Hydrochemistry and Critical Loads of Acidity for Lakes and Ponds in the Canadian Arctic

Hydrochemistry and critical loads of acidity for lakes and ponds in the Canadian Arctic

A Thesis Submitted to the Committee of Graduate Studies in Partial Fulfillment of the Requirements for the Degree of Master of Science in the Faculty of Arts and Science Trent University Peterborough, Ontario, Canada

Abstract

Thesis: Hydrochemistry and Critical Loads of Acidity for Lakes in The Canadian Arctic

Author: Tanner Liang

Threats such as climate change and increased anthropogenic activity such as shipping, are expected to negatively affect the Arctic. Lack of data on Arctic systems restricts our current understanding of these sensitive systems and limits our ability to predict future impacts. Lakes and ponds are a major feature of the Arctic landscape and are recognized as ‘sentinels of change’, as they integrate processes at a landscape scale. A total of 1300 aquatic sites were assessed for common chemical and physical characteristics. Geology type was found to be the greatest driver of water chemistry for Arctic lakes and ponds.

Acid-sensitivity was assessed using the Steady State Water Chemistry model and a subset of

1138 sites from across the Canadian Arctic. A large portion of sites (40.0%, n = 455) were classified as highly sensitive to acidic deposition, which resulted in a median value of 35.8 meq·m―2·yr―1 for the Canadian Arctic. Under modelled sulphur deposition scenarios for the year 2010, exceedances associated with shipping is 12.5% (n = 142) and 12.0% (n = 136) for without shipping, suggesting that impacts of shipping are relatively small.

Keywords: Critical Loads, Arctic Lakes, Water Chemistry, Shipping Emissions, Acidic Deposition, Steady-State Water Chemistry Model

ii Acknowledgments

First and foremost, I want to express my great appreciation for the patience, guidance, and encouragement of Julian Aherne throughout my Masters. His unwavering positive attitude, knowledge, and humor was a great source of inspiration for me, and to whom I owe my R skills, many coffees, and many hands of bananas. I am grateful to

Julian for giving me the opportunity to study and visit the Arctic – an awe-inspiriting place to be - with many memories that I will always treasure. In addition, I must thank my supervisory committee Peter Lafleur and Céline Guéguen, for their patience and

Arctic expertise that they bring into this thesis. Also, I thank my external examiner Colin

Whitfield for their valuable contribution and participation in my thesis defence.

This research would not have been possible without the funding and logistical support from a number of sources: Environment and Climate Change Canada, the

Northern Scientific Training Program, the Natural Sciences and Engineering Research

Council of Canada, the Polar Continental Shelf Project, the Nunavut Research Institute

(especially Rick Armstrong and Jamal Shirley) and the McLean Foundation.

I would like to acknowledge the territory of the ᐃᓄᐃᑦ (Inuit) and Dene peoples

(specifically Akaitcho First Nations) that the research for this project took place in;

ᕿᑭᖅᑖᓗᒃ (Omanerdjuark; Baffin Island), Akpatordjuark (Coats Island), Prince Charles

Island, and Northwest Territories. Specifically (on ᕿᑭᖅᑖᓗᒃ) the communities surrounding ᑭᒻᒥᕈᑦ (Kimmirut), ᐃᖃᓗᐃᑦ (Iqaluit; Jean Allen), ᐸᖕᓂᖅᑑᖅ (Pangniqtuuq,

Pangnirtung; Billy Etooangat), and ᒥᑦᑎᒪᑕᓕ (Mittimatalik; Pond Inlet). I’d also like to acknowledge the works of other polar research groups, e.g., John Smol’s Paleoecological iii Environmental Assessment and Research Laboratory; Scott Lamoureux’s Cape Bounty

Arctic Watershed Observatory, producers of many limnological data in the Arctic. Their work has been on of the largest contribution to polar limnology and have set the foundation for future work in the Arctic.

Also, I’d like thank Max Posch (Coordination Centre for Effects, Netherlands) for his comments on the modified critical loads equation in Chapter 3. His knowledge and experience were a great addition to this thesis. Special thanks goes to Hazel Cathcart for her awesome computer skills (Long live the computer queen), and Kevin Adkinson for his laboratory assistance and for running many analyses. A special mention to Chris

Furgal and his lab Nasivvik, whom have provided priceless knowledge and suggestion on community engagement and presentations.

I thank the members of the Environmental GeoSciences Research Group, and to my peers at Trent Office SC104: Max Debuse, Phaedra Cowden, Conor Gaffney, Erin

Hayward and Abby Wynia, for sharing the experience of grad life, all the coffee runs, and for moral support. And lastly, on a personal note, to my family and friends, for your love and support.

vi Table of Contents

Abstract ...... ii Acknowledgements ...... iii List of Figures ...... vii List of Tables ...... ix Abbreviations...... xi

1.0 General Introduction ...... 1 1.1 Acidic Deposition ...... 1 1.1.1 Environmental Impacts ...... 3 1.1.2 Critical Loads Concept ...... 6 1.2 Canadian Region of the Arctic ...... 7 1.2.1 Emission of Air Pollution in the North ...... 8 1.2.2 Shipping in the Arctic ...... 10 1.2.3 Regulation of Atmospheric Emissions in the Arctic...... 13 1.3 Thesis Objective ...... 15 References ...... 17

2.0 Physical and chemical characteristics of 1300 lakes and ponds across the Canadian Arctic ...... 26 2.1 Abstract ...... 26 2.2. Introduction ...... 28 2.3 Methods ...... 30 2.3.1 Study Area ...... 30 2.3.2 Arctic hydrochemistry dataset ...... 34 2.3.3 Statistics ...... 37 2.4 Results ...... 38 2.4.1 Physical Characteristics ...... 42 v 2.4.2 Water pH and Conductivity ...... 47 2.4.3 Major Cations and Anions ...... 56 2.4.4 Phosphorus, Nitrogen, and Carbon ...... 58 2.4.5 Trace Metals ...... 60 2.4.6 Drivers and relationship of water chemistry ...... 62 2.5 Discussions ...... 63 2.6 Conclusion ...... 72 References ...... 74

3.0 Critical Loads of Acidity and Exceedances of 1138 Lake and Ponds in the Canadian Arctic ...... 83 3.1 Abstract ...... 83 3.2 Introduction ...... 85 3.3 Methods ...... 88 3.3.1 Study Area ...... 88 3.3.2 Arctic water chemistry database ...... 91 3.3.3 Steady-State Water Chemistry model ...... 94 3.3.4 Exceedances ...... 98 3.4 Results ...... 99 3.4.1 Critical loads of acidity ...... 99 3.4.2 Sulphur deposition and Exceedances ...... 105 3.5 Discussions ...... 109 3.6 Conclusions ...... 115 References ...... 117

4.0 General Conclusion ...... 124 4.1 Study Conclusion ...... 124 4.2 Contribution to research ...... 127 4.3 Recommendations and further work ...... 128 vi References ...... 129

Appendix A Supplementary information for Chapter 2 ...... 131 Appendix B Supplementary information for Chapter 3 ...... 138

vi List of Figures

1.0 General Introduction Figure 1.1 Lake trout conditions in Lake 223 water pH of 5.4 (above) Page 4 and 5.1 (below) (Source: Schindler et al., 1985). Figure 1.2 Annual sulphur oxide emission in Canada by source for the Page 9 year 1990 to 2015 (Source: Environment and Climate Change Canada, 2017a) Figure 1.3 Sulphur emissions (in tonnes) for the year 2000 and 2014 for Page 10 each northern Canadian region, i.e., Nunavut (above), Northwest Territories (middle), and Yukon (bottom). Amount of marine sulphur emissions are depicted as pink stacks. Source: Air Pollutant Emission Inventory by Environment and Climate Change Canada (2018) Figure 1.4 Map of the Canadian Arctic as well as the general pathway Page 12 of the Northwest Passage (shown in green) (Source: Commissioner of the Environment and Sustainable Development, 2014) Figure 1.5 Area within the North American Emission Control Zones Page 14 (ECA) in agreement with IMO MARPOL Annex VI (source: Dupont, n.d)

2.0 Physical and Chemical Characteristics of 100 Lakes and Ponds in The Arctic Figure 2.1 Names of islands and regions within the study region with Page 33 the AMAP boundary depicted in red (above). Location of all sites (sites from published sources are depicted as red dots, while sites collected from this study are depicted as yellow dots) within the study region. Figure 2.2 Location of all sites (published sources = red dots, this study Page 34 = yellow dots) among the different geology (Top) and ecoregions (bottom) types. Figure 2.3 Frequency of sites among different Elevation (Top), Page 46 Distance to coast (Middle), and Depth (Bottom). Figure 2.4 Correlation matrix of selected chemical variables with Page 47 Spearman’s rank correlation coefficient. Significant (p<0.01) correlations coefficient are highlighted in colour. Figure 2.5 Boxplot of conductivity (Left top), pH (Left middle), and DIC Page 49 (Left bottom) among geology types, and DOC (Right top), K (Right middle), and TN (Right bottom) among ecoregions. Significant differences (Kruskal–Wallis rank sum test and Dunn’s post hoc test with Bonferroni adjustment) between mean concentrations are indicated by * = p values <0.05, ** = p-value <0.01, and *** = p-value < 0.001.

vii Figure 2.6 Principal component analysis (PCA) of 27 physical and Page 63 chemical variables, and 613 sites from across 17 regions in the Canadian Arctic. Sites are depicted in four geological type from Harrison et al. (2011); Igneous (red), Sedimentary (Green), Supracrustal (Blue), and Unclassified (purple).

3.0 Critical loads of acidity for 1138 lakes and ponds across the Canadian Arctic Figure 3.1 Location names of islands and regions within the Canadian Page 90 Arctic (above). Locations of all sites (below) from both the database (red dots) and from the lake surveys (yellow dots), with the AMAP boundary depicted in red. Figure 3.2 Cumulative frequency distribution for critical loads of Page 100 acidity for the three ANClimits values. Where ANClimit= 8 µeq·L–1 for Brown Trout, 11 µeq·L–1 for Arctic Char, and 20 µeq·L–1 for Ecosystem. Figure 3.3 Critical loads of Acidity for 1138 sites using an ANClimit for Page 103 Arctic Char (11 µeq·L–1). Figure 3.4 Critical loads exceedances under 2010 sulphur deposition Page 107 scenario from the with shipping scenario. Figure 3.5 Critical loads exceedances under 2010 sulphur deposition Page 108 scenario from the without shipping scenario.

Appendix B Supplementary information for Chapter 2 Figure A1 Water DIC concentrations (mg·L–1) as a function of Page 137 alkalinity (mg·L–1) in arctic lakes and ponds (n = 98) in the Canadian Arctic. Figure A2 Ion balance of the major cations and anions for 224 lakes in Page 137 the Canadian Arctic

Appendix B Supplementary information for Chapter 3 Figure B1 Critical loads of Acidity for 1138 sites using an ANClimit Page 142 value for Brown Trout (8 µeq L–1), and relative sensitivity to acidification. Figure B2 Critical loads of Acidity for 1138 sites using an ANClimit Page 143 value for Ecosystem (not organic acid adjusted; 20 µeq L–1), and relative sensitivity to acidification. – Figure B3 Critical loads (with ANClimit value of Brown Trout, 8 µeq L Page 144 1) exceedances under 2010 with shipping sulphur deposition scenario. – Figure B4 Critical loads (with ANClimit value of Ecosystem, 20 µeq L Page 145 1) exceedances under 2010 with shipping sulphur deposition scenario. – Figure B5 Critical loads (with ANClimit value of Brown Trout, 8 µeq L Page 146 1) exceedances under 2010 without shipping sulphur deposition scenario

vii – Figure B6 Critical loads (with ANClimit value of Ecosystem, 20 µeq L Page 147 1) exceedances under 2010 without shipping sulphur deposition scenario Figure B7 Sulphur deposition (meq·m–2·yr–1) from both model; with Page 148 shipping (top) and without shipping (bottom) among sites in the Canadian Arctic.

vii List of Tables

2.0 Physical and Chemical Characteristics of 100 Lakes and Ponds in The Arctic Table 2.1 Summary of sampling sites per region with sampling year. Page 36 Table 2.2 Descriptive statistics of 26 parameters including; units, count, Page 39 mean, standard deviation (SD), minimum, maximum, and percentile (5th and 9th) values. Table 2.3 Summary of sampling sites per region with sampling year Page 41 Table 2.4 Median (mean) values and site count for selected water Page 43 chemistry variables and ratios for each geology type. Table 2.5 Median (mean) values and site count for selected water Page 44 chemistry variables and ratios for each ecoregion; Arctic Cordillera (AC), Northwestern Forested Mountains (NWF), Taiga (TA), and Tundra (TU). Table 2.6 Median (mean) values and site count for selected water Page 50 chemistry variables and ratios for each region. Table 2.7 Significant differences calculated with Kruskal–Wallis rank Page 54 sum test and Dunn’s post hoc test with Bonferroni adjustment for chemical variables among the geology types and are indicated by * = p values <0.05, ** = p-value <0.01, and *** = p-value < 0.001. Table 2.8 Significant differences calculated with Kruskal–Wallis rank sum test Page 55 and Dunn’s post hoc test with Bonferroni adjustment for chemical variables among the ecoregion types and are indicated by * = p values <0.05, ** = p-value <0.01, and *** = p-value < 0.001. AC = Arctic Cordillera, TU = Tundra, TA = Taiga, and NWF = Northwestern Forested Mountains.

