
1669 Finizio, A., Villa, S., Raffaele, F., Vighi, M., 2006. Variation of POP concentrations in fresh-fallen
1670 snow and air on an Alpine glacier (Monte Rosa). Ecotoxicology and Environmental Safety,
1671 63, 25-32.
1672 Flanner, M.G., Zender, C.S., Randerson, J.T., Rasch, P., 2007. Present-day climate forcing and
1673 response from black carbon in snow. Journal of Geophysical Research, 112, 345-349.
1674 Foster, K.L., Stern, G.A., Pazerniuk, M.A., Hickie, B., Walkusz, W., Wang, F., Macdonald, R.W.,
1675 2012. Mercury biomagnification in marine zooplankton food webs in Hudson Bay.
1676 Environmental Science & Technology, 46, 12952-12959.
1677 French, T.D., Campbell, L.M., Jackson, D.C., Casselman, J.M., Scheider, W.A., Hayton, A., 2006.
1678 Long-term changes in legacy trace organic contaminants and mercury in Lake Ontario salmon
1679 in relation to source controls, trophodynamics, and climatic variability. Limnology and
1680 Oceanography, 51, 2794-2807.
1681 Friedman, C.L., Zhang, Y.X., Selin, N.E., 2013. Climate change and emissions impacts on
1682 atmospheric PAH transport to the Arctic. Environmental Science & Technology, 48(1),
1683 429–437.
1684 Friedman, C.L., Pierce, J.R., Selin, N.E., 2014. Assessing the influence of secondary organic versus
1685 primary carbonaceous aerosols on long-range atmospheric polycyclic aromatic hydrocarbon
1686 transport. Environmental Science & Technology, 48(6), 3293–3302.
1687 Friedman, C.L., Selin, N.E., 2016. PCBs in the Arctic atmosphere: determining important driving
1688 forces using a global atmospheric transport model. Atmospheric Chemistry and Physics, 16,
1689 3433-3448.
1690 Fuglei, E., Bustnes, J.O., Hop, H., Mork, T., Bjornfoth, H., van Bavel, B., 2007. Environmental
1691 contaminants in arctic foxes (Alopex lagopus) in Svalbard: relationships with feeding ecology
1692 and body condition. Environmental Pollution, 146, 128–138.
1693 Gaden, A., Ferguson, S.H., Harwood, L., Melling, H., Stern, G.A., 2009. Mercury trends in ringed
1694 seals (Phoca hispida) from the western Canadian Arctic since 1973: Associations with length
1695 of ice-free season. Environmental Science & Technology, 43, 3646-3651.
1696 Galbán-Malagón, C., Berrojalbiz, N., Ojeda, M.J., Dachs, J., 2012. The oceanic biological pump
1697 modulates the atmospheric transport of persistent organic pollutants to the Arctic. Nature
1698 Communications, 3(3), 199-202. DOI: 10.1038/ncomms1858.
1699 Gao, H., Ma, J., Cao, Z., Dove, A., Zhang, L., 2010. Trend and climate signals in seasonal air
1700 concentration of organochlorine pesticides over the Great Lakes. Journal of Geophysical
1701 Research, 115, D15307. DOI: 10.1029/2009JD013627.
1702 Garrett, T.J., Zhao, C., 2006. Increased Arctic cloud longwave emissivity associated with pollution
1703 from mid-latitudes. Nature, 440 (7085),787-789.
1704 Geisz, H.N., Dickhut, R.M., Cochran, M.A., Fraser, W.R., Ducklow, H.W., 2008. Melting glaciers: A
1705 probable source of DDT to the Antarctic marine ecosystem. Environmental Science &
1706 Technology, 42, 3958-3962.
1707 Gilbert, D., 2011. Buried by bad decisions. Nature, 474(7351), 275-277.

69
1708 Gioia, R., Lohmann, R., Dachs, J., Temme, C., Lakaschus, S., Schulz-Bull, D., Hand, I., Jones, K.C.,
1709 2008. Polychlorinated biphenyls in air and water of the North Atlantic and Arctic Ocean.
1710 Journal of Geophysical Research,113, D19302. DOI: 10.1029/2007JD009750.
1711 Gobas, F.A.P.C., Graggen, M.N., Zhang, X., 1995. Time response of the Lake Ontario ecosystem to
1712 virtual elimination of PCBs. Environmental Science & Technology, 29, 2038-2046.
1713 Gobas, F.A.P.C., MacLean, L.G., 2003. Sediment-water distribution of organic contaminants in
1714 aquatic ecosystems: The role of organic carbon mineralization. Environmental Science &
1715 Technology, 37, 735-741.
1716 Goldberg, E.D., Bourne, W.R.P., Boucher, E.A., Preston, E., 1975. Synthetic organohalides in the sea.
1717 Proceedings of the Royal Society of London Series, 189, 277–289.
1718 Goss, K.U., 1997. Conceptual model for the adsorption of organic compounds from the gas phase to
1719 liquid and solid surfaces. Environmental Science & Technology, 31, 3600-3605.
1720 Gouin, T., Armitage, J.M., Cousins, I.T., Muir, D.C.G., Ng, C.A., Reid, L., Tao, S., 2013. Influence of
1721 global climate change on chemical fate and bioaccumulation: The role of multimedia models.
1722 Environmental Toxicology and Chemistry, 32, 20-31.
1723 Grannas, A.M., Bogdal, C., Hageman, K.J., Halsall, C., Harner, T., Hung, H., Kallenborn, R., Klan,
1724 P., Klanova, J., Macdonald, R.W., Meyer, T., Wania, F., 2013. The role of the global
1725 cryosphere in the fate of organic contaminants. Atmospheric Chemistry and Physics, 13,
1726 3271–3305.
1727 Grynkiewicz, M., Polkowska, Z., Gorecki, T., Namiesnik, J., 2001. Pesticides in precipitation in the
1728 Gdansk region (Poland). Chemosphere, 43, 303-312.
1729 Grynkiewicz, M., Polkowska, Z., Namiesnik, J., 2002. Determination of polycyclic aromatic
1730 hydrocarbons in bulk precipitation and runoff waters in an urban region (Poland).
1731 Chemosphere, 36, 361-369.
1732 Hanot, L., Donime, F., 1999. Evolution of the surface area of a snow layer. Environmental Science &
1733 Technology, 33, 4250-4255.
1734 Hansen, K.H, Christensen, J.H, Geels, C., Silver, J.D., Brandt, J., 2015. Modelling the impact of
1735 climate change on the atmospheric transport and the fate of persistent organic pollutants in the
1736 Arctic. Atmospheric Chemistry and Physics, 15, 6549-6559.
1737 Heintzenberg, J., 1989. Arctic haze: air pollution in polar regions, AMBIO: A Journal of the Human
1738 Environment, 18, 50-55.
1739 Henson, S.A., Sarmiento, J.L., Dunne, J.P., Bopp, L., Lima, I., Doney, S.C., John, J., Beaulieu, C.,
1740 2010. Detection of anthropogenic climate change in satellite records of ocean chlorophyll and
1741 productivity. Biogeosciences, 7, 621-640.