3.0 Critical loads of acidity for 1138 lakes and ponds across the Canadian Arctic Table 3.1 Number of lakes, and average values for runoff and selected Page 92 chemical variables for regions of the Canadian Arctic. Table 3.2 Selected statistics (mean, standard deviation, minimum, Page 100 maximum, median, 5th percentile, and 95th percentile) of CL(A) –1 for the three ANClimits. Where ANClimit= 8 µeq·L for Brown Trout, 11 µeq·L–1 for Arctic Char, and 20 µeq·L–1 for Ecosystem Table 3.3 Site count and mean CL(A) (median) values in meq·m–2·yr–1 Page 100 per geological type and among the three ANClimits values. –1 –1 Where ANClimit= 8 µeq·L for Brown Trout, 11 µeq·L for Arctic Char, and 20 µeq·L–1 for Ecosystem. Table 3.4 Percentage (and count) of sites within different acid sensitivity Page 101 –2 –1 classes (meq·m ·yr ) among the three ANClimits values. –1 –1 Where ANClimit= 8 µeq·L for Brown Trout, 11 µeq·L for Arctic Char, and 20 µeq·L–1 for Ecosystem.

ix Table 3.5 Percentage (and number) of sites (under Arctic Char ANClimit) Page 102 within different acid sensitivity classes (values are expressed in meq·m–2·yr–1) among different regions of the Canadian Arctic. Table 3.6 Number of sites mean CL(A) (calculated with Arctic Char Page 104 –2 –1 –2 – ANClimit; meq·m ·yr ), mean sulphur deposition (meq·m ·yr 1) from both model scenarios, and exceedances (percentage of total sites and number of sites under both models) for each region of the Canadian Arctic. Table 3.7 Selected statistics (Mean, standard deviation (SD), minimum, Page 106 maximum, median, 5th percentile, and 95th percentile) of total sulphur deposition values (meq·m–2·yr–1) for both scenarios.

Appendix A Supplementary information for Chapter 2 Table A1 Descriptive statistics of 59 parameters including; count, mean, Page 132 standard deviation (SD), minimum, maximum, and percentile (5th and 9th) values. Table A2 List of PCA variables and their loadings for the five principal Page 134 components with eigenvalues above 1. Table A3 Median (mean) sea-salt corrected concentrations for Ca, Mg, Page 135 Na, K, and SO4, per region, ecoregion, and geology type.

Appendix B Supplementary information for Chapter 3 Table B1 Percentage of sites (under Brown Trout ANClimit) within Page 138 different acid sensitivity classes (values are expressed in meq·m–2·yr–1) among different regions of the Canadian Arctic. Table B2 Percentage of sites (under Ecosystem ANClimit) within Page 139 different acid sensitivity classes (values are expressed in meq·m–2·yr–1) among different regions of the Canadian Arctic. Table B3 Number of sites, mean CL(A) (calculated with Brown Trout Page 140 –2 –1 ANClimit value; meq·m ·yr ), mean sulphur deposition (meq·m–2·yr–1) from both model, and exceedances (percentage of total sites and number of sites under both models) for each region of the Canadian Arctic. Table B4 Number of sites, mean CL(A) (calculated with the Ecosystem Page 141 –2 –1 ANClimit value; meq·m ·yr ), mean sulphur deposition (meq·m–2·yr–1) from both model, and exceedances (percentage of total sites and number of sites under both models) for each region of the Canadian Arctic.

x List of abbreviations Acronyms AC Arctic Cordillera ANC Acid Neutralization Capacity ANClimit Acid Neutralization Capacity limit ANCoaa.limit Organic Acid Adjusted Acid Neutralization Capacity limit BAU Business-As-Usual scenario Cond Conductivity CL Critical Load CL(A) Critical Load of Acidity DistC Distance to Coast ECA Emission Control Area Elev Elevation Ex Exceedances GEM-MACH Global Environmental Multi-scale - Modelling Air quality and CHemistry HG High-Growth scenarios IMO International Maritime Organization Is. Island LRTAP Long-Range Transboundary Air Pollution MARPOL International Convention for the Prevention of Pollution from Ships NOx Nitrogen Oxides NWF Northwestern Forested Mountains NWP North-West Passage PCA Principal Component Analysis Sdep Sulphur deposition SOx Sulphur Oxides SSWC Steady State Water Chemistry model TA Taiga TU Tundra

Nomenclature Al Aluminum ALK Alkalinity [BC*]0 Pre-acidification sea-salt corrected catchment-average net base cation flux [BC*†]t Current observed base cation concentration corrected both for sea-salts and geological sulphate Ca Calcium Ca* Sea-Salt Corrected Calcium concentration Cl Chloride xii DIC Dissolve Inorganic Carbon DOC Dissolve Organic Carbon F F-Factor, the change in non-marine base cations concentration in relationship to deposition of strong acid anions H Protons HNO3 Nitric Acid H2SO4 Sulphuric Acid K Potassium K* Sea-Salt Corrected Potassium concentration Mg Magnesium Mg* Sea-Salt Corrected Magnesium concentration Na Sodium Na* Sea-Salt Corrected Sodium concentration Na:Cl Sodium – Chloride ratios Na:k Sodium - Potassium NH3 Ammonia NH4 Ammonium NO Nitrogen Oxide NO2 Nitrite NO3 Nitrate POC Particulate Organic Carbon Q Annual runoff Sdep.0 Pre-industrial concentration sulphur deposition Sdep.t Current observed sulphur deposition SiO2 Silica SO4 Sulphate SO4* Sea-Salt Corrected Sulphate concentration

[SO4]geo Geological sulphate [SO4*]t Current observed sea-salt corrected sulphate concentration TKN Total Kjeldahl Nitrogen TN Total Nitrogen TN:TP Total Nitrogen – Total Phosphorus ratios TP Total Phosphorous

xii 1.0 General Introduction

1.1 Acid Deposition

The industrial revolution brought power to steam engines and heat to homes in the late 1700s, but also initiated the beginning of wide scale environmental degradation owing to the burning of fossil fuels. This production of power directly resulted in emissions of sulphur dioxide (SO2), which is one of the main precursors (other than nitrogen dioxide, NO2) of acidic deposition. Anthropogenic emissions of nitrogen oxides

(NOx ≡ NO2 + NO) are primarily associated with vehicular emissions. When SO2 is released into the atmosphere, it combines with water vapor to form sulphuric acid

(H2SO4). Emissions of NO2 produce nitric acid (HNO3), also through interactions with water vapor. The term acidic deposition (also known as acid rain) refers to dry (ash, gases, or particles) and wet deposition (rain, snow, or hail) of acidifying compounds.

European emissions peaked during the 1980s, when emissions of anthropogenic sulphur

6 oxides (SOx) were estimated to be ~3.5×10 kt of S (Stern, 2005) and nitrogen oxides

(NOx = NO + NO2) were estimated at ~0.02 kt of N (Dignon and Hameed, 1989). Global emissions of SO2 were approximately 2.1 Gg in the 1850, and increased to 3.8 Gg in

2005, with the greatest contributor being coal combustion (Smith et al., 2011). In contrast, natural emissions of SO2 can be derived from natural (grassland and forest fires; Smith et al., 2001) and geological sources (Smoking Hills, NWT; Mathews and Bustin, 1984) and they are insignificant when compared to anthropogenic emissions (Smith et al., 2001).

Acidic deposition dissociates into strong acidic anions (sulphate [SO4] and nitrate

[NO3]), and into protons [H]. The release of [H] displaces base cations (calcium [Ca], magnesium [Mg], potassium [K], and sodium [Na]), and aluminum [Al]. Studies have shown that enhanced deposition of H2SO4 is associated with decreasing exchangeable base cation concentrations in soils (Lawrence et al., 1999), and depletion in base cation pools (Cronan and Grigal, 1995; Likens et al., 1996). If soils contain an insufficient base cation pool to neutralise incoming acidity, the result is soil acidification. This leaching of base cations (and acidic anions) from soil to surface waters (Likens et al., 1996; Molot and Dillion, 2008) has been widely observed in streams (Kirchner et al., 1992). For example, increased base cations concentrations were found to be associated with higher

[SO4] and [NO3] in streams, in response to high inputs of acidity at the Hubbard Brook long-term monitoring catchment in New Hampshire (Likens et al., 1996). In lentic aquatic systems, i.e., freshwater lakes and ponds, the ratio of anions to base cations determines the system’s sensitivity to acidity. The Acid Neutralizing Capacity (ANC) of surface waters is often used to describe the ability of an aquatic system’s ability to tolerate inputs of acidity. Following Reuss and Johnson (1986), ANC is defined as the sum of base cations and ammonia minus the sum of acid anions, as follows

ANC = ∑Cation – ∑Anions = ([Ca] + [Mg] + [K] + [Na] + [NH4]) – ([Cl] + [SO4] +

[NO3])

Surface waters with ANC values <50 µeq·L-1 are sensitive to acidification leading to possible damage to biota, while sites with ANC >50 µeq·L-1 are considered to be relatively insensitive to acidic deposition (Driscoll et al., 2001; Arctic Monitoring and

Assessment Programme, 2006). The input of base cations to lakes is dependent on the 2

leaching of base cations from their surrounding catchment; lakes with higher ANC are often associated with soils with higher cation pools. Houle et al. (2006) best described this through their findings that higher Ca and Mg pools were associated with lake Ca (r2 =

0.83) concentrations, and lake ANC (r2 = 0.75) for 21 sites in Quebec, Canada.

Acidification of surface waters occurs when elevated acidity (from inputs of protons, through the dissociation of H2SO4) lowers the pH and ANC. Early lake surveys in the late 1970s observed that inputs of acidic precipitation (pH <4; Beamish and

Harvey, 1972; Likens and Bormann, 1974; Dillon et al., 1978) were linked to decreased pH (4.0–4.4, Beamish and Harvey, 1972; <5, Kretser et al., 1989), declined ANC (<0

µeq·L-1), elevated sulphate concentrations (Landers et al., 1988; Baker et al., 1991), and decreased alkalinity levels (Henriksen, 1979) in surface waters.

1.1.1 Environmental Impacts

Robert Angus Smith (1872) first described the effects of acidic deposition, in large coal powered towns during the 19th century. However, it was not until the mid-20th century when concerns regarding acidic deposition grew through the observations of reduced forest growth (Abrahamsen et al., 1975; Tamm, 1976), loss of fish populations

(Jensen and Snekvik, 1972; Wright et al., 1976; Hesthagen 1989, 1995), and changes in water chemistry (Odén, 1976; Wright and Henriksen, 1977), in central Europe and

Southern Fennoscandia, were linked to the long-range transport of SO2 emissions in the

United Kingdom and Germany .

The acidification of surface waters has a negative impact on aquatic biota.

Although, some species of adult fish have been found in naturally occurring acidic lakes (e.g.,

Largemouth Bass (Micropterus salmoides), Walleye (Sander vitreus), Northern Pike (Esox lucius) at pH < 5; Webster et al.,

1993), those most sensitive to acidification are lower level trophic organisms such as molluscs, crustaceans, smaller Figure 1.1 Lake trout conditions in Lake 223 water fish species, and young fish pH of 5.4 (above) and 5.1 (below) (Source: Schindler et al., 1985). (Økland and Økland, 1986; Mills and Schindler, 1986). In an 8-year whole lake acidification (reduction of pH from 6.5 to

5.0) experiment (Lake 223, ELA, Ontario), Schindler et al. (1985) found that the reduction of the Lake trout (Salvelinus namaycush) population was the result of two drivers; 1) the lost of lower trophic level organism such as opossum shrimp (Mysida sp.), fathead minnows (Pimephales promelas), copepod (diaptomus spp.), bluntnose minnows

(Pimephales sp.), crayfish (Orconectes spp.), and leeches (Hirudinea spp.), which resulted in cannibalism and the deterioration of health (Figure 1.1) of the lake trout

population, and 2) chronic stress (from starvation) resulted in the female fish’s failure to release embryos, ultimately reducing recruitment of the next generation of fish. Although aluminum (Al3+) concentrations associated elevated acidity have been found to affect fish populations (Driscoll et al., 1980; Schofield and Trojnar, 1980; Baker and Schofield,

1992), aluminum was found at low concentrations in Lake 223 (7–36 µg·L-1; Schindler et al., 1985). Further, lack of recruitment was found to be the driver for the crash of fish stocks in the La Cloche Mountains, with many of the fish stocks (lake trout, lake herring

(Coregonus artedii), and white suckers (Catostomus commersoni)) absent by the early

1970s (Beamish and Harvey, 1972). In Norwegian watersheds, years of acidification affected an estimated area of 1417 km2 in 1940, which increased to 51,530 km2 in 1990, resulting in the loss of >8,000 Brown Trout (Salmo trutta) populations, and a further

>3,000 populations at different stages of decline (Hesthagen et al., 1999). In

Fennoscandia, loss of fish populations were estimated at 5,122 (16.1%), 1202 (6.5%), and

223 (4.0%) for Norway, Sweden, and Finland during the 1990’s, respectively (Tammi et al., 2003).

Ecosystem recovery from acidification is a slow process with symptoms of acidification still present long after acidic deposition has been reduced. For example, in catchments, acid loading through stored acid (melting of acidic snow, acidic precipitation from storms) of soils will continue to leach out base cations (Kirchner and Lydersen,

1995; Watmough et al., 2005b), into surface waters (Watmough et al., 2005a).

Improvements of ecosystem health can be obtained through emissions reduction policies such as the 1979 Convention on Long-Range Transboundary Air Pollution (LRTAP;

Economic Commission for Europe, 1979). Under LRTAP Convention (via its protocols),

European emissions have decreased by 73–80% for SOx, and 31–41% for NOx between

1980–2000 (Fowler et al., 2007). Long-term monitoring sites in forest ecosystems have

2− shown signs of recovery in soils though decreasing SO4 , decreasing of soil water base cations (leaching), increasing pH, and deceasing Al3+ concentration, in both Europe

(Löfgren and Zetterberg, 2011; Akselsson et al., 2013; Johnson et al., 2018), and North

American (Lawrence et al., 2015). However, different climatic conditions and biogeochemistry in soils may inhibit recovery, especially soils in deeper horizons (Berger et al., 2016; Johnson et al., 2018). Similarly, trends in water chemistry of lakes have

2− indicated a decrease in [SO4 ], decrease in base cation, and an increase in ANC across

North America and Europe (Skjelkvåle et al., 1999; Stoddard et al., 1999; Driscoll et al.,

2003; Skjelkvåle et al., 2005; Futter et al., 2014; Driscoll et al., 2016). Some biological

(macrophyte, Baastrup‐Spohr et al., 2017; benthic, Gunn and Keller, 1990; fish, Snucins and Gunn, 1998; Caputo et al., 2016) populations in these regions have shown signs of recovery from acid deposition.

1.1.2 Critical Loads Concept

The critical load (CL) concept is used to reinforce emissions reduction policies by linking ecosystems health to the deposition of atmospheric pollutants. A CL is an estimate of that amount of pollutant that an ecosystem can tolerate without causing significant harm on sensitive biological components (Nilsson and Grennfelt, 1988). A critical load of acidity [CL(A)] is the maximum amount of acidity that an ecosystem can tolerant without causing harm to sensitive components. Ecosystems with higher CL(A) are able to tolerate higher amounts of acidity and are less sensitive to inputs of acidity (while those with lower CL(A) are sensitive to inputs of acidity). Exceedance occurs when acidic inputs 6

(i.e., deposition) are greater than the ecosystem’s CL, leading to a greater risk of negative impacts. Steady-state models have been developed to quantify CL for both soils (De

Vries, 1991) and surface waters (Henriksen et al., 1992). The use of CL was adopted under the LRTAP to underpin emissions reduction protocols in response to concerns regarding air pollution impacts to forest soils and surface waters. For example, mapping of critical loads and their exceedance in Europe (Hettelingh et al., 1991) lead to the 1994

Oslo Protocol on Further Reduction of Sulphur Emissions.

1.2 The Canadian Arctic

The Canadian Arctic (area > 60⁰N) is approximate 4.0 ×106 km2 (AMAP, 1998) and includes an estimated 42.3% (n = 1.49×106; Paltan et al., 2015) of all lakes in the

Arctic region. This vast area includes the Territories of Nunavut, Northwest Territories,

(and their islands) and Yukon, which contains an estimated 36.8% (Nunavut = 17.6%,

Northwest Territories = 18.3, Yukon = 0.9%) of Canada’s freshwater area (in km2;

Statistic Canada, 2012). Many of these lakes are situated on either the Arctic Platform or the Precambrian (Canadian) Shield. The Arctic platform consists of Paleozoic sedimentary geology that originated from historical coastal plains and shallow marine beds, while the Shield consists of crystalline and igneous rock. Common geological materials found on the Arctic Platform are shale, siltstone, sandstone, limestone, and dolomite, while gabbros, gneisses, granitic, and volcanic rocks are common on the Shield

(Fulton 1989). Low year-round temperatures (–33 to –35⁰C in January, and 8 to 10⁰C in

July) and low precipitation (< 500 mm) (Maxwell, 1981) render soils to be poorly developed. Thus, lakes on the Arctic Platform typically have more alkaline lakes and ponds, e.g., mean pH = 8.1–8.5 (Michelutti et al., 2010; Rautio, 2011). Whereas those on 7

more acid-sensitive terrain (Precambrian Shield) tend to have a mean pH of 6.1–7.0

(Rühland and Smol, 1998; Joynt III and Wolfe, 2001; Rühland et al., 2003). However, on both geological types, base cations concentrations are typically found in low concentrations when compared to other regions of Canada (McNeely et al., 1979). This suggest that many Arctic lakes, especially those on Precambrian Shield, may have low

CL(A) and are sensitive to acidic depositon.

1.2.1 Emissions of Air Pollutants in the North

The successful ratification of the LRTAP Convention and the Canada–U.S. Air

Quality Agreement has resulted in the reduction in sulphur dioxide (SO2) emission in

North America. As part of the process, CL have been mapped for much of Canada (Forest

Soils, Carou et al., 2008; Lakes, Aherne and Jeffries, 2015) and the United States (Dupont et al., 2005; McNulty et al., 2007; Pardo et al., 2011) and have contributed to the decline of emissions of SOx by 55% (1,054 kt) from 1990–2015 in Canada (Figure 1.2;

Environment and Climate Change Canada, 2017a) and by 73% (16,800 Kt) from 1990–

2011 in the United States (Environmental Protection Agency, 2015).

3500

3000

2500

2000

1500

1000

500 Total sulphur oxide emissions Totalsulphur oxideemissions (Kt)

0 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014

Figure 1.2 Annual sulphur oxide emission in Canada by source for the year 1990 to 2015 (Source: Environment and Climate Change Canada, 2017a) Within the northern region of Canada (Nunavut, Northwest Territories, and

Yukon) emissions of sulphur have been relatively stable (Figure 1.3) when compared with the national emissions (Figure 1.2). Of the three northern territories, Nunavut was regarded as the largest emitter in 2014 (2,468 tonnes), followed by Northwest Territories

(1,051 tonnes) and Yukon (983 tonnes) (Figure 1.3). Generally, the largest contributors of sulphur emission for each territory are marine transportation (Nunavut), mining

(Northwest Territories), and oil industry (Yukon) (Environment and Climate Change

Canada, 2018). Marine transportation accounted for the third largest source of sulphur for

Northwest Territories (289 tonnes) and Yukon (289 tonnes) for 2015 (Figure 1.3). This is expected, as marine access regions (such as Iqaluit in Nunavut and Tuktoyaktuk in

Northwest Territories) are reliant on marine shipping for the transportation of goods.

Figure 1.3 Sulphur emissions (in tonnes) for the year 2000 and 2014 for each northern Canadian region, i.e., Nunavut (above), Northwest Territories (middle), and Yukon (bottom). Amount of marine sulphur emissions are depicted as pink stacks. Source: Air Pollutant Emission Inventory by Environment and Climate Change Canada (2018).

1.2.2 Shipping in the Arctic

Although the Arctic is relatively pristine, long-range transport by the atmosphere can enhance contaminant concentrations in Arctic ecosystems. Sources of acidic air pollutants in the Arctic occur as point source, e.g., the Severonickel smelter, Kola

Peninsula, Northern Russia, or through long-range transport and cold condensation

(Wania and Mackay, 1993) where pollutants are able to move into the Arctic and condense onto Arctic ecosystems. It was estimated that > 90% of anthropogenic sulphur

2− originates from Europe and Asia (Barrie et al., 1989). Although trends of SO4 among

Arctic air and precipitation have show a decreasing trend (Hole et al., 2009), there has been a growing concern regarding the impacts of ship-sourced sulphur emissions. Sea ice extent within the Canadian Arctic decreased by 75% between 1969 and 2015

(Environment and Climate Change Canada, 2016), which corresponded similarly to other regions of the Arctic (Perovich et al., 2017). The reduction of sea ice will increase opportunities for resource extraction, tourism, fishing, and shipping in the Northern Sea

Route (Atlantic to Pacific Ocean through the Russian Arctic), the Northwest Passage

(Pacific to Atlantic Ocean through the Canadian Arctic) and the Hudson Strait (Atlantic

Ocean to Churchill, Manitoba) (Figure 1.4).

Figure 1.4 Map of the Canadian Arctic as well as the general pathway of the Northwest Passage (shown in green) (Source: Commissioner of the Environment and Sustainable Development, 2014) Between the year 1990 and 2015, shipping (in km travelled) increased from

345,567 km to 793,684 km for the territory of Nunavut (Daweson et al., 2017) and from

50 to 450 km for the Beaufort Sea (Pizzolate et al., 2016). Most marine fuels are residual fuels from the refining of crude oils, which have higher concentrations of impurities and sulphur content (Corbett and Fischbeck, 1997). Heavy fuels (>0.5% sulphur content) are often less expensive than those with more sulphur content, in which it is estimated that approximately 0.5% of all (global) marine fuels, have a sulphur content less than 0.5%

(International Maritime Organization, 2009). The increase in ship traffic in the Arctic is

-1 estimated to release approximate 34,000 and 231,000 Mt·yr of SO2 and NO2 by the year

2020, with container shipping accounting for 27% of all emissions (Corbette et al., 2010).

Increased emissions of SO2 and NO2 are expected to increased sulphur and nitrogen deposition in the surrounding landscape. For example, shipping emissions were found to contribute 10–25% and 30–50% of sulphate and nitrate deposition along the Norwegian coast (Dalsøren et al., 2007). Within Canada, shipping emissions are expected to contribute 20–100% and 10–50% of ambient SO2 and NO2 concentrations (Gong et al.,

2018). This is expected to impact air quality in northern communities, for example,

Aliabadi et al. (2015) have reported that shipping emissions have contribute to 18.1 and

17.5% of ambient SO2 and NO2 concentrations in two northern communities (Cape

Dorset and Resolute, Nunavut). Currently, very few studies have assessed the CL(A) in the Canadian region of the Arctic (AMAP, 2006; Forsius et al., 2010).

1.2.3 Regulation of Atmospheric Emissions in the Arctic

To regulate emissions from ships, the United Nations International Maritime

Organization (IMO) developed the International Convention for the Prevention of

Pollution from Ships (MARPOL) in 1973, which is currently ratified by 156 nation states that represent 99% of the world’s shipping tonnage (International Maritime Organization,

2018). Under MARPOL, there are six Annexes that aim to prevent and minimize specific pollutants; oil, harmful substances, sewage, air pollution, etc. To regulate shipping emissions, Canada adopted MARPOL’s Annex VI Prevention of Air Pollution from Ships

(19 May 2005) into its 2001 Shipping Act (Regulations for the Prevention of Pollution from Ships and for Dangerous Chemicals), which came into force in July 1, 2010, aims to limit the emissions of SOx and NOx from ship origin (Environment and Climate Change

Canada, 2017b). In addition to emissions regulations, the Annex VI also established the

North America Emission Control Areas (NA-ECAs; MARPOL Annex VI Regulation 13

13.6.1; Figure 1.5) in which stricter regulation on ship emissions (SOx and NOx) are established from the coast of Labrador to Texas, and from California to Anchorage,

Alaska. Within the NA-ECAs, usage of marine fuel must have a sulphur content of 0.1%, as of January 1, 2015 (International Maritime Organization, n.d). Outside of the NA-

ECAs, marine fuel must have a sulphur content of 3.5% and will decrease to 0.5% as of

January 2020 (International Maritime Organization, n.d). Currently, the Canadian Arctic is not under an ECA, but the establishment of one would further reduce SOx emission from shipping and limit the protected increase in shipping emissions of SOx.

Figure 1.5 Area within the North American Emission Control Zones (ECA) in agreement with IMO MARPOL Annex VI (source: Dupont, n.d)

1.3 Thesis Objectives

The primary objectives of this study were to assess the hydrochemical characteristics of lakes and ponds in the Canadian Arctic, and to determine their acid sensitivity through the CL approach. This thesis was written in manuscript style, which includes a general introduction (Chapter 1), two manuscript style chapters (Chapter 2 and

3), and a general conclusion (Chapter 4). Chapters 2 and 3 address the primary objectives of the thesis, where Chapter 3 builds on the results of Chapter 2.

Chapter 2, titled Physical and chemical characteristics of 1300 lakes and ponds in the

Canadian Arctic, evaluated the hydrochemical and physical properties (latitude, longitude, elevation, depth, area, distance to coast, ecoregion, and geology type) of lakes and ponds across the Canadian Arctic. Hydrochemical data for Arctic lakes and ponds (n

= 1200) were collated from a combination of published articles (n = 27), reports (n = 2), and graduate theses (n = 3), along with new survey data (Baffin Is. (n = 80), Coats Is. (n

= 10), Prince Charles Is. (n = 4), and eastern Northwest Territories (n = 6). Sites sampled on Baffin Is. were collected by the author of this thesis, while those collected on Coats Is.