1742 Herbert, B.M.J., Villa, S., Halsall C.J., 2006. Chemical interactions with snow: understanding the
1743 behaviour of semi-volatile organic compounds in snow. Ecotoxicology and Environmental
1744 Safety, 63, 3-16.
1745 Hoekstra, P.F., O'Hara, T.M., Pallant, S., Solomon, K.R., Muir, D.C.G., 2002. Bioaccumulation of
1746 organochlorine contaminants in bowhead whales (Balaena mysticetus) from Barrow, Alaska.
1747 Archives of Environmental Contamination and Toxicology, 42, 497-507.
1748 Hoerling, M.P., Hurrell, J.W., Xu, T., 2001. Tropical origins for recent North Atlantic climate change.
1749 Science, 292, 90–92.
1750 Hoff, R.M., Strachan, W.M.J., Sweet, C.W., Chan, C.H., Shackleton, M., Bidleman, T.F., Brice, K.A.,
1751 Burniston, D.A., Cussion, S., Gatz, D.F., Harlin, K., Schroeder, W.H., 1996. Atmospheric

70
1752 deposition of toxic chemicals to the Great Lakes: A review of data through 1994.
1753 Atmospheric Environment, 30, 3505-3527.
1754 Holland, M.M., Bitz, C.M., Tremblay, B., 2006. Future abrupt reductions in the summer Arctic sea
1755 ice. Geophysical Research Letters, 2006, 33(23), 265-288.
1756 Honjo, S., Krishfield, R.A., Eglinton, T.I., Manganini, S.J., Kemp, J.N., Doherty, K., Hwang, J.,
1757 McKee, T.K., Takizawa, T., 2010. Biological pump processes in the cryopelagic and
1758 hemipelagic Arctic Ocean: Canada Basin and Chukchi Rise. Progress in Oceanography, 85,
1759 137-170.
1760 Houde, M., De Silva, A.O., Muir, D.C.G., Letche, R.J., 2011. Monitoring of perfluorinated
1761 compounds in aquatic biota: An updated review. Environmental Science & Technology, 45,
1762 7962-7973.
1763 Hung, H., Blanchard, P., Halsall, C.J., Bidleman, T.F., Stern, G.A., Fellin, P., Muir, D.C.G., Barrie,
1764 L.A., Jantunen, L.M., Helm, P.A., Ma, J., Konoplev, A., 2005. Temporal and spatial
1765 variabilities of atmospheric POPs in the Canadian Arctic: Results from a decade of
1766 monitoring. Science of the Total Environment, 342, 119-144.
1767 Hung, H., Kallenborn, R., Breivik, K., Su, Y., Brorström-Lundén, E., Olafsdottir, K., Thorlacius, J.M.,
1768 Leppänen, S., Bossi, R., Skov, H., Manø, S., Patton, G.W., Stern, G., Sverko, E., Fellin, P.,
1769 2010. Atmospheric monitoring of organic pollutants in the Arctic under the Arctic Monitoring
1770 and Assessment Programme (AMAP): 1993–2006. Science of the Total Environment, 408,
1771 2854-2873.
1772 Hung, H., Katsoyiannis, A. A., Brorström-Lundén, E., Olafsdottir, K., Aas, W., Breivik, K.,
1773 Bohlin-Nizzetto, P., Sigurdsson, A., Hakola, H., Bossi, R., Skov, H., Sverko, E., Barresi, E.,
1774 Fellin, P., Wilson, S. (2016) Temporal Trends of Persistent Organic Pollutants (POPs) in
1775 Arctic Air: 20 Years of Monitoring under the Arctic Monitoring and Assessment Programme
1776 (AMAP). Environmental Pollution, Article in Press.
1777 Hwang, J., Kim, M., Manganini, S.J., McIntyre, C.P., Haghipour, N., Park, J., Krishfield, R.A.,
1778 Macdonald, R.W., McLaughlin, F.A., Eglinton, T.I., 2015. Temporal and spatial variability of
1779 particle transport in the deep Arctic Canada Basin. Journal of Geophysical Research-Oceans,
1780 120, 2784-2799.
1781 IPCC, 2005. Carbon dioxide capture and storage. Bert Metz, Ogunlade Davidson, Heleen de Coninck,
1782 Manuela Loos and Leo Meyer (Eds.) Cambridge University Press, UK. pp. 431
1783 IPCC, 2013. Summary for Policy makers, in: Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M.,
1784 Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P.M. (Eds.), Climate Change
1785 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment
1786 Report of the Intergovernmental Panel on Climate Change. Cambridge University Press,
1787 United Kingdom, pp. 28
1788 Jantunen, L., and T. Bidleman, T.,. 1995. Reversal of the air-water gas exchange direction of
1789 hexachlorocyclohexanes in the Bering and Chukchi seas: 1993 versus 1988. Environmental
1790 Science and Technology, 29, 1081-1089.Jantunen, L., Bidleman, T. F., 2006. Henry’s law
1791 constants for hexachlorobenzene, p,p’-DDE and components of technical chlordane and
1792 estimates of gas exchange for Lake Ontario. Chemosphere, 62, 1689–1696.
1793 Jantunen, L.M., Helm, P.A., Kylin, H., Bidleman, T.F., 2008. Hexachlorocyclohexanes (HCHs) in the
1794 Canadian Archipelago. 2. Air-water gas exchange of a- and γ-HCH. Environmental Science
1795 & Technology, 42(2), 465-470.
1796 Jenssen, B.M., 2006. Endocrine-disrupting chemicals and climate change: A worst-Case combination
1797 for Arctic marine mammals and seabirds? Environmental Health Perspectives, 14, 76-80.

71
1798 Jeremiason, J.D., Eisenreich, S.J., Baker, J.E., Eadie, B.J., 1998. PCB decline in settling particles and
1799 benthic recycling of PCBs and PAHs in Lake Superior. Environmental Science &
1800 Technology, 32, 3249-3256.
1801 Kallenborn, R., Macdonald, R., Halsall, C., Pawlak, J., Bidleman, T, 2011. Contaminant pathways and
1802 change in the cryosphere. Chapter 11.3. In Snow, water, ice and permafrost in the Arctic
1803 (SWIPA): Climate Change and the Cryosphere. Arctic Monitoring and Assessment
1804 Programme (AMAP), Oslo, Norway xii + 538 pp. Arctic Monitoring and Assessment
1805 Programme, Oslo, Norway. pp. 11-19 to 11-27.
1806 Kallenborn, R., Halsall, C., Dellong, M., Carlsson, P., 2012. The influence of climate change on the
1807 global distribution and fate processes of anthropogenic persistent organic pollutants. Journal
1808 of Environment Monitoring, 4, 2854-2869.