(n = 10), Prince Charles Is. (n = 4), and eastern Northwest Territories (n = 6) were collected in collaboration with other research groups. The author did all the laboratory work.

The objectives of Chapter 2 were to assess the physical and chemical data of 1300 lakes and ponds, provide regional baseline hydrochemical data, and to evaluate drivers of water chemistry in the Canadian Arctic environment.

Chapter 3, titled Critical loads and Exceedances of 1138 Lakes and Ponds in the

Canadian Arctic, evaluated the acid sensitivity of lakes and ponds in the Canadian Arctic.

The study focused on a subset of 1138 sites from Chapter 1 using a modified critical loads model for aquatic systems (Steady State Water Chemistry Model; Henriksen, 1979) and evaluated exceedance under two sulphur deposition scenarios; with shipping and without shipping for the year 2010.

The objectives of Chapter 3 were to calculate critical loads of acidity for lakes and ponds in the Canadian Arctic and to determine the risk of acidification (exceedances) for with and without shipping scenario for the year 2010

Significance of research

The spatial capacity of this research will integrate existing small-scale lake surveys into a single assessment. The evaluation of water chemistry from 1300 lakes will contribute to a better understanding of baseline water chemistry data, which will be valuable in an environment heavily influence by climate change. When applied with the critical loads approach, results will inform Government organizations (Environment and

Climate Change Canada and Transport Canada), and international organization [Arctic

Monitoring and Assessment Programme (AMAP), Conservation of Arctic Flora and

Fauna (CAFF), and International Cooperative Programme for assessment and monitoring of the effects of air pollution on rivers and lakes (ICP-Waters)], on the potential acidification of Arctic lake systems under elevated marine shipping emissions and will support the International Maritime Organization’s (IMO) assessment on the potential establishment of an ECA within the Canadian Arctic.

References

Abrahamsen, G., Horntvedt, R., & Tveite, B. (1977). Impacts of acid precipitation on coniferous forest ecosystems. Water, Air, & Soil Pollution, 8(1): 57-73.

Aherne, J., & Jeffries, D. (2015). Critical load assessments and dynamic model applications for lakes in North America. In Critical Loads and Dynamic Risk Assessments (pp. 485-503). Springer, Dordrecht.

Akselsson, C., Hultberg, H., Karlsson, P.E., Karlsson, G.P., & Hellsten, S. (2013). Acidification trends in south Swedish forest soils 1986–2008—Slow recovery and high sensitivity to sea-salt episodes. Science of the Total Environment. 444:271-287.

Aliabadi, A.A., Staebler, R.M., & Sharma, S. (2015). Air quality monitoring in communities of the Canadian Arctic during the high shipping season with a focus on local and marine pollution. Atmospheric Chemistry and Physics, 15(5), 2651-2673.

Arctic Monitoring and Assessment Programme. (1998). AMAP Assessment Report: Arctic Pollution Issues, Arctic Monitoring and Assessment Programme AMAP, Oslo, Norway. 859p.

Arctic Monitoring and Assessment Programme. (2006). Arctic Monitoring and Assessment Programme Assessment Report: Acidifying Pollutants, Arctic Haze, and Acidification in the Arctic, Arctic Monitoring and Assessment Programme AMAP, Oslo, Norway. Xii+859p

Baastrup‐Spohr, L., Sand‐Jensen, K., Olesen, S.C., & Bruun, H.H. (2017). Recovery of lake vegetation following reduced eutrophication and acidification. Freshwater Biology, 62(11): 1847-1857.

Baker, J.P., & Schofield, C.L. (1982). Aluminum toxicity to fish in acidic waters. In Long-Range Transport of Airborne Pollutants. (289-309). Springer, Dordrecht.

Baker, L.A., Herlihy, A.T., Kaufmann, P.R., & Eilers, J.M. (1991). Acidic lakes and streams in the United States: the role of acidic deposition. Science, 1151-1154.

Barrie, L.A., Olson, M.P., & Oikawa, K.K. (1989). The flux of anthropogenic sulphur into the Arctic from mid-latitudes in 1979/80. Atmospheric Environment (1967), 23(11), 2505-2512.

Beamish, R.J., & Harvey, H.H. (1972). Acidification of the La Cloche Mountain Lakes, Ontario, and resulting fish mortalities. Journal of the Fisheries Board of Canada, 29(8): 1131-1143.

Berger, T.W., Türtscher, S., Berger, P., & Lindebner, L. (2016). A slight recovery of soils from Acid Rain over the last three decades is not reflected in the macro nutrition of

beech (Fagus sylvatica) at 97 forest stands of the Vienna Woods. Environmental pollution, 216:624-635.

Caputo, J., Beier, C.M., Fakhraei, H., & Driscoll, C.T. (2016). Impacts of acidification and potential recovery on the expected value of recreational fisheries in Adirondack lakes (USA). Environmental science & technology, 51(1): 742-750.

Carou, S., Aherne, J., Ouimet, R., Arp, P.A., Watmough, S.A., DeMerchant, I., Shaw, M., Vet, R., Bouchet, V., & Moran, M. (2008). A National Picture of Acid Deposition Critical Loads for Forest Soils in Canada. A report for the CCME Acid Rain Task Group. Canadian Council of Ministers of the Environment

Commissioner of the Environment and Sustainable Development. (2014). Chapter 3— Marine Navigation in the Canadian Arctic. 2014 Fall Report of the Commissioner of the Environment and Sustainable Development. Office of the Auditor General of Canada. ISBN: 978-1-100-24947-6. Available Online: http://www.oag- bvg.gc.ca/internet/English/parl_cesd_201410_e_39845.html

Corbett, J.J., Lack, D.A., Winebrake, J.J., Harder, S., Silberman, J.A., & Gold, M. (2010). Arctic shipping emissions inventories and future scenarios. Atmospheric Chemistry and Physics, 10(19):9689-9704.

Cronan, C.S. and Grigal, D.F. (1995). Use of calcium/aluminum ratios as indicators of stress in forest ecosystems. Journal of Environmental Quality, 24(2): 209-226.

Dalsøren, S.B., Endresen, Ø., Isaksen, I.S., Gravir, G., & Sørgård, E. (2007). Environmental impacts of the expected increase in sea transportation, with a particular focus on oil and gas scenarios for Norway and northwest Russia. Journal of Geophysical Research: Atmospheres, 112(D2).

Dawson, J., Copland, L., Mussells, O., and Carter, N. 2017. Shipping Trends in Nunavut 1990-2015: A report prepared for the Nunavut General Monitoring Program. Ottawa, Canada and Iqaluit, Nunavut.p68

De Vries, W. (1991). Methodologies for the assessment and mapping of critical loads and of the impact of abatement strategies on forest soils. Wageningen (The Netherlands), DLO The Winand Staring Centre. Report 46. 109p.

Dignon, J., & Hameed, S. (1989). Global emissions of nitrogen and sulfur oxides from 1860 to 1980. JAPCA, 39(2):180-186.

Dillon, P.J., Jeffries, D.S., Snyder, W., Reid, R., Yan, N.D., Evans, D., Moss, J., & Scheider, W.A. (1978). Acidic precipitation in south-central Ontario: recent observations. Journal of the Fisheries Board of Canada, 35(6):809-815.

Dranga, S.A., Hayles, S., & Majewskis, K. (2017). Synthesis of limnological data from lakes and ponds across Arctic and Boreal Canada. Arctic Science. DOI: 10.1139/AS- 2017-0039

Driscoll Jr, C.T., Baker, J.P., Bastogne Jr, J.J., & Schofield, C.L. (1980). Effect of aluminium speciation on fish in dilute acidified waters. Nature, 284(5752): 161.

Driscoll, C.T., Lawrence, G.B., Bulger, A.J., Butler, T.J., Conan, C.S., Eagar, C., Lambert, K.F., Likens, G.E., Stoddard, J.L. and Weathers, K.C. (2001). Acidic Deposition in the Northeastern United States: Sources and Inputs, Ecosystem Effects, and Management Strategies: The effects of acidic deposition in the northeastern United States include the acidification of soil and water, which stresses terrestrial and aquatic biota. AIBS Bulletin, 51(3): 180-198.

Driscoll, C.T., Driscoll, K.M., Roy, K.M., & Mitchell, M.J. (2003). Chemical response of lakes in the Adirondack region of New York to declines in acidic deposition. Environmental science & technology, 37(10):2036-2042.

Driscoll, C.T., Driscoll, K.M., Fakhraei, H., & Civerolo, K. (2016). Long-term temporal trends and spatial patterns in the acid-base chemistry of lakes in the Adirondack region of New York in response to decreases in acidic deposition. Atmospheric Environment, 146:5-14.

Dupont, J., Clair, T. A., Gagnon, C., Jeffries, D. S., Kahl, J. S., Nelson, S. J., & Peckenham, J. M. (2005). Estimation of critical loads of acidity for lakes in northeastern United States and eastern Canada. Environmental monitoring and assessment, 109(1–3): 275–92.DOI: 10.1007/s10661-005-6286-x

Dupont. (N.d.). Is your ship ready to go in accordance with IMO Annex VI. http://www.dupont.ru/products-and-services/clean-technologies/products/belco- clean-air/sub-products/belco-marine-systems/belco-marine-regulatory-issues.html

Economic Commission for Europe. (1979). 1979 Convention on Long-Range Transboundary Air Pollution. Available Online: https://www.unece.org/fileadmin/DAM/env/lrtap/full%20text/1979.CLRTAP.e.pdf

Environment and Climate Change Canada (2016). Canadian Environmental Sustainability Indicators: Sea Ice in Canada. Available at: www.ec.gc.ca/indicateursindicators/default.asp?lang=en&n=1F76A73A-1.

Environment and Climate Change Canada. (2017a.) Canadian Environmental Sustainability Indicators: Air pollutant emissions. Available at: www.ec.gc.ca/indicateursindicators/default.asp?lang=en&n=E79F4C12-1

Environment and Climate Change Canada. (2017b). Compendium of Canada’s Engagement In International Environmental Agreements. Environment and Climate Change. Available online:

https://www.canada.ca/content/dam/eccc/migration/main/international/85fbcf8a- 229e-4861-b58c-6c9f68ab96ca/a13-202017-20iea-20factsheet-20marpol-20en- 20final.pdf

Environment and Climate Change Canada. (2018). Air Pollutant Emission Inventory (APEI) Historical Trends. Environment and Climate Change Canada. Available Online: http://donnees.ec.gc.ca/data/substances/monitor/air-pollutant-emission- inventory-apei-historical-trends/?lang=en

Environmental Protection Agency. (2015). EPA's Report on the Environment - Technical Documentation for Sulfur Dioxide Emissions. Available at: https://cfpub.epa.gov/roe/indicator.cfm?i=22

Forsius, M., Posch, M., Aherne, J., Reinds, G. J., Christensen, J., & Hole, L. (2010). Assessing the Impacts of Long-Range Sulfur and Nitrogen Deposition on Arctic and Sub-Arctic Ecosystems. Ambio, 39(2), 136–147.DOI: 10.1007/s13280-010-0022-7

Fowler, D., Smith, R., Muller, J., Cape, J.N., Sutton, M., Erisman, J.W., & Fagerli, H. (2007). Long term trends in sulphur and nitrogen deposition in Europe and the cause of non-linearities. Water, Air, & Soil Pollution: Focus, 7(1-3), 41-47.

Fulton, R.J. (1989). Quaternary Geology of Canada and Greenland. Geological Survey of Canada, Geology of Canada Series no. p1–839

Futter, M.N., Valinia, S., Löfgren, S., Köhler, S.J., & Fölster, J. (2014). Long-term trends in water chemistry of acid-sensitive Swedish lakes show slow recovery from historic acidification. Ambio, 43(1), 77-90.

Gunn, J.M., & Keller, W. (1990). Biological recovery of an acid lake after reductions in industrial emissions of sulphur. Nature, 345(6274): 431.

Henriksen A. (1979). A simple approach for identifying and measuring acidification of freshwater. Nature, 278(5704): 542–545. doi: 10.1038/278542a0.

Hesthagen, T. (1989). Episodic fish kills in an acidified salmon river in southwestern Norway. Fisheries, 14(3): 10-17

Hesthagen, T., Berger, H.M., Larsen, B.M., & Saksgård, R. (1995). Monitoring fish stocks in relation to acidification in Norwegian watersheds. Water, Air, & Soil Pollution, 85(2), 641-646.

Hesthagen, T., Sevaldrud, I.H., & Berger, H.M. (1999). Assessment of damage to fish populations in Norwegian lakes due to acidification. Ambio, 28(2): 112-117.

Hettelingh, J.P., Downing, R.J., de Smet, P.A., de Vries, W., Schopp, W., Chadwick, M.J., Kuylenstierna, J.C.I., & Gough, C.A. (1991). Mapping Critical Loads for Europe. CCE Technical Report no. 1. Rijksinstituut voor Volksgezondheid en Milieu RIVM.

Hettelingh, J.P., Posch, M., De Smet, P.A.M., & Downing, R.J. (1995). The use of critical loads in emission reduction agreements in Europe. Water, Air, and Soil Pollution, 85(4): 2381-2388.