1809 Karhu, K., Auffret, M.D., Dungait, J.A.J., Hopkins, D.W., Prosser, J.I., Singh, B.K., Subke, J.,
1810 Wookey, P.A., Agren, G.I., Sebastia, M.T., Gouriveau, F., Bergkvist, K., Meir, P.,
1811 Nottingham, A.T., Salinas, N., Hartley, I.P., 2014. Temperature sensitivity of soil respiration
1812 rates enhanced by microbial community response. Nature, 513 (7516): 81-84. DOI:
1813 10.1038/nature13604
1814 Kelly, B.C., 2006. Bioaccumulation potential of organic contaminants in an arctic marine food web.
1815 PhD thesis, Simon Fraser University, available at http://summit.sfu.ca /item/2326.
1816 Komprda, J., Komprdová, K., Sáňka, M.,Možný, M., Nizzetto, L., 2013. Influence of climate and land
1817 use change on spatially resolved volatilization of persistent organic pollutants (POPs) from
1818 background soils. Environmental Science & Technology, 47, 7052–7059
1819 Kong, D., MacLeod, M., Hung, H., Cousins, I. T., 2014. Statistical analysis of long-term monitoring
1820 data for persistent organic pollutants in the atmosphere at 20 monitoring stations broadly
1821 indicates declining concentrations. Environmental Science & Technology, 48 (21),
1822 12492-12499.
1823 Lamon, L., von Waldow, H., MacLeod, M., Scheringer, M., Marcomini, A., Hungerbuhler, K., 2009.
1824 Modeling the global levels and distribution of polychlorinated biphenyls in air under a
1825 climate change scenario. Environmental Science & Technology, 43, 5818-5824.
1826 Lavoie, D., Denman, K.L. Macdonald, R.W., 2010. Effects of future climate change on primary
1827 productivity and export fluxes in the Beaufort Sea. Journal of Geophysical Research, 115,
1828 C04018.
1829 Le Treut, H., Somerville, R., Cubasch, U., Ding, Y., Mauritzen, C., Mokssit, A., Peterson, T., Prather,
1830 M. et al., Historical Overview of Climate Change. In: Climate Change 2007: The Physical
1831 Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the
1832 Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M.
1833 Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. Cambridge University Press,
1834 Cambridge, United Kingdom and New York, NY, USA.
1835 Lei, Y.D., Wania, F., 2004. Is rain or snow a more efficient scavenger of organic chemicals?
1836 Atmospheric Environment, 38, 3557−3571.
1837 Lenton, T.M., 2011. Early warning of climate tipping points. Nature Climate Change, 1(4), 201–209.
1838 Letcher, R.J., Bustnes, J.O., Dietz, R., Jenssen, B.M., Jørgensen, E.H., Sonne, C., Verreault, J.,
1839 Vijayan, M.M., Gabrielsen, G.W., 2010. Exposure and effects assessment of persistent
1840 organohalogen contaminants in arctic wildlife and fish. Science of the Total Environment,
1841 408, 2995–3043.
1842 Li, Y.F., Macdonald, R.W., Jantunen, L.M.M., Harner, T., Bidleman, T.F., Strachan, W.M.J., 2002.
1843 The transport of beta-hexachlorocyclohexane to the western Arctic Ocean: a contrast to
1844 alpha-HCH. Science of the Total Environment, 291 (1-3), 229-246.

72
1845 Li, Y.F., Macdonald, R.W., Ma, J., Hung, H., Venkatesh, S., 2004. Historical a-HCH budget in the
1846 Arctic Ocean: The Arctic mass balance box model (AMBBM). Science of the Total
1847 Environment, 324, 115-139.
1848 Li, Y.F., Macdonald, R.W., 2005. Sources of selected organochlorine pesticides in the Arctic and the
1849 effect of pathway divergence on HCH trends in biota: A review. Science of the Total
1850 Environment, 342, 87-106.
1851 Li, N., Wania, F., Lei, Y. D., Daly, G. L., 2003. A comprehensive and critical compilation, evaluation,
1852 and selection of physical–chemical property data for selected polychlorinated biphenyls.
1853 Journal of Physical and Chemical Reference Data, 32, 1545-1590.
1854 Livina, V.N., Lenton, T.M., 2013. A recent tipping point in the Arctic sea ice cover: abrupt and
1855 persistence increase in the seasonal cycle since 2007. Cryosphere, 7, 275-286.
1856 Lohmann, R., Breivik, K., Dachs, J., Muir, D.C.G., 2007. Global fate of POPs: Current and future
1857 research directions. Environmental Pollution, 150, 150-165.
1858 Loseto, L., Stern, G.A., Macdonald, R.W., 2015. Distant drivers or local signals: where do mercury
1859 trends in western Arctic belugas originate? Science of the Total Environment, 509-510,
1860 226-236.
1861 Lowry, K.E., van Dijken, G.L., Arrigo, K.R., 2014. Evidence of under-ice phytoplankton blooms in
1862 the Chukchi Sea from 1998 to 2012. Deep-Sea Research Part II-Topical Studies in
1863 Oceanography, 105, 105-117.
1864 Luchin, V., Paneled, G., 2014. Thermal regimes in the Chukchi sea from 1941–2008. Deep-Sea
1865 Research II, 109, 14-26.
1866 Lucia, M., Verboven, N., Strom, H., Miljeteig, C., Gavrilo, M.V., Braune, B.M., Boertmann, D.,
1867 Gabrielsen, G.W., 2015. Circumpolar contamination in eggs of the High-Arctic ivory gull
1868 Pagophila ebunea. Environmental Toxicology and Chemistry, 34, 1552-1561.
1869 Ma, J., Venkatesh, S., Jantunen, L. M. M. 2003. Evidence of the impact of ENSO events on temporal
1870 trends of Hexachlorobenzene air concentrations over the Great Lakes, Science of the Total
1871 Environment, 313(1-3), 177-184.
1872 Ma, J., Hung, H., Blanchard, P., 2004a. How do climate fluctuations affect persistent organic pollutant
1873 distribution in North America? Evidence from a decade of air monitoring. Environment Science
1874 & Technology, 38, 2538-2543.
1875 Ma, J., Cao, Z., Hung, H., 2004b. North Atlantic Oscillation signatures in the atmospheric
1876 concentrations of persistent organic pollutants-An analysis using Integrated Atmospheric
1877 Deposition Network-Great Lakes data. Journal of Geophysical Research-Atmosphere, 109,
1878 D12305, doi: 10.1029/2003JD004435.
1879 Ma, J., Venkatesh, S., Li, Y., Cao, Z., Daggupaty, S.M., 2005. Tracking toxaphene in the North
1880 American Great Lakes basin – 2. A strong episodic long-range transport event. Environmental
1881 Science & Technology, 39, 8123-8131.
1882 Ma, J., Li, Y., 2006. Interannual variation of persistent organic pollutants over the Great Lakes
1883 induced by tropical Pacific sea surface temperature anomalies. Journal of Geophysical
1884 Research, 111, D04302, doi:10.1029/2005JD006014.
1885 Ma, J., 2010. Atmospheric transport of persistent semi-volatile organic chemicals to the Arctic and
1886 cold condensation in the mid-troposphere–Part 1: 2-D modeling in mean atmosphere.
1887 Atmospheric Chemistry and Physics, 10, 7303-7314.
1888 Ma, J., Cao, Z., 2010. Quantifying the perturbations of persistent organic pollutants induced by
1889 climate change. Environmental Science & Technology, 44, 8567–8573.

73
1890 Ma, J., Hung, H., Tian, C., Kallenborn, R., 2011. Revolatilization of persistent organic pollutants in
1891 the Arctic induced by climate change. Nature Climate Change, 1, 256–260.
1892 Ma, J., Sverko, E., Su., Yu., Zhang, J., Gao, H., 2013. Uptake and mobilization of organic chemicals
1893 with clouds: Evidence from a hail sample. Environmental Science & Technology, 47,
1894 9715−9721.
1895 Macdonald, R.W., Bewers, J.M., 1996. Contaminants in the arctic marine environment: priorities for
1896 protection. ICES Journal of Marine Science, 53, 537-563.
1897 Macdonald, R.W., Mackay, D., Hickie, B., 2002. Contaminant amplification in the environment:
1898 Revealing the fundamental mechanisms. Environmental Science & Technology, 36,
1899 457A-462A.
1900 Macdonald, R.W., Mackay, D., Li, Y.F., Hickie, B., 2003. How will global climate change affect risks
1901 from long-range transport of persistent organic pollutants? Human and Ecological Risk
1902 Assessment, 9(3), 643-660.
1903 Macdonald, R.W., Harner T., Fyfe, J., 2005. Recent climate change in the Arctic and its impact on
1904 contaminant pathways and interpretation of temporal trend data. Science of the Total
1905 Environment, 342, 5-86.
1906 Macdonald, R.W., 2007. Contaminants, climate change and cold regions, in: Orbaek, J.B., Tombre, I.,
1907 Kallenborn, R., Heggseth, E.N., Falk-Petersen, S., Hoel, A.H. (Eds.), Arctic-Alpine
1908 Ecosystems and People in a Changing Environment. Springer-Verlag, Berlin, pp. 305-327.
1909 Macdonald, R.W., Kuzyk, Z.A., Johannessen, S.C., 2015. It is not just about the ice: A geochemical
1910 perspective on the changing Arctic Ocean. Journal of Environmental Studies and Sciences, 5,
1911 288-301.
1912 Mackay, D., Sang, S., Vlahos, P., Diamond, D., Gobas, F., Dolan, D., 1994. A rate constant model of
1913 chemical dynamics in a lake ecosystem: PCBs in Lake Ontario. Journal of Great Lakes
1914 Research, 20, 625–642.
1915 Mackay, D., Wania, F., 1995. Transport of contaminants to the Arctic: partitioning, processes and
1916 models. Science of the Total Environment, 160-161, 25-38.
1917 Mackay, D., 2001. Multimedia Environmental Models: the fugacity approach, 2nd Edition, Lewis
1918 Publishers, New York, pp. 261
1919 MacLeod, M., Riley, W.J., Mckone, T.E., 2005. Assessing the influence of climate variability on
1920 atmospheric concentrations of polychlorinated biphenyls using a Global-Scale Mass Balance
1921 Model (BETR-Global). Environmental Science & Technology, 39, 6749-6756.
1922 Manciocco, A., Calamandrei, G., Alleva, E., 2014. Global warming and environmental contaminants
1923 in aquatic organisms: The need of the etho-toxicology approach. Chemosphere, 100, 1-7.
1924 Mao, X., Gao, H., Huang, T., Zhang, L., Ma, J., 2014. Modeling global persistent organic chemicals in
1925 clouds. Atmospheric Environment, 96, 345-352.
1926 Masson, O., Baeza, A., Bieringer, J., Brudecki, K., Bucci, S., Cappai, M., Carvalho, F.P., Connan, O.,
1927 Cosma, C., Dalheimer, A., Didier, D., Depuydt, G., De Geer, L.E., De Vismes, A., Gini, L.,
1928 Groppi, F., Gudnason, K., Gurriaran, R., Hainz, D., Halldorsson, O., Hammond, D., Hanley,
1929 O., Holey, K., Homoki, Z., Ioannidou, A., Isajenko, K., Jankovic, M., Katzlberger, C.,
1930 Kettunen, M., Kierepko, R., Kontro, R., Kwawman, P.J.M., Lecomte, M., Leon Vintro,L., .
1931 Leppanen, A.P., Lind, B., Lujaniene, G., Mc Ginnity, R., Mc Mahon, C., Mala, H., Manenti,
1932 S., Manolopoulou, M., Mattila, A., Mauringm, A., Papastefanou, C., Penev, I., Pham, M.K.,
1933 Povinec, P.P., Rameback, H., Reis, M.C., Ringer, W., Rodriguez, A., Rulík, P., Saey, P.R.J.,
1934 Samsonov, V., Schlosser, C., Sgorbati, G., Silobritiene, B.V., Soderstrom, C., Sogni, R.,
1935 Solier, L., Sonck, M., Steinhauser, G., Steinkopf, T., Steinmann, P., Stoulos, S., Sykora, I.,

74
1936 Todorovic, D., Tooloutalaie, N., Tositti, L., Tschiersch, J., Ugron, A., Vagena, E., Vargas, A.,
1937 Wershofen, H., Zhukova, O., 2011. Tracking of airborne radionuclides from the damaged
1938 Fukushima Dai-Ichi Nuclear Reactors by European Networks. Environmental Science &
1939 Technology, 45, 7670-7677.
1940 McGuire, A.D., Anderson, L.G., Christensen, T.R., Dallimore, S., Guo, L., Hayes, D.J., Heimann, M.,
1941 Lorenson, T.D., Macdonald, R.W., Roulet, N., 2009. Sensitivity of the carbon cycle in the
1942 Arctic to climate change. Ecological Monographs, 79(4), 523-555.
1943 McKinney, M.A., Peacock, E., Letcher, R.J., 2009. Sea ice-associated diet change increases the levels
1944 of contaminants in polar bears. Environmental Science & Technology, 43, 4334-4339.
1945 McKinney, M.A., McMeans, B.C., Tomy, G.T., Rosenberg, B., Ferguson, S.H., Morris, A., Muir,
1946 D.C.G., Fisk, A.T., 2012. Trophic transfer of contaminants in a changing arctic marine food
1947 web: Cumberland Sound, Nunavut, Canada. Environmental Science & Technology, 46,
1948 9914-9922.
1949 McKinney, M.A., Iverson, S.J., Fisk, A.T., Sonne, C., Rigét, F.F., Letcher, R.J., Arts, M.T., Born,
1950 E.W., Rosing-Asvid, A., Dietz, R., 2013. Global change effects on the long-term feeding
1951 ecology and contaminant exposures of East Greenland polar bears. Global Change Biology,
1952 19 (8), 2360-2372. DOI: 10.1111/gcb.12241.
1953 McKinney, M.A., Pedro, S., Dietz, R., Sonne, C., Fisk, A.T., Roy, D., Jenssen, B.M., Letcher, R.J.,
1954 2015. A review of ecological impacts of global climate change on persistent organic pollutant
1955 and mercury pathways and exposures in arctic marine ecosystems. Current Zoology, 61,
1956 617-628.