Hole, L.R., Christensen, J.H., Ruoho-Airola, T., Tørseth, K., Ginzburg, V., & Glowacki, P. (2009). Past and future trends in concentrations of sulphur and nitrogen compounds in the Arctic. Atmospheric Environment, 43(4): 928-939.

Houle, D., Ouimet, R., Couture, S., & Gagnon, C. (2006). Base cation reservoirs in soil control the buffering capacity of lakes in forested catchments. Canadian Journal of Fisheries and Aquatic Sciences, 63(3): 471-474.

International Maritime Organization. (2009). Prevention of air pollution from ships. Sulphur monitoring for 2008. MEPC 59/4/1.

International Maritime Organization. (2018). Status of Treaties. http://www.imo.org/en/About/Conventions/StatusOfConventions/Documents/Status OfTreaties.pdf

International maritime Organization. (N.d). Sulphur oxides (SOx) and Particulate Matter (PM) – Regulation 14. Available Online: http://www.imo.org/en/OurWork/Environment/PollutionPrevention/AirPollution/Pag es/Sulphur-oxides-(SOx)-%E2%80%93-Regulation-14.aspx

Jensen, K., W., & Snekvik, E. (1972). Low pH-Levels Wipe Out Salmon and Trout Populations in Southernmost Norway. Ambio, 1(6): 223-225.

Johnson, J., Graf Pannatier, E., Carnicelli, S., Cecchini, G., Clarke, N., Cools, N., Hansen, K., Meesenburg, H., Nieminen, T.M., Pihl Karlsson, G., & Titeux, H. (2018). The response of soil solution chemistry in European forests to decreasing acid deposition. Global change biology.

Joynt III, E.H., & Wolfe, A.P. (2001). Paleoenvironmental inference models from sediment diatom assemblages in Baffin Island lakes (Nunavut, Canada) and reconstruction of summer water temperature. Canadian Journal of Fisheries and Aquatic Sciences, 58(6): 1222-1243.

Kirchner, J.W., Dillon, P.J., & LaZerte, B.D. (1992). Predicted response of stream chemistry to acid loading tested in Canadian catchments. Nature, 358(6386):.478.

Kirchner, J.W., & Lydersen, E. (1995). Base cation depletion and potential long-term acidification of Norwegian catchments. Environmental science & technology, 29(8): 1953-1960.

Kretser, W., Gallagher, J. and Nicolette, J. 1989. Adirondack Lakes Study. 1984-1987. An evaluation offish communities and water chemistry. Ray Brook (NY): Adirondacks Lakes Survey Corporation.

Lander, D.H., Overton, W.S., Linthurst, R.A., & Brakke, D.F. (1988). Eastern Lake Survey, Regional Estimates of Lake Chemistry. Environmental Science & Technology. 22(2), 128-135

Lawrence, G.B., David, M.B., Lovett, G.M., Murdoch, P.S., Burns, D.A., Stoddard, J.L., Baldigo, B.P., Porter, J.H., & Thompson, A.W. (1999). Soil calcium status and the response of stream chemistry to changing acidic deposition rates. Ecological Applications, 9(3): 1059-1072.

Lawrence, G.B., Hazlett, P.W., Fernandez, I.J., Ouimet, R., Bailey, S.W., Shortle, W.C., Smith, K.T., & Antidormi, M.R. (2015). Declining acidic deposition begins reversal of forest-soil acidification in the northeastern US and eastern Canada. Environmental science & technology, 49(22), 13103-13111.

Likens, G.E., & Bormann, F.H. (1974). Acid rain: a serious regional environmental problem. Science, 184(4142): 1176-1179.

Likens, G.E., Driscoll, C.T., & Buso, D.C. (1996). Long-term effects of acid rain: response and recovery of a forest ecosystem. Science, 272(5259): 244-246.

Löfgren, S., & Zetterberg, T. (2011). Decreased DOC concentrations in soil water in forested areas in southern Sweden during 1987–2008. Science of the Total Environment, 409(10): 1916-1926.

Mathews, W.H., & Bustin, R.M. (198)4. Why do the Smoking Hills smoke?. Canadian Journal of Earth Sciences, 21(7), 737-742.

Maxwell, J.B. (1981). Climatic regions of the Canadian Arctic islands. Arctic: 225–240.

McNulty, S.G., Cohen, E.C., Myers, J.A.M., Sullivan, T.J., & Li, H. (2007). Estimates of critical acid loads and exceedances for forest soils across the conterminous United States. Environmental Pollution, 149(3): 281-292.

McNeely, R.N., Dwyer, L., & Neimanis, V.P. (1979). Water quality sourcebook: a guide to water quality parameters. Environment Canada, Inland Waters Directorate, Water Quality Branch, Ottawa, pp.89

Michelutti, N., Blais, J.M., Mallory, M.L., Brash, J., Thienpont, J., Kimpe, L.E., Douglas, M.S., & Smol, J.P. (2010). Trophic position influences the efficacy of seabirds as

metal biovectors. Proceedings of the National Academy of Sciences, 107(23):10543– 10548.

Mills, K.H., & Schindler, D.W. (1986). Biological indicators of lake acidification. In Acidic Precipitation (pp. 779-789). Springer, Dordrecht.

Molot, L.A., & Dillon, P.J. (2008). Long-term trends in catchment export and lake concentrations of base cations in the Dorset study area, central Ontario. Canadian Journal of Fisheries and Aquatic Sciences, 65(5): 809-820.

Moran, M.D., Lupu, A., Zhang, Z., Savic-Jovicic, V., & Gravel, S. (2016). A comprehensive performance evaluation of the next generation of the Canadian operational regional air quality deterministic prediction system. Proc 35th International Technical Meeting on Air Pollution modelling and its Application. 3-7 Oct., Chania, Crete

Nilsson, J., & P. Grennfelt. (1988). Critical loads for sulphur and nitrogen: report from a workshop held at Stockloster, Sweden, 19-24 March 1988. Technical Report, Nordic Council of Ministers.

Odén, S. (1968). The acidification of air and precipitation and its consequences on the natural environment. State National Science Research Council

Økland, J., & Økland, K.A. (1986). The effects of acid deposition on benthic animals in lakes and streams. Experientia, 42(5): 471-486.

Oliver, J.G.J., Bouwman, A.F., Van der Hoek, K.W. and Berdowski, J.J.M., 1998. Global air emission inventories for anthropogenic sources of NOx, NH3 and N2O in 1990. In Nitrogen, the Confer-Ns, 135-148.

Paltan, H., Dash, J., & Edwards, M. (2015). A refined mapping of Arctic lakes using Landsat imagery. International Journal of Remote Sensing, 36(23): 5970-5982.

Pardo, L.H., Fenn, M.E., Goodale, C.L., Geiser, L.H., Driscoll, C.T., Allen, E.B., Baron, J.S., Bobbink, R., Bowman, W.D., Clark, C.M., & Emmett, B. (2011). Effects of nitrogen deposition and empirical nitrogen critical loads for ecoregions of the United States. Ecological Applications, 21(8): 3049-3082.

Perovich, D., Meier, W., Tschudi, M., Farrell, S., Hendricks, S., Gerland, S., Haas, C., Krumpen, T., Polashenski, C., Ricker, R., & Webster, M. (2017): Sea Ice [in NOAA Arctic Report Card 2017], https://www.Arctic.noaa.gov/Report-Card/

Pizzolato, L., Howell, S. E., Dawson, J., Laliberté, F., & Copland, L. (2016). The influence of declining sea ice on shipping activity in the Canadian Arctic. Geophysical Research Letters: 43(23).

Rautio, M., Dufresne, F., Laurion, I., Bonilla, S., Vincent, W.F., & Christoffersen, K.S., (2011). Shallow freshwater ecosystems of the circumpolar Arctic. Ecoscience, 18(3): 204–222.

Reuss, J.O., & Johnson, D.W. (1986). Acid Deposition and the Acidification of Soils and Waters. Ecological Studies. (59):119. DOI: 10.1002/iroh.19880730113

Rühland, K.M., & Smol, J.P. (1998). Limnological Characteristics of 70 Lakes Spanning Arctic Treeline from Coronation Gulf to Great Slave Lake in the Central Northwest Territories, Canada. Int. Rev. Hydrobiol.(141): 137–141.doi: 10.1002/iroh.19980830302

Rühland, K.M., Smol, J.P., Wang, X., & Muir, D.C.G. (2003). Limnological characteristics of 56 lakes in the Central Canadian Arctic Treeline Region. Journal of Limnology. 62: 9–27. doi:10.4081/jlimnol.2003.9.

Schindler, D.W., Mills, K.H., Malley, D.F., Findlay, D.L., Shearer, J.A., Davies, I.J., Turner, M.A., Linsey, G.A., & Cruikshank, D.R. (1985). Long-term ecosystem stress: the effects of years of experimental acidification on a small lake. Science, 228(4706): 1395-1401.

Schofield, C.L., & Trojnar, J.R. (1980). Aluminum toxicity to brook trout (Salvelinus fontinalis) in acidified waters. In Polluted rain (pp. 341-366). Springer, Boston, MA.

Skjelkvåle, B.L., Mannio, J., Wilander, A., & Andersen, T. (2001). Recovery from acidification of lakes in Finland, Norway and Sweden 1990? 1999. Hydrology and Earth System Sciences Discussions, 5(3), 327-338.

Skjelkvåle, B.L., Stoddard, J.L., Jeffries, D.S., Tørseth, K., Høgåsen, T., Bowman, J., Mannio, J., Monteith, D.T., Mosello, R., Rogora, M., & Rzychon, D. (2005). Regional scale evidence for improvements in surface water chemistry 1990– 2001. Environmental Pollution, 137(1), 165-176.

Smith, R. A. (1872). Air and rain: the beginnings of a chemical climatology. Longmans, Green, and Company. Retrieved from https://archive.org/details/airrainbeginning00smitiala.

Smith, S.J., Pitcher, H., & Wigley, T.M. (2001). Global and regional anthropogenic sulfur dioxide emissions. Global and planetary change, 29(1-2): 99-119.

Snucins, E., & Gunn, J. (1998). Chemical and biological status of Killarney Park lakes (1995–1997). A study of lakes in the early stages of recovery from acidification. Ontario Ministry of Natural Resources Cooperative Freshwater Ecology Unit.

Statistic Canada (2009). EnviroStat. 3(2). Statistics Canada, Ottawa, Canada. Catalogue no.16-002-X

Statistic Canada (2012). Canada Year Book 2012. Statistics Canada, Ottawa, Canada. Catalogue no. 11-402-X.

Stern, D.I. (2005). Global sulfur emissions from 1850 to 2000. Chemosphere, 58(2), 163- 175. ISSN 0068-8142

Stoddard, J.L., Jeffries, D.S., Lükewille, A., Clair, T.A., Dillon, P.J., Driscoll, C.T., Forsius, M., Johannessen, M., Kahl, J.S., Kellogg, J.H., & Kemp, A. (1999). Regional trends in aquatic recovery from acidification in North America and Europe. Nature, 401(6753), 575.

Tamm, C.O. (1976). Acid precipitation: biological effects in soil and on forest vegetation. Ambio, 235-238.

Tammi, J., Appelberg, M., Beier, U., Hesthagen, T., Lappalainen, A., & Rask, M. (2003). Fish status survey of Nordic lakes: effects of acidification, eutrophication and stocking activity on present fish species composition. AMBIO: A Journal of the Human Environment, 32(2) 98-105.

Wania, F., & Mackay, D. (1993). Global fractionation and cold condensation of low volatility organochlorine compounds in polar regions. Ambio,.10-18.

Watmough, S.A., Aherne, J., & Dillon, P.J. (2005a). Effect of declining lake base cation concentration on freshwater critical load calculations. Environmental science & technology, 39(9), 3255-3260.

Watmough, S.A., Aherne, J., Alewell, C., Arp, P., Bailey, S., Clair, T., Dillon, P., Duchesne, L., Eimers, C., Fernandez, I., & Foster, N. (2005b). Sulphate, nitrogen and base cation budgets at 21 forested catchments in Canada, the United States and Europe. Environmental Monitoring and Assessment, 109(1-3): 1-36.

Webster, K.E., Eilers, J.M., Wiener, J.G., Glass, G.E., & Garrison, P.J. (1993). Chemical and biotic characteristics of two low-alkalinity lakes in northern wisconsin: relation to atmospheric deposition. Technical bulletin no. 184 (No. PB-95-111878/XAB). Wisconsin Dept. of Natural Resources, Madison, WI (United States).

Wright, R.F., Dale, T., Gjessing, E.T., Hendrey, G.R., Henriksen, A., Johannessen, M., & Muniz, I.P. (1976). Impact of acid precipitation on freshwater ecosystems in Norway. Water, air, and soil pollution, 6(2-4): 483-499.

Wright, R.F., & Henriksen, A. (1978). Chemistry of small Norwegian lakes, with special reference to acid precipitation. Limnology and Oceanography, 23(3): 487-498.