1957 McKone, T.E., Daniels, J.I., Goldman, M., 1996. Uncertainties in the link between global climate
1958 change and predicted health risks from pollution: Hexachlorobenzene (HCB) case study using
1959 a fugacity model. Risk Analysis, 16(3): 377-393.
1960 Meijer, S.N., Ockenden, W.A., Sweetnam, A., Breivik, K., Grimalt, J.O., Jones, K.C., 2003. Global
1961 distribution and budget of PCBs and HCB in background surface soils: implications for
1962 sources and environmental processes. Environmental Science & Technology, 37, 667–672.
1963 Meijer, S., Grimalt, J., Fernandez, P., Dachs, J., 2009. Seasonal fluxes and temperature-dependent
1964 accumulation of persistent organic pollutants in lakes: The role of internal biogeochemical
1965 cycling. Environmental Pollution, 157, 1815–1822.
1966 Meyer, T., Wania, F., 2008. Organic contaminant amplification during snowmelt. Water Research, 42,
1967 1847-1865.
1968 Moeckel, C., Nizzetto, L., Guardo, A.D., Steinnes, E., Freppaz, M., Filippa, G., Camporini, Paolo.,
1969 Benner, J., Jones, K.C., 2008. Persistent organic pollutants in boreal and montane soil
1970 profiles: distribution, evidence of processes, and implications for global cycling.
1971 Environmental Science & Technology, 42(22), 8374-8380
1972 Moeckel, C., Nizzetto, L., Berg, B., Lindroth, A., Jones, K.C., 2009. Air-boreal forest transfer and
1973 processing of polychlorinated biphenyls. Environmental Science & Technology, 43(14),
1974 5282–5289
1975 Muir, D.C.G., Norstrom, R.J., 1994. Persistent organic contaminants in Arctic marine and freshwater
1976 ecosystems. Arctic Research of the United States, 8, 136-146.
1977 Muir, D., Riget, F., Cleemann, M., Skaare, J., Kleivane, L., Nakata, H., Dietz, R., Severinsen, T.,
1978 Tanabe, S., 2000. Circumpolar trends of PCB and organochlorine pesticides in the arctic
1979 marine environment inferred from levels in ringed seals. Environmental Science &
1980 Technology, 34(12), 2431–2438.

75
1981 Müller-Herold, U., Caderas, U., Funck, P., 1997. Validity of global lifetime estimates by a simple
1982 general limiting law for the decay of organic compounds with long-range pollution potential.
1983 Environmental Science & Technology, 31, 3511-3515.
1984 Bengtson Nash, S., 2011. Persistent organic pollutants in Antarctica: current and future research
1985 priorities. Journal of Environmental Monitoring, 13, 497-504.
1986 NCP, 2013. Canadian Arctic Contaminants Assessment Report On Persistent Organic Pollutants –
1987 2013.Muir D, Kurt-Karakus P, Stow J. (Eds). Northern Contaminants Program, Aboriginal
1988 Affairs and Northern Development Canada, Ottawa ON. xxiii + 487 pp + Annex.
1989 Ng, C.A., Gray, K.A., 2011. Forecasting the effects of global change scenarios on bioaccumulation
1990 patterns in great lakes species. Global Change Biology, 17, 720-733
1991 Nizzetto, L., Jarvis, A., Brivio, P.A., Jones, K.C., Di Guardo, A., 2008. Seasonality of air-forest
1992 canopy POP exchange. Environmental Science & Technology, 42, 8778-8783.
1993 Nizzetto, L., MacLeod, M., Borga, K., Cabrerizo, A., Dachs, J., Guardo, A.D., Ghirardello, D.,
1994 Hansen, K.M., Jarvis, A., Lindroth, A., Ludwig, B., Monteith, D., Perlinger, J.A., Scheringer,
1995 M., Schwendenmann, L., Semple, K.t., Wick, L.Y., Zhang, G., Jones, K.C., 2010. Past,
1996 present, and future controls on levels of persistent organic pollutants in the global
1997 environment. Environmental Science & Technology, 44, 6526–6531.
1998 Nizzetto, L., Gioia, R., Li, J., Borgå, K., Pomati, F., Bettinetti, R., Dachs, J., Jones, K.C., 2012.
1999 Biological pump control of the fate and distribution of hydrophobic organic pollutants in
2000 water and plankton. Environmental Science & Technology, 46, 3204−3211.
2001 Noyes, P.D., McElwee, M.K., Miller, H.D., Clark, B.W., Van Tiem, L.A., Walcott, K.C., Erwin, K.N.,
2002 Levin, E.D., 2009. The toxicology of climate change: environmental contaminants in a
2003 warming world. Environment International, 35, 971-986.
2004 Noyes, P.D., Lima, S.C., 2015. Forecasting the impacts of chemical pollution and climate change
2005 interactions on the health of wildlife. Current Zoology, 61. 669-689.
2006 O’Driscoll, K., Mayer, B., Su, J., Mathis, M., 2014. The effects of climate change on persistent
2007 organic pollutants (POPs) in the North Sea. Ocean Science, 10, 397-409.
2008 Octaviani, M., Stemmler, I., Lammel, G., Graf, F., 2015. Atmospheric transport of persistent organic
2009 pollutants to and from the Arctic under present-day and future climate. Environmental
2010 Science & Technology, 49(6), 3593-3602.
2011 Ottar, B., 1981. The transfer of airborne pollutants to the Arctic region. Atmospheric Environment, 15,
2012 1439-1445.
2013 Overpeck, J.T., Sturm, M., Francis, J.A., Perovich, D.K., Serreze, M.C., Benner, R., Carmack, E.C.,
2014 Chapiun III, F.S., Gerlach, S.C., Hamilton, L.C., Hinzman, L.D., Holland, M., Huntington,
2015 H.P., Key, J.R., Lloyd, A.H., Macdonald, G.M., McFadden, J., Noone, D., Prowse, T.D.,
2016 Schlosser, P., Vörösmarty, C., 2005. Arctic system on trajectory to new, seasonally ice-free
2017 state. EOS, Transactions, American Geophysical Union, 86, 309-316.
2018 Pachauri, R., Reisinger, A., IPCC Climate Change 2007: Synthesis Report. Cambridge Univ. Press,
2019 2007. Available at http://www.ipcc.ch/publications_ and_data/ publications_and_
2020 data_reports.shtml.
2021 Pacyna, J.M., Cousins, I.T., Halsall, C., Rautio, A., Pawlak, J., Pacyna, E.G., Sundseth, K., Wilson, S.,
2022 Munthe, J., 2015. Impacts on human health in the Arctic owing to climate-induced changes in
2023 contaminant cycling – The EU ArcRisk project policy outcome. Environmental Science &
2024 Policy, 50, 200-213.Palmer, M.A., Saenz, B.T., Arrigo, K.R., 2014. Impacts of sea ice retreat,
2025 thinning, and melt pond proliferation on the summer phytoplankton bloom in the Chukchi Sea,
2026 Arctic Ocean. Deep-Sea Research II, 105, 85-104.

76
2027 Pekar, M., Pavlova, N., Gusev, A., Shatalov, V., Vulikh, N., Ioannisian, D., Dutchak, S., Berg, T.,
2028 Hjellbrekke, A., 1999. Long-range transport of selected persistent organic pollutants.
2029 Development of transport models for polychlorinated biphenyls, benzo[a]pyrene,
2030 dioxins/furans and lindane, Joint report of EMEP Centres: MSC-E and CCC, EMEP Report
2031 4/99.