2.0: Physical and chemical characteristics of 1300 lakes and ponds across the Canadian Arctic

2.1 Abstract

There are large spatial and temporal gaps in baseline water chemistry for the Canadian

Arctic. Nonetheless, lakes and ponds are a major feature of the Arctic landscape and are recognized as effective ‘sentinels of change’, as they integrate processes at the catchment and landscape scale. We attempt to determine baseline water chemistry characteristic of

Arctic lakes and ponds (n = 1300) across the Canadian Arctic (and within each region of the Canadian Arctic, i.e., mainland territories and the Queen Elizabeth Islands), and to determine key drivers of water chemistry for Arctic lakes and ponds. Water chemistry data in the Canadian Arctic (1200 sites with 26 variables) were collected from published studies and supplemented with data from recent surveys on Baffin Is. (n = 80), Coats Is.

(n = 10), Prince Charles Is. (n = 4), and eastern Northwest Territories (n = 6). In general, most sites were shallow (85.4%, < 10 m), located at low elevation (66.5%, < 200 m.a.s.l), close to coastlines (72.5%, 0–50 km), and underlain by sedimentary geology (66.5%).

The first two components from Principal Component Analysis (based on 613 sites with

17 variables) explained 49.3% of the variation in the dataset; the first component was dominated by conductivity/carbonate materials, and the second component dominated by weathering of non-carbonate soils. In general, bedrock geology is the primary driver of water chemistry characteristics, with major differences between igneous and sedimentary rocks. Those on a sedimentary base tend to have high pH, nutrients (DOC, TP, and TN)

and higher inorganic solutes (both cations and anions) concentrations. Sea-salt aerosols drive Na and Cl concentrations in lakes and ponds, while vegetation litter and surface soils drive DOC (most carbon existing as DOC than POC).

Key Words: Limnology, Arctic lakes, water chemistry, nutrients, metals, geology

2.2 Introduction Lentic systems, i.e., freshwater lakes and ponds, are a major feature of the Arctic landscape. It is estimated, that there are approximately 3.5 million lakes and ponds within the Arctic circle (≥ 66.6 ⁰N), with > 40% located within the Canadian Arctic (Paltan et al., 2015). The vast majority of these aquatic systems are generally small (< 10 ha) and shallow (< 12 m in depth) ponds (Hamilton et al., 2001; Rautio et al., 2011; Paltan et al.,

2015), but can also be large and deep systems (Great Bear Lake in Northwest Territories at 114,717 km2 and a max depth 446 m; Vincent et al., 2012). In some regions lakes and ponds can cover up to 90% of the surface land area (Pienitz et al., 2008). Arctic lakes and ponds are exposed to harsh climatic conditions, i.e., low temperatures, reduced precipitation, and seasonally low inputs of solar radiation, which limit the development of vegetation and the chemical weathering of soils within the catchment. Most precipitation occur in the form of snow or ice (Maxwell, 1981) and during the melting period bring large amounts of water and other components (particulates and dissolved compounds) into these (often isolated) systems, which results in dilute systems that are further modified by terrestrial processes. This runoff-dominated process, results in lakes and ponds with hydrochemical characteristics that are unique to the Arctic region (Hamilton et al., 2001; Wetzel, 2001; Pienitz et al., 2004). These systems are vital habitat for many biological communities. In addition, they provide vital resources (hunting, fishing, and drinking water) for indigenous groups. Moreover, it is well established that aquatic systems such as lakes and ponds are effective indicators or ‘sentinels of change’, as they reflect process changes at the catchment scale, and can provide critical spatial and temporal information on the impacts of anthropogenic activity (Adrian et al. 2009).

Air temperature in the Arctic has increased twice as fast as the rest of the world

(Overland et al., 2017) and is expected to greatly impact the physical, chemical, and biological characteristics of Arctic lentic ecosystems. Observations of sea-ice coverage have reported a loss of 50% since 1979 (Meier et al., 2006), a decrease in thickness from

3.1 to 1.8 m (Rothrock et al., 1999) and loss of multi-year ice (Polyakov et al., 2012).

These changes present opportunities for anthropogenic activities such as natural resource extraction (Haley et al., 2011) and shipping (Pizzolato et al. 2016; Dawson and Carter,

2017) and may lead to elevated inputs of pollutants in the Arctic (Law and Stohl, 2007;

Corbett et al., 2010; Peters et al., 201l). It is well established that sulphur emissions have altered (acidified) the surface water chemistry of lakes in Europe and North America

(Neary and Dillon, 1988; Moiseenko, 1994) and degraded air quality in urban areas

(Eckhardt et al., 2013; Aliabadi, et al., 2015). In the Arctic, there is concern that increased emissions of acidifying compounds may acidify freshwater systems, especially those located near major shipping routes, i.e., the Northwest Passage and Hudson Strait, and those on acid-sensitive geology, i.e., Baffin Is. Across the Arctic region, including Alaska

(Osterkamp and Romanovsky, 1999), Canada (Payette et al., 2004; Smith et al., 2005), and Russia (Romanovsky et al., 2010), thawing of permafrost (from higher air temperatures) has led to both the expansion (through the creation of depressions) and the disappearance (through draining) of thermokart lakes (Plug et al., 2008). The change in permafrost has also changed the hydrochemical characteristics of Arctic lakes and ponds

(Prowse et al., 2006; Walvoord and Strieg, 2007;Thienpont et al., 2013; Roberts et al.,

2017). Changes in the water chemistry have both positively (Pienitz et al., 2004;

Michelutti et al., 2005, 2007; Thienpont et al., 2013) and negatively (Reist et al., 2006;

Robert et al., 2017) impacted biological communities.

Although there have been many limnological studies in the Canadian Arctic

(Pienitz et al., 1997a, b; Rühland and Smol, 1998; Hamilton et al., 2001; Michelutti et al.,

2002; Lim and Douglas, 2003; Michelutti et al., 2002a, b; Mallory et al., 2006; Westover et al., 2009; Côté et al., 2010; Stewart and Lamoureux, 2011; Medeiros et al., 2012;

Roberts et al., 2017), there are still large (spatial and temporal) data gaps, with few studies integrating existing observations to provide baseline limnological data and even fewer (Roberts et al., 2017) providing long-term monitoring data.

The objective of this study was to assess the physical and chemical characteristics of lakes and ponds across the Canadian Arctic, to provide baseline hydrochemical data

(for 1300 sites) and to evaluate the drivers of water chemistry. This was carried out by collating published hydrochemical data for 1200 sites from 32 peer-reviewed articles and supplementing these data with new hydrochemical observations for 100 lakes and ponds on Baffin, Coats, and Prince Charles Islands, and the Northwest Territories.

2.3 Methods

2.3.1 Study area

In this study, the Canadian Arctic was defined as all Canadian territory within the boundary of the Arctic Monitoring and Assessment Program (Stonehouse, 1989). This region is approximately 4.0 ×106 km2 (AMAP, 1998) covering all areas north of 60 ⁰N.

This includes the Canadian Arctic Archipelago, the territory of Yukon, Northwest

Territories and Nunavut, and parts of northern Quebec and Labrador (Figure 2.1 Top). 30

Much of the Canadian Arctic Archipelago region rests upon the Arctic Platform, which consists of sedimentary geology comprised of shale, siltstone, sandstone, limestone, and dolomite (Fulton, 1989; Trettin, 1991; Figure 2.2 Top). The eastern perimeter (eastern

Ellesmere, eastern Devon, Baffin Is., eastern Northwest Territories, Nunavut, northern

Quebec, and northern Labrador) rests upon Precambrian (Canadian) Shield, which consist of crystalline and metamorphic rock, that include greenstone, gabbro, gneisses, granitic, and volcanic rocks (Fulton, 1989; Trettin, 1991; Harrison et al., 2011; Figure 2.2 Top). In general, soils are poorly developed and greatly influenced by cryogenic processes (the formation of ice in soils) such as the freeze-thaw, ice build-up, thermal cracking, and frost heave, which leads to broken soil horizons (Tarnocai, 2003). Chemical properties of soils are greatly influence by the parent material. Soils nutrients (N, P, K) are generally locked up by surface organics (Tarnocai, 2003). Peat bogs and organic soils are common among depressions in the southern Arctic landscape and are the results of many years of plant growth. The study area encompasses four level I ecoregions: Arctic Cordillera, Tundra,

Taiga, and Northwestern Forested mountains (Figure 2.2 Bottom). The Arctic Cordillera

(AC) consist of the mountainous regions of the eastern Arctic, while the tundra (TU) covers most of the Canadian Arctic Archipelago (Commission for Environmental

Cooperation, 1997). The Taiga (TA) ecoregion is south of the TU and encompasses the tree line, while the Northwestern Forested Mountains (NWF) mostly lies in the southwestern portion of the study area (Commission for Environmental Cooperation,

1997). Low temperatures and precipitation, extended long winters, short summers, and extreme seasonal light exposure (24 hrs of darkness in the winter and light during the summer) are common climate characteristics of the Arctic. Climatic conditions vary

among regions (Maxwell, 1981), with colder temperature occurring in the north (–28 to –

35⁰C in January and 0–3⁰C in July), compared with the south (–20 to –25⁰C in the winter and 5 to 8⁰C in July). Similarly, precipitation ranges from <100 mm in the north, to 200–

500 mm annually in the south, with much of the precipitation (20–50%) falling as snow or ice (Maxwell 1981).

Figure 2.1 Names of islands and regions within the study region with the AMAP boundary depicted in red (Top). Location of all sites (sites from published sources are depicted as red dots, while sites collected from this study are depicted as yellow dots) within the study region (Bottom).

Figure 2.2 Location of all sites (published sources = red dots, this study = yellow dots) among the different geology (Top) and ecoregions (Bottom) types.

2.3.2 Arctic water chemistry dataset

In general, site selection in most studies in the Canadian Arctic is limited by logistical and weather conditions. During collation it was further noted that individual datasets did not have similar universal unit systems, i.e., coordinate systems (Decimal degrees vs. Degree minute seconds), concentration (µg·L–1 vs mg·L–1), or a consistent suite of chemical parameters. The initial dataset consisted of >1500 water bodies and

>110 physical and chemical variables.

Published articles (n = 27), reports (n = 2), and graduate theses (n = 3) containing water chemistry observations for Arctic lakes and ponds (n = 1200) were compiled into one dataset (Table 2.1). In general, sites were primarily sampled during the ice-free season of July and August; details on sampling and analytical methods can be found in their corresponding papers (Table 2.1). In general, most studies followed methods as outlined in Environment Canada (1994a and b). In addition, recent water chemistry data

(2015 and 2016) for lakes and ponds in Baffin Is. (n = 80), Coats Is. (n = 10), Prince

Charles Is. (n = 4), and eastern Northwest Territories (n = 6) were included (Figure 2.1

Bottom), bringing the total number of water bodies sampled to 1300 (Table 2.1).

The ionic balance check was used to assess the quality of the water chemistry dataset following the ICP Waters Programme Centre (2010). The ion balance check compares the sum of cations to the sum of anions (in µmolc·L-1), with differences

(between cations and anions) of ≤10% assumed to be acceptable (ICP Waters Programme

Centre, 2010).

Ion balance = [sum of cations: ∑([Ca] + [Mg] + [K] + [Na] + [H]) – sum of anions:

∑([Cl] + [SO4] + [NO3] + [ALK]) *100] /sum of cations

The concentration of protons was calculated from pH, and ALK was estimated from DIC (Supplementary information Figure A1). Only 224 sites had observations of

Ca, Mg, Na, K, ALK, Cl, NO3, SO4, and pH, to perform the ionic balance check

(Supplementary information Figure A2).

Table 2.1 Summary of published sources with their respectable site count.

No. Source Site Count 1 Antoniades et al., 2003a 66 2 Antoniades et al., 2003b 25 3 Babaluk et al., 1999 8 4 Babaluk et al., 2009 14 5 Bouchard et al., 2004 48 6 Brimble et al., 2009 26 7 Bunbury and Gajewski, 2009 9 8 Bunbury and Gajewski, 2005 33 9 Côte et al., 2010 27 10 Delvin MSc 2010 20 11 Hadley et al., 2013 40 12 Hadley MSc 2007 6 13 Hamilton et al., 2010 181 14 Keatley et al., 2007 55 15 Keatley Ph.D 2007 46 16 Lim and Douglas, 2003 23 17 Lim et al., 2005 45 18 Lim et al., 2001 9 19 Mallory et al., 2006 32 20 Medeiros et al.., 2012 93 21 Michelutti et al., 2002 34 22 Michelutti et al., 2002 38 23 Michelutti et al., 2007 33 24 Michelutti et al., 2010 2 25 Moser et al., 1993 8 26 Pienitz et al., 1997a 59 27 Pienitz et al., 1997b 24 36

28 Ruhland and Smol., 1998 70 29 Ruhland et al., 003 56 30 Stewart and Lamoureux., 2011 2 31 Westover et al., 2009 61 32 Wilson and Gajewski, 2002 7 Sub Total 1200 This Study 100 Grand Total 1300

Physical (location, elevation, depth, area) and chemical data for each site were extracted and unified into a common database. The most recent sampling period was chosen to represent sites where multiple observation was available. Missing values for elevation (m) were determined using Google Maps’ Elevation Application Programming

Interface. Distance to coast (km) was calculated using the Grass GIS 7 plugin in QGIS and a coastal shapefile of Canada. Geological data was obtained from Harrison et al.

(2011). The initial database had more than 90 physical and chemical variables.

Observations below detection, primarily trace element concentrations, were assigned a random number between zero and the associated detection limit. When values for total kjeldahl nitrogen (TKN), nitrates (NO3), nitrites (NO2) and ammonium (NH4) were present, TN concentration was calculated by the sum of TKN (Organic Nitrogen +

Ammonia), NO3, and NO2 values. Lakes (n > 650) with a common set of key variables (n

= 26) were selected for statistical analysis: Latitude, Longitude, Elevation, Distance to

Coats, Area, pH, Conductivity, Ca, K, Mg, Na, Cl, SO4, SiO2, DOC, POC, DIC, NH3,

TKN, TN, TP, Al, Ba, Fe, Mn, and Sr.

2.3.3 Statistics

All statistical analysis was preformed using R (Version 3.3.2). Variables were tested for normality (lilliefor test from the package nortest), homogeneity of variance

(Levene’s test from the package car), and linearity (quantile-quantile plots from the package stats) prior to statistical analysis. However, due to the consistent non-normal distribution among variables (found by testing for normality, homogeneity of variance, and linearity), non-parametric testing was used. To determine correlations between physical and chemical variables, the spearman’s rank correlation (rs) from the package

Hmisc was used. To determine if there are statistical difference concentrations between regions (geological type, ecoregions, and geographical region), the Kruskal–Wallis rank sum test (from the package stats) was used with a Dunn’s post hoc test (Bonferroni adjustment; from the package dunn.test). Physical and chemical data were summarised

(for baseline concentrations) across the entire dataset and by region, geology, and ecoregion type. Principal Component Analysis (PCA) was performed with transformed variables (log) in R, with the packages ggbiplot and ggfortify to determine possible relationships, potential drivers of hydrochemical variables, and the variability within the dataset. For the PCA, a subset of chemical variables (Latitude, Longitude, Elev, DistC,

Area, pH, Cond, Ca, K, Mg, Na, SO4, Cl, DOC, DIC, TN, TP, Al, Fe, and Mn) were used.

2.4 Results

A subset of 1300 sites with 26 physical and chemical parameters was used in this study (Table 2.2), other parameters (n = 59) i.e., Nitrite, Total Dissolved Nitrogen,

Arsenic, Nitrogen Isotope 15 can be found in the supplementary information for this study (Supplementary information Table A1). The quality of the water chemistry database was checked by using the ion balance check (Supplementary information Figure A2). The 38

ion balance check indicated a mean cation deficit of 8.4% (median = 3.8%) for the 224 sites and is most likely due to laboratory error. However, only nine sites had an ion difference >10%, with four sites >15%.

Table 2.2 Descriptive statistics of 26 parameters including; units, count, mean, standard deviation (SD), minimum, maximum, and percentile (5th and 9th) values.

Parameter Symbol Units Count Mean SD Latitude Decimal Degree 1300 71.1 6.71 Longitude Decimal Degree 1300 -96.8 19.6 Elevation Elev m.a.s.l 1299 198 207.2 Distance to Coast DistC km 1300 69.4 128.2 Area Area ha 877 1088 18453.2 pH pH 1253 6.01 4.82 Conductivity Cond µS·cm–1 1235 186 487.9 Calcium Ca mg·L–1 1253 20.5 29.5 Potassium K mg·L–1 1208 1.61 4.74 Magnesium Mg mg·L–1 1144 8.24 17.5 Sodium Na mg·L–1 1255 12.3 73.2 Chloride Cl mg·L–1 1251 16.2 103.7 –1 Sulphate SO4 mg·L 1251 28.7 121 –1 Silica SiO2 mg·L 984 1.04 1.55 Dissolve Organic Carbon DOC mg·L–1 1135 6.44 13.9 Dissolve Inorganic Carbon DIC mg·L–1 1032 13.8 12.8 Particulate Organic Carbon POC mg·L–1 702 0.56 0.73 –1 Ammonia NH3* µg·L 744 25.1 40.0 Total Kjeldahl Nitrogen TKN* µg·L–1 802 376 380 Total Nitrogen TN* µg·L–1 864 424 426 Total Phosphorous TP* µg·L–1 1247 11.14 31.1 Aluminum Al µg·L–1 872 113 588 Barium Ba µg·L–1 814 14.3 23.7 Iron Fe µg·L–1 1015 199 645 Manganese Mn µg·L–1 875 529 3301 Strontium Sr µg·L–1 819 72 193

Table 2.2 continued

Percentile Symbol Min Max Median CV 5th 95th Lat 60 83.1 72.2 9.44 61.3 81.84 Long -140 -61.5 -95.7 -20.3 -133 -68.4 Elev 0 1387 134 104 7.9 657 DistC 0.01 750 13.5 184 0.56 395 Area 0 506300 6.38 1695 0.03 553 pH 10.9 3.4 7.9 -119 8.7 6.31 Cond 1.46 13200 97.4 263 9.61 554 Ca 0 451 14.6 144 0.67 52.4 K 0 109 0.57 295 0.11 5.97 Mg 0.01 273 3.7 212 0.32 28.2 Na 0.01 1650 1.65 594 0.3 35.1 Cl 0 2850 2.05 642 0.3 44.5

SO4 0.03 2100 3.1 422 0.4 117

SiO2 0 13.9 0.51 150 0.07 3.7 DOC 0.02 332 3.5 216 0.70 18.5 DIC 0.06 134 10.9 92.5 0.70 36.5 POC 0.01 9.89 0.41 130 0.12 1.34

NH3* 0.06 459 12 159 2.00 83.9 TKN* 0.06 2760 263 101 49.05 1100 TN* 7.40 5324 312 100 71.8 1068 TP* 0 761 7.05 280 1.35 28.1 Al 0.02 11200 17 521 2.76 328 Ba 0.15 272 6.45 166 1.00 54 Fe 0.03 11500 48 325 3.00 765 Mn 0 52600 5.3 624 0.46 2490 Sr 0.24 3150 23.9 268 3.09 245

Result presented as parentages (in regard to the chemical parameter) in this study is in reference to the total observed count for the parameters (which is shown in Table 40

2.2) as not all 1300 sites have observations for all 26 chemical and physical parameters.

Only percentages shown for Location, Elevation, Distance to Coast, Geology, Ecoregion, and region of the Canadian Arctic are in reference to the 1300 sites. For example, 14.5%

(n = 127) of sites have an area >100 ha, this means that for all sites with observations of area (n = 877; Table 2.2), 14.5% (n = 127) have area >100 ha. While 82.0% (n = 1028) of sites have a pH >7 (1028 of total sites with observations for pH = 1253; Table 2.2).

A large portion of sites were sampled in the years 1993 (n = 185), 2003 (n = 101), and 2016 (n = 81; this study) (Table 2.3). However, some datasets provided a range of years for their sampling date, i.e., 1989–2002 (Bouchard et al. 2004), 2006–2010

(Medeiros et al. 2012), which render some unknowns as to when the site was actually sampled. This accounted for a large number (n = 176) of sites. Hamilton et al. (2001) reported the earliest site observation in 1979 on Ellesmere Is., whereas the most recent observation was reported under this study on Baffin Is. in 2016 (Table 2.3).

Table 2.3 Summary of sampling sites per region with sampling year.

Region Site count Years sampled Axel Heiberg Is. 47 1995/1996/1998 Baffin Is. 132 1980/1984/1985/1993/2015/2016 Banks Is. 45 2000 Bathurst Is. 67 1992/1994/1997/1998/1999/2000/2001/2002/2005 Bylot Is. 47 2005/2008 Coats Is. 10 2016 Cornwallis Is. 47 1980/1992/1993 Crozier Is. 2 2008 Devon Is. 66 1980/1994/1996/2000/2004/2005/2006/2006/2007 Ellef Ringnes Is. 25 1996 Ellesmere Is. 170 1979/1989/1990/1992/1995/1996/1997/1998/1999 /2001/2003/2007/2008 King William Is. 4 1982 Little Cornwallis Is. 1 1981

Melville Is. 49 1992/2003/2004 Northwest 1990/1991/1993/2015 Territories 153 1982/1983/1991/1999/2004/2006/2007/2008/2009/ Nunavut 190 2010 Prince Charles Is. 5 1985/2016 Prince of Wales Is. 5 1994/1995 Prince Patrick Is. 35 1999 Somerset Is. 13 1980/1989/1990/1991/1993/1994/1995/1996 Southampton Is. 37 1983/2001/2002 Victoria Is. 88 1982/1997/2000/2004 Yukon 64 1990/1996/2000/2002 Grand Total 1300 Range: 1979–2016

2.4.1 Physical characteristics

Most sites were in Nunavut (n = 190), Ellesmere Is. (n = 170), and Northwest

Territories (n = 153) (Figure 2.1 Bottom; Table 2.3). Islands with the least number of sites were Little Cornwallis Is. (n = 1; Hamilton et al. 2001), Prince Charles Is. (n = 1;

Hamilton et al., 2001; without the addition of sites from this study), and Crozier Is. (n =

2; Michelutti et al. 2010) (Figure 2.1 Bottom; Table 2.3). Approximately 66.5% (n = 864) of all sites were on sedimentary geology, 26.4% (n = 343) were on igneous geology, and

6.2% (n = 80) on supracrustal geology (Figure 2.2 Top; Table 2.4). Thirteen sites were on unclassified geology (1.0%; Figure 2.2; Table 2.4) and were found on Somerset (n = 4),

Bathurst Is. (n = 1), Yukon (n = 4), Devon Is. (n = 1), and Ellesmere Is. (n = 3).

Unclassified geology consists primarily of metamorphic rock of granite gneiss, tonalite gneiss, granodiorite, paragneiss lithology (Harrison et al., 2011). Within ecoregion type, most sites (84.8%, n = 1102) were located within the Tundra (TU), which covers most of the Arctic Archipelago, mainland Nunavut, northern section of Northwest Territories and

Quebec (Figure 2.2 Bottom; Table 2.5). This was followed by the Taiga (TA) at 9.1% (n 42

= 118), Northwestern Forested Mountains (NWF) at 4.3% (n = 56), and lastly, the Arctic

Cordillera (AC) ecoregion at 1.8% (n = 24) (Figure 2.2 Bottom; Table 2.5).

Table 2.4 Median (mean) values and site count for selected water chemistry variables and ratios for each geology type. Geology Parameter Units Igneous Sedimentary Supracrustal Unclassified Count 343 864 80 13 Elev m.a.s.l 178 (230) 117 (163) 365 (406) 240 (383) DistC km 19.3 (102) 10.5 (54.5) 101 (90.0) 7 (72.7) Area ha 10.3 (335) 4.25 (1508) 30.9 (491) 4.75 (9.11) pH 7.31 (7.32) 8.06 (7.91) 8.08 (7.74) 8.06 (7.84) Cond µS·cm-1 7.41 (6.85) 8.06 (5.86) 8.31 (6.61) 8.06 (7.16) Ca mg·L-1 40.0 (66.8) 132 (232) 98.2 (181) 87 (248) K mg·L-1 4.11 (7.72) 20.8 (26) 13.4 (17.3) 16 (20.2) Mg mg·L-1 0.4 (0.69) 0.6 (1.95) 1.17 (1.92) 0.6 (2.3) Na mg·L-1 0.93 (1.99) 5.4 (10.3) 3.13 (11.5) 5.9 (21.5) Cl mg·L-1 0.63 (3.34) 2.22 (16.7) 2.77 (5.87) 2.2 (3.02) -1 SO4 mg·L 0.9 (6.08) 2.93 (21.4) 1.52 (6.75) 1.37 (2.54) -1 SiO2 mg·L 1.99 (5.29) 3.9 (39.2) 3.9 (12.0) 2.9 (55.2) DOC mg·L-1 0.4 (0.7) 0.58 (1.15) 0.48 (1.1) 0.32 (0.55) POC mg·L-1 2.9 (5.6) 3.8 (6.83) 3.9 (6.04) 2.2 (3.12) DIC mg·L-1 3.7 (5.27) 15.9 (17.1) 5.95 (10.1) 5 (7.17) -1 NH3* µg·L 18.7 (33.1) 11 (20.8) 23 (29.8) 11 (13.4) TKN* µg·L-1 176 (320) 281 (386) 290 (445) 230 (211) TN* µg·L-1 211 (289) 376 (486) 323 (331) 140 (190) TP* µg·L-1 6.00 (8.11) 7.4 (12.5) 7.7 (10.4) 7.71 (7.32) Al µg·L-1 12.8 (30.8) 17.7 (150) 29 (34.6) 16 (28.8) Ba µg·L-1 4 (5.96) 7.89 (16.6) 18.7 (27.4) 3.3 (14.4) Fe µg·L-1 47.7 (147) 49.4 (226) 31.2 (81.5) 47 (44.1) Mn µg·L-1 6.57 (523) 4.6 (575) 7.35 (15.2) 8.95 (113) Sr µg·L-1 14.5 (22.2) 30.1 (86.7) 116 (125) 28 (121) TN:TP Ratio 57.4 (79.84) 43.8 (54.2) 49.2 (57.5) 35.1 (44.7) Na:Cl Ratio 0.62 (1.16) 0.64 (1.27) 0.67 (1.16) 1.17 (12.1) Na:K Ratio 0.36 (1.21) 0.39 (2.82) 0.48 (0.99) 0.46 (0.56)