2032 Prentice, I.C., 2001. The carbon cycle and atmospheric carbon dioxide. In: Climate Change 2001: The
2033 Scientific Basis (IPCC). Cambridge: Cambridge University Press, pp.184-237
2034 Pućko, M., Stern, G.A., MacDonald, R., Barner, D., 2010a. α- and γ-Hexachlorocyclohexane
2035 measurements in the brine fraction of sea ice in the Canadian High Arctic using a sump-hole
2036 technique. Environmental Science & Technology, 44, 9258–9264.
2037 Pućko, M., Stern, G.A., Barber, D.G., Macdonald, R.W., Rosenberg, B., 2010b. The International
2038 Polar Year (IPY) Circumpolar Flaw Lead (CFL) System Study: the importance of brine
2039 processes for α- and β-hexachlorocyclohexane (HCH) accumulation/rejection in the sea ice.
2040 Atmosphere-Ocean, 48, 244-262.
2041 Pućko, M., Stern, G.A., Macdonald, R.W., Barber, D.G., Rosenberg, B., 2011. The influence of the
2042 atmosphere-snow-ice-ocean interactions on the levels of hexachlorocyclohexanes (HCHs) in
2043 the Arctic cryosphere. Journal of Geophysical Research, 116,
2044 C02035.doi:10.1029/2010JC00661.
2045 Pućko, M., Macdonald, R.W., Barber, D.G., Rosenberg, B., Gratton, Y., 2012a. α-HCH enantiomer
2046 fraction (EF) – a novel approach to calculate the ventilation age of water in the Arctic Ocean?
2047 Journal of Geophysical Research, 117, C08038.
2048 Pućko, M., Stern, G.A., Barber, D., Macdonald, R.W., Warner, K., Fuchs, C., 2012b. Mechanisms and
2049 implications of α-HCH enrichment in melt pond water on Arctic sea ice. Environmental
2050 Science & Technology, 46, 11862-11869.
2051 Pućko, M., Stern, G.A., Macdonald, R.W., Barber, D.G., Rosenberg, B., Walkusz, W., 2013a. When
2052 will α-HCH disappear from the Arctic Ocean? Journal of Marine Systems, 127, 88-100.
2053 Pućko, M., Walkusz, W., Macdonald, R.W., Barber, D., Fuchs, C., Stern, G.A., 2013b. Importance of
2054 arctic zooplankton seasonal migrations for α-hexachlorocyclohexane (α-HCH)
2055 bioaccumulation dynamics. Environmental Science & Technology, 46, 11862-11869.
2056 Pućko, M., Stern, G.A., Macdonald, R.W., Jantunen, L.M., Bidleman, T.F., Wong, F., Barber, D.G.,
2057 Rysgaard, S., 2015. The delivery of organic contaminants to the Arctic food web: why sea ice
2058 matters. Science of the Total Environment, 506-507, 444-452.
2059 Rahn, K.A., Heidam, N.Z., 1981. Progress in Arctic air chemistry 1977–1980: a comparison of the
2060 first and second symposium. Atmospheric Environment, 15, 1345-1348.
2061 Rigét, F., Bignert, A., Braune, B., Stow, J., Wilson, S., 2010. Temporal trends of legacy POPs in
2062 Arctic biota, an update. Science of the Total Environment, 408, 2874-2884.
2063 Rigét, F., Vorkamp, K., Hobson, K.A., Muir, D.C.G., Dietz, R., 2013. Temporal trends of selected
2064 POPs and the potential influence of climate variability in a Greenland ringed seal population.
2065 Environmental Science: Processes & Impacts, 15, 1706-1716.
2066 Rockström, J., Steffen, W., Noone, K., Persson, A., Chapin, S., Lambin, E.F., Lenton, T.M., Scheffer,
2067 M., Folke, C., Schellnhuber, H.J., Nykvist, B., de Wit, C.A., Hughes, T., van der Leeuw, S.,
2068 Rodhe, H., Sörlin, S., Snyder, P.K., Costanza, R., Svedin, U., Falkenmark, M., Karlberg, L.,
2069 Corell, R.W., Fabry, V.J., Hansen, J., Walker, B., Liverman, D., Richardson, K., Crutzen, P.
2070 Foley, J.A., 2009a. A safe operating space for humanity. Nature, 461 (24 September):
2071 472-475.

77
2072 Rockström, J., Steffen, W., Noone, K., Persson, A., Chapin, S., Lambin, E.F., Lenton, T.M., Scheffer,
2073 M., Folke, C., Schellnhuber, H.J., Nykvist, B., de Wit, C.A., Hughes, T., van der Leeuw, S.,
2074 Rodhe, H., Sörlin, S., Snyder, P.K., Costanza, R., Svedin, U., Falkenmark, M., Karlberg, L.,
2075 Corell, R.W., Fabry, V.J., Hansen, J., Walker, B., Liverman, D., Richardson, K., Crutzen, P.,
2076 Foley, J.A., 2009b. Planetary boundaries: Exploring the safe operating space for humanity.
2077 Ecology and Society, 14, 32.
2078 Roemmich, D., Church, J., Gibson, J., Monselesan, D., Sutton, P., Wijfels, S., 2015. Unabated
2079 planetary warming and its ocean structure since 2006. Nature Climate Change, 5, 240-245.
2080 Ross, P.S., Kennedy, C., Shelley, L.K., Tierney, K.B., Patterson, D.A., Fairchild, W.L., Macdonald,
2081 R.W., 2013. The trouble with salmon: Relating pollutant exposure to toxic effect in species
2082 with transformational life histories and lengthy migrations. Canadian Journal of Fisheries and
2083 Aquatic Sciences, 70, 1252-1264.
2084 Roth, C.M., Goss, K. U., Schwarzenbach, R.P., 2002. Adsorption of a diverse set of organic vapors on
2085 the bulk water surface. Journal of Colloid Interface Science, 252, 21-30.
2086 Roth, C.M., Goss, K.U., Schwarzenbach, R.P., 2004. Sorption of diverse organic vapors to snow.
2087 Environmental Science & Technology, 38, 4078-4084.
2088 RSC (The Royal Society), 2005. Ocean acidification due to increasing atmospheric carbon dioxide,
2089 Plymouth, UK, pp. 57
2090 Schenker, U., MacLeod, M., Scheringer, M., Hungerbuhler, K., 2005. Improving data quality for
2091 environmental fate models: a least-squares adjustment procedure for harmonizing
2092 physicochemical properties of organic compounds. Environment Science & Technology, 39,
2093 8434-8441.