Table 2.5 Median (mean) values and site count for selected water chemistry variables and ratios for each ecoregion; Arctic Cordillera (AC), Northwestern Forested Mountains (NWF), Taiga (TA), and Tundra (TU). Ecoregion Parameter Units AC NWF TA TU Count 24 56 118 1102 43

Elev m.a.s.l 153 (231) 782 (774) 306 (338) 118 (153) DistC km 6.43 (13.4) 138 (190) 401 (374) 8.29 (31.9) Area ha 7.7 (139) 14.7 (46.2) 13.5 (103) 4.91 (1290) pH 7.66 (7.38) 8.37 (8.36) 7.5 (7.55) 7.95 (7.74) Cond µS·cm-1 7.66 (6.72) 8.36 (8.19) 7.5 (7.18) 7.95 (5.94) Ca mg·L-1 33.8 (201) 330 (373) 39.1 (90.7) 100 (186) K mg·L-1 2.7 (20.2) 30.6 (35.0) 5.25 (13.3) 15.3 (20.6) Mg mg·L-1 0.36 (2.38) 3.02 (4.15) 0.7 (1.31) 0.5 (1.48) Na mg·L-1 0.7 (6.89) 18.5 (28.5) 1.6 (4.77) 3.9 (7.75) Cl mg·L-1 1.36 (6.18) 4.6 (10.1) 0.92 (2.56) 1.65 (13.7) -1 SO4 mg·L 1.17 (7.24) 1.48 (4.27) 0.80 (2.44) 2.38 (18.5) -1 SiO2 mg·L 2.28 (71.5) 24.2 (74.4) 3.00 (12.7) 3.00 (27.1) DOC mg·L-1 0.52 (1.02) 4.03 (4.68) 0.40 (0.92) 0.52 (0.98) POC mg·L-1 1.5 (2.95) 10.6 (12.7) 12.9 (25) 3.15 (4.58) DIC mg·L-1 3.6 (10.6) 32.0 (34.9) 3.95 (9.34) 11.8 (13.9) -1 NH3* µg·L 13 (19.7) 6.00 (11.6) 10.0 (13.6) 13.0 (26.6) TKN* µg·L-1 120 (294) 540 (671) 423 (609) 230 (329) TN* µg·L-1 75.0 (109) 613 (729) 437 (612) 287 (403) TP* µg·L-1 6.00 (8.89) 10.8 (13.1) 7.85 (10.3) 6.85 (11.2) Al µg·L-1 16.2 (34.4) 30.5 (30.1) 30.0 (51.0) 13.8 (122) Ba µg·L-1 1.46 (5) 32.0 (30.2) 5.00 (16.6) 6.37 (13.5) Fe µg·L-1 100 (428) 32.5 (72.5) 41.0 (179) 49.0 (204) Mn µg·L-1 8.34 (12.8) 12.9 (19.9) 61.0 (4849) 4.43 (94.1) Sr µg·L-1 3.68 (10.3) 174 (195) 19.0 (106) 23.0 (64.7) TN:TP Ratio 53.5 (64.5) 37.6 (46.1) 54.4 (66.0) 46.6 (60.6) Na:Cl Ratio 0.53 (6.66) 0.58 (1.09) 0.69 (1.19) 0.64 (1.25) Na:K Ratio 0.63 (1.35) 1.17 (1.79) 0.64 (2.48) 0.25 (2.25)

Most sites sampled were situated at low elevations (66.5%, n = 865, <200 m.a.s.l;

Figure 2.3 Top) and close to the coast (72.5%, n = 938, 0–50 km; Figure 2.3 Middle), indicative of site accessibility rather than the geographic distribution of lakes and ponds in the Arctic. Using the available data for depth (only 680 sites with this observation) and area (only 877 observation), 85.4% of sites (n = 581; Figure 2.3 Bottom) are classified as shallow (≤10 m; Hamilton et al. 2001), and 58.0% (n = 509) small (≤10 ha; Hamilton et al. 2001), only 31.6% (n = 215) can be further classified as ponds (≤ 2 m; Lim et al.

2001). An interesting significant (p < 0.01) weak correlation between lake depth and Mg

(rs = –0.31), Cond (rs = –0.31), Ca (rs = –0.25), Cl (rs = –0.42), Fe (rs = –0.59), NH3 (rs = –

0.35), TKN (rs = –0.38), TN (rs = –0.39), POC (rs = –0.35), and TP (rs = –0.39) (Figure

2.4 ) suggest that deep sites tend to have less concentrations of majors ions and nutrients, a dilution effect.

Figure 2.3 Frequency of sites among different Elevation (Top), Distance to coast

(Middle), and Depth (Bottom).

Figure 2.4 Correlation matrix of selected chemical variables with Spearman’s rank correlation coefficient. Significant (p < 0.01) correlations coefficient are highlighted in colour.

2.4.2 Water pH and conductivity

Water pH ranged from 3.4 to 10.9, however, median pH for all sites was slightly alkaline (pH = 7.9, mean = 6.0, n = 1253) most sites (82.0%, n = 1028) had pH > 7, 47

while the rest (18%; n = 225) had pH ≤ 7 (Table 2.2). Regions with the lowest median pH were: Bylot Is. = 6.5, Ellef Ringnes Is. = 6.8, and Baffin Is. = 6.9, while regions with most alkaline medians were: Cornwallis Is. = 8.6, Prince of Wales Is 8.4, and Yukon =

8.4 (Table 2.6). Median surface water pH among geology types were: Igneous = 7.3,

Sedimentary = 8.1, Supracrustal = 8.1, and Unclassified = 8.1 (Table 2.4).

A significant positive correlation (p < 0.01) was found between pH and Ca (rs =

0.61), Mg (rs = 0.59), and DIC (rs = 0.64) (Figure 2.4), which may suggest that carbonate content (either underneath or in the catchment) influence pH levels in the water. Among geology, pH was significantly (p < 0.001) different between Igneous with Sedimentary and Supracrustal geology (Figure 2.5 Left middle; Table 2.7) and among ecoregions;

NWF–AC, NWF–TA, NWF– TU, TA–TU (Table 2.8) were significantly different.

Figure 2.5. Boxplot of conductivity (Left top), pH (Left middle), and DIC (Left bottom) among geology types, and DOC (Right top), K (Right middle), and TN (Right bottom) among ecoregions. Significant differences (Kruskal–Wallis rank sum test and Dunn’s post hoc test with Bonferroni adjustment) between mean concentrations are indicated by * = p values <0.05, ** = p- value <0.01, and *** = p-value < 0.001.

Table 2.6 Median (mean) values and site count for selected water chemistry variables and ratios for each region.

m.a.s.l km ha µS·cm-1 Region Count Elev DistC Area pH Cond Axel Heiberg Is. 47 171 (214) 5.94 (7.37) 15.5 (20.8) 8.05 (5.03) 128 (402) Baffin Is. 132 139 (153) 5.62 (12.1) 4.85 (6632) 6.89 (5.23) 50.8 (88.1) Banks Is. 45 128 (136) 29.6 (34.1) 1.00 (16.8) 8.10 (7.90) 109 (182) Bathurst Is. 67 61 (94.3) 5.27 (7.72) 19.6 (34.0) 8.20 (7.85) 131 (127) Bylot Is. 47 179 (237) 4.91 (7.22) 5.86 (25.7) 6.50 (6.50) 15.9 (34.8) Coats Is. 10 41.5 (41.3) 7.81 (7.81) 103 (111) Cornwallis Is. 47 64 (80.6) 2.00 (4.79) 18.1 (32.9) 8.60 (8.24) 127 (451) Devon Is. 66 81 (89.5) 4.76 (7.87) 0.72 (19.1) 8.28 (8.00) 99.5 (121) Ellef Ringnes Is. 25 43 (38.4) 1.51 (1.41) 0.06 (1.32) 6.80 (6.21) 228 (405) Ellesmere Is. 170 170 (224) 11.9 (17.8) 1.84 (810) 8.20 (7.65) 216 (356) Melville Is. 49 76.2 (125) 3.82 (4.98) 2.06 (15.0) 8.00 (7.81) 56.0 (149) Northwest Territories 153 311 (289) 312 (315) 18.9 (81.1) 7.5 (7.02) 48.0 (91.2) Nunavut 190 111 (180) 79.1 (86.3) 15.0 (1567) 7.55 (6.50) 51.0 (75.7) Prince Charles Is. 5 10 (8.64) 2.48 (2.58) 2500 (2500) 7.98 (7.98) 76.2 (98.6) Prince of Wales Is. 5 122 (110) 19.0 (19.4) 0.90 (172) 8.40 (8.18) 65.0 (114) Prince Patrick Is. 35 8.00 (21.4) 1.17 (1.20) 0.05 (1.43) 7.80 (7.67) 75.0 (115) Somerset Is. 13 170 (156) 6.54 (9.45) 11.0 (12.36) 7.90 (7.81) 60.0 (81) Southampton Is. 37 40.0 (65.1) 6.59 (13.6) 16.0 (259) 7.96 (7.83) 222 (251) Victoria Is. 88 139 (159) 22.6 (37.5) 11.1 (32.4) 7.80 (7.74) 140 (156) Yukon 62 787 (794) 152 (211) 14.8 (45.1) 8.37 (7.97) 305 (343) Note: Na:K ratios are sea-salt corrected.

Table 2.6 continued

-1 -1 -1 -1 -1 -1 mg·L mg·L mg·L mg·L mg·L mg·L Region Ca K Mg Na Cl SO4 Axel Heiberg Is. 17.8 (32.1) 1.10 (4.22) 4.70 (14.0) 3.40 (53.6) 2.16 (80.6) 9.30 (62.9) Baffin Is. 5.75 (8.64) 0.18 (0.67) 0.83 (1.57) 0.59 (3.37) 0.98 (6.74) 2.91 (8.43) Banks Is. 17.2 (19.2) 0.60 (1.38) 9.10 (12.6) 1.20 (15.6) 2.86 (29.4) 9.30 (15.0) Bathurst Is. 24.7 (22.9) 0.30 (0.55) 4.00 (4.52) 1.78 (4.50) 2.83 (7.69) 3.50 (5.05) Bylot Is. 1.77 (3.33) 0.53 (0.68) 1.29 (1.64) 1.71 (2.18) 1.42 (2.37) 0.70 (2.12) Coats Is. 19.4 (22.6) 0.50 (0.62) 5.90 (6.62) 13.4 (13.1) 25.3 (33.7) 2.61 (4.00) Cornwallis Is. 26.0 (23.9) 0.30 (0.60) 5.15 (5.62) 2.47 (8.81) 4.30 (11.4) 1.55 (5.00) Devon Is. 24.5 (25.1) 0.20 (0.26) 5.45 (6.01) 1.20 (1.85) 2.47 (3.82) 2.50 (13.9) Ellef Ringnes Is. 21.3 (49.1) 2.10 (4.26) 12.4 (35.6) 17.0 (71.6) 7.30 (47.4) 131 (303) Ellesmere Is. 31.0 (42.3) 1.00 (3.64) 7.57 (15.5) 3.25 (27.1) 2.74 (29.8) 7.90 (76.8) Melville Is. 5.8.0 (14.9) 0.85 (1.50) 3.00 (8.10) 2.00 (18.1) 4.12 (35.4) 1.80 (13.1) Northwest Territories 5.10 (12.7) 0.77 (1.25) 1.10 (4.30) 0.90 (2.97) 0.80 (3.75) 2.70 (10.9) Nunavut 4.08 (6.40) 0.50 (0.78) 1.32 (2.43) 1.12 (5.67) 1.77 (10.8) 1.71 (2.69) Prince Charles Is. 17.0 (17.3) 0.54 (0.60) 3.23 (3.20) 1.52 (5.35) 1.32 (7.47) 0.90 (2.57) Prince of Wales Is. 26.0 (27.0) 0.50 (0.48) 10.3 (10.6) 1.70 (1.54) 3.30 (3.03) 6.70 (9.52) Prince Patrick Is. 12.1 (13.2) 0.80 (1.06) 3.30 (5.15) 4.10 (12.9) 9.32 (25.0) 4.00 (7.7) Somerset Is. 10.0 (13.4) 0.24 (0.24) 3.60 (3.97) 0.71 (1.21) 1.31 (2.18) 1.80 (3.46) Southampton Is. 31.9 (29.8) 0.58 (1.32) 6.10 (6.87) 6.46 (19.2) 3.53 (8.60) 13.2 (39.6) Victoria Is. 22.5 (21.3) 0.34 (0.45) 9.44 (10.9) 0.75 (1.75) 1.70 (4.08) 2.04 (4.45) Yukon 30.0 (33.1) 2.60 (3.78) 18.5 (28.5) 4.04 (9.30) 1.38 (3.91) 20.5 (68.0)

Table 2.6 continued

-1 -1 -1 -1 -1 -1 mg·L mg·L mg·L mg·L µg·L µg·L Region SiO2 DOC POC DIC NH3* TKN* Axel Heiberg Is. 1.17 (1.74) 