2094 Scheringer, M., 1997. Characterization of the environmental distribution behaviour of organic
2095 chemicals by means of persistence and spatial range. Environmental Science & Technology,
2096 31, 2891-2897.
2097 Scheringer, M., Hungerbühler, K., Matthies, M., 2001. The spatial scale of organic chemicals in
2098 multimedia fate modeling- Recent developments and significance for chemical..
2099 Environmental Science and Pollution Research, 8, 150-155
2100 Schiedek, D., Sundelin, B., Readman, J.W., Macdonald, R.W., 2007. Interactions between climate
2101 change and contaminants. Marine Pollution Bulletin, 54, 1845-1856.
2102 Schindler, D.W., Smol, J.P., 2006. Cumulative effects of climate warming and other human on
2103 freshwaters of Arctic and subarctic freshwaters of North America. AMBIO: A Journal of the
2104 Human Environment ,35(4), 160-168.
2105 Semple, K.T., Doick, K.J., Wick, L.Y., Harms, H., 2007. Microbial interactions with organic
2106 contaminants in soil: definitions, processes and measurement. Environmental Pollution,
2107 150(1), 166–176.
2108 Serreze, M.C., Walsh, J.E., Chapin, F.S.I., Osterkamp, T., Dyurgerov, M., Romanovsky, V., Oechel,
2109 W.C., Morison, J., Zhang, T., Barry, R.G., 2000. Observational evidence of recent change in
2110 the northern high-latitude environment. Climatic Change, 46, 159-207.
2111 Serreze, M.C., Stroeve, J., 2015. Arctic sea ice trends, variability and implications for seasonal ice
2112 forecasting. Philosophical Transactions of the Royal Society a-Mathematical Physical and
2113 Engineering Sciences, 373.
2114 Shen, L., Wania, F., Lei, Y.D., Teixeira, C., Muir, D.C.G., Bidleman, T.F., 2004.
2115 Hexachlorocyclohexanes in the North American atmosphere. Environmental Science &
2116 Technology, 38, 965-975.

78
2117 Shen, L., Wania, F., 2005. Compilation, evaluation, and selection of physical-chemical
2118 property data for organochlorine pesticides. Journal of Chemical Engineering Data, 50,
2119 742-768.
2120 Shoeib, M., Harner, T., 2002. Using measured octanol-air partition coefficients to explain
2121 environmental partitioning of organochlorine pesticides. Environmental Toxicology and
2122 Chemistry, 21(5), 984- 990.
2123 Simonich, S.L., Hites, R., 1994. Importance of vegetation in removing polycyclic aromatic
2124 hydrocarbons from the atmosphere. Nature, 370, 49-51
2125 Simonich, S.L., Hites, R.A., 1995. Global distribution of persistent organochlorine compounds.
2126 Science, 269, 1851–1854.
2127 Smith, J.N., Brown, R.M., Williams, W.J., Robert, M., Nelson, R., Moran, S.B., 2015. Arrival of the
2128 Fukushima radioactivity plume in North American continental waters. Proceedings of the
2129 National Academy of Sciences of the United States of America, 112, 1310-1315.
2130 Sobek, A., Gustafsson, O., 2014. Deep water masses and sediments are main compartments for PCBs
2131 in the Arctic Ocean. Environmental Science & Technology, 48, 6719-6725.
2132 Steele, M., Ermold, W., Zhang, J., 2008. Arctic Ocean surface warming trends over the past 100 years.
2133 Geophysical Research Letter, 35, L0261.
2134 Stemmler, I., Lammel, G., 2009. Cycling of DDT in the global oceans 1950-2002: World ocean
2135 returns the pollutant. Geophysical Research Letters, 36, L24602.
2136 Stern, G.A., Braekevelt, E., Helm, P.A, Bidleman, T.F., Outridge, P.M., Lockhart, W.L., McNeeley,
2137 R., Rosenberg, B., Ikonomou, M.G., Hamilton, P., Tomy, G.T. Wikinson, P., 2005. Modern
2138 and historical fluxes of halogenated organic contaminants to a lake in the Canadian Arctic as
2139 determined from annually laminated sediment cores. Science of the Total Environment,
2140 342(1-3): 223-243.
2141 Stocker, J., Scheringer, M., Wegmann, F., Hungerbuhler, K., 2007. Modeling the effect of snow and
2142 ice on the global environmental fate and long-range transport potential of semivolatile organic
2143 compounds. Environmental Science & Technology, 41, 6192–6198.
2144 Stohl, A., Seibert, P., Wotawa, G., Arnold, D., Burkhart, J.F., Eckhardt, S., Tapia, C., Vargas, A.,
2145 Yasunari, T.J., 2012. Xenon-133 and caesium-137 releases into the atmosphere from the
2146 Fukushima Dai-ichi nuclear power plant: determination of the source term, atmospheric
2147 dispersion, and deposition. Atmospheric Chemistry and Physics, 12, 2313-2343.
2148 Stroeve, J.C., Kattsov, V., Barrett, A., Serreze, M., Pavlova, T., Holland, M., Meier, W.N., 2012.
2149 Trends in Arctic sea ice extent from CMIP5, CMIP3 and observations. Geophysical Research
2150 Letters, 39, L16502.
2151 Sumaila, U.R., Cheung, W.W.L., Lam, V.W.Y., Pauly, D., Herrick, S., 2011. Climate change impacts
2152 on the biophysics and economics of world fisheries. Nature Climate Change, 1, 449-456.
2153 Teng, Y., Xu, Z., Luo, Y., 2012. How do persistent organic pollutants be coupled with biogeochemical
2154 cycles of carbon and nutrients in terrestrial ecosystems under global climate change? Journal
2155 of Soils and Sediments, 12, 411-419.
2156 Teran, T., Lamon, L., Marcomini, M., 2012. Climate change effects on POPs’ environmental
2157 behaviour: a scientific perspective for future regulatory actions. Atmospheric Pollution
2158 Research, 3(4), 466-476.
2159 Thies, H., Kickus, U., Mair, V., Tessadri, R., Tait, D., Thaler, B., Psenner, R., 2007. Unexpected
2160 response of high alpine lake waters to climate warming. Environmental Science &
2161 Technology, 41, 7424-7429.

79
2162 Timothy, D.A., Wong, C.S., Barwell-Clarke, J.E., White, L.A., Page, J.S., Macdonald, R.W., 2013.
2163 Climatology of sediment flux and composition in the subarctic Northeast Pacific Ocean with
2164 biogeochemical implications. Progress in Oceanography, 116, 95-129.
2165 UNEP/AMAP, 2011. Climate Change and POPs: Predicting the Impacts. Report of the UNEP/AMAP
2166 Expert Group. Secretariat of the Stockholm Convention, Geneva, pp. 62
2167 van Pul, W., de Leeuw, F., van Jaarsveld, J., van der Gagg, M., Sliggers, C., 1998. The potential for
2168 long-range transboundary atmospheric transport. Chemosphere, 37, 113-141.
2169 Villa, S., Negrelli, C., Finizio, A., Flora, O., Vighi, M., 2006. Organochlorine compounds in ice melt
2170 water from Italian Alpine rivers. Ecotoxicology and Environmental Safety, 63, 84-90.
2171 von Waldow, H., MacLeod, M., Jones, K., Scheringer, M., Hungerbühler, K., 2010. Remoteness from
2172 emission sources explains the fractionation pattern of polychlorinated biphenyls in the
2173 Northern Hemisphere. Environmental Science & Technology, 44, 6183-6188.
2174 Vorkamp, K., Christensen, J.H., Rigét, F., 2004. Polybrominated diphenyl ethers and organochlorine
2175 compounds in biota from the marine environment of East Greenland. Science of the Total
2176 Environment, 331, 143-155.
2177 Vorkamp, K., Rigét, F.F., Bossi, R., Dietz, R., 2011. Temporal trends of hexabromocyclododecane,
2178 polybrominated diphenyl ethers and polychlorinated biphenyls in ringed seals from East
2179 Greenland. Environmental Science & Technology, 45, 1243-1249.
2180 Wallace, J.M., Thompson, D.W.J., 2002. Annular modes and climate prediction. Physics Today, 55,
2181 28–33.
2182 Wang, X., Gong, P., Zhang, Q., Yao, T., 2010a. Impact of climate fluctuations on deposition of DDT
2183 and hexachlorocyclohexane in mountain glaciers: Evidence from ice core records.
2184 Environmental Pollution, 158(2), 375-380.
2185 Wang, F., Macdonald, R.W., Stern, G.A., Outridge, P.M., 2010b. When noise becomes the signal:
2186 Chemical contamination of aquatic ecosystems under a changing climate. Marine Pollution
2187 Bulletin, 60, 1633-1635.
2188 Wania, F., Mackay, D., 1993. Global fractionation and cold condensation of low volatility
2189 organochlorine compounds in polar regions. AMBIO: A Journal of the Human Environment,
2190 22, 10-18.
2191 Wania, F., Mackay, D., 1996. Tracking the distribution of persistent organic pollutants. Environmental
2192 Science & Technology, 30, 390-396A.
2193 Wania, F., 1997. Modelling the fate of non-polar organic chemicals in an ageing snow pack.
2194 Chemosphere, 35, 2345-2363.
2195 Wania, F., Haugen, J., Lei, Y., Mackay, 1998. Temperature dependence of atmospheric concentrations
2196 of semivolatile organic compounds. Environmental Science & Technology, 32, 1013-1021.
2197 Wania, F., Mackay, D., 1999. Global chemical fate of a-hexachlorocyclohexane. 2. Use of a global
2198 distribution model for mass balancing, source apportionment, and trend prediction.
2199 Environmental Toxicology and Chemistry, 18, 1400-1407.
2200 Wania, F., 2003. Assessing the potential of persistent organic chemicals for long-range transport and
2201 accumulation in polar regions. Environmental Science & Technology 37, 1344-1351.
2202 Wania, F., Su. Y., 2004. Quantifying the global fractionation of polychlorinated biphenyls. AMBIO,
2203 33(3), 161-168.
2204 Watson, R.T., Patz, J., Gubler, D.J., Parson, E.A., Vincent, J.H., 2005. Environmental health
2205 implications of global climate change. Journal of Environmental Monitoring, 7, 834-843.

80
2206 Weingartner, T.J., Danielson, S., Sasaki, Y., Pavlov, V., Kulakov, M., 1999. The Siberian coastal
2207 current: A wind- and buoyancy-forced Arctic coastal current. Journal of Geophysical
2208 Research, 104, 29697-29713.
2209 Wöhrnschimmel, H., MacLeod, M., Hungerbühler, K., 2013. Emissions, fate and transport of
2210 persistent organic pollutants to the Arctic in a changing global climate. Environmental
2211 Science & Technology, 47, 2323-2330.
2212 Wong, F., Jantunen, L. M., Pućko, M., Papakyriakou, T., Stern, G. A., Bidleman, T. F., 2010.
2213 Air-water exchange of anthropogenic and natural organohalogenson: International Polar Year
2214 (IPY) expeditionsin the Canadian Arctic. Environmental Science & Technology, 45, 876-881.
2215 Wood, K.R., Bond, N.A., Danielson, S.L., Overland, J.E., Salo, S.A., Stabeno, P., Whitefield, J., 2015.
2216 A decade of environmental change in the Pacific Arctic region. Progress in Oceanography,
2217 136, 12-31.
2218 Woodgate, R.A., Aagaard, K., Wiengartner, T.J., 2006. Interannual changes in the Bering Strait fluxes
2219 of volume, heat and freshwater between 1991 and 2004. Geophysical Research Letters 33,
2220 L15609.
2221 World Health Organization. Persistent Organic Pollutants: Impact on Child Health. 2010. Geneva,
2222 Switzerland, pp. 59
2223 Xiao, H., Li, N.Q., Wania, F., 2004. Compilation, evaluation, and selection of physical-chemical
2224 property data for alpha-, beta-, and gamma-hexachlorocyclohexane. Journal of Chemical
2225 Engineering Data, 49, 173-185.
2226 Xu, Z.H., Chen, C.R., 2006. Fingerprinting global climate change and forest management within
2227 rhizosphere carbon and nutrient cycling processes. Environmental Science and Pollution
2228 Research, 13, 293-298.
2229 Yao, Y., Harner, T. Ma, J., Tuduri, L., Blanchard, P., 2007. Sources and occurrence of Dacthal in the
2230 Canadian atmosphere. Environmental Science & Technology, 41(3), 688-694.
2231 Zhang, L., Ma, J., Venkatesh, S., Li, Y., Cheung, P., 2008. Modeling evidence of episodic
2232 intercontinental long-range transport of lindane. Environmental Science & Technology, 42,
2233 8791-8797.
2234 Zhang, L., Lohmann, R., 2010. Cycling of PCBs and HCB in the surface ocean-lower atmosphere of
2235 the open pacific. Environmental Science & Technology, 44(10), 3832–3838.
2236 Zhang, Y., Jacob, D.J., Dutkiewicz, S., Amos, H.M., Long, M.S., Sunderland, E.M., 2015.
2237 Biogeochemical drivers of the fate of riverine mercury discharged to the global and Arctic
2238 Oceans. Global Biogeochemical Cycles, 29, 854-864. doi:10.1002/2015GB005124.
2239 Zhao, Y., Huang, T., Gao, H., Ma, J., 2015. Step changes in persistent organic pollutants over the
2240 Arctic and their implications. Atmospheric Chemistry and Physics, 15, 3479–3495.
2241 Zheng, Q., Nizzetto, L., Mulder, M.D., Sáňka, O., Lammel, G., Li, J., Bing, H., Liu, X., Jiang, Y.,
2242 Luo, C., Zhang, G., 2014. Does an analysis of polychlorinated biphenyl (PCB) distribution in
2243 mountain soils across China reveal a latitudinal fractionation paradox? Environmental
2244 Pollution, 195, 115-122.

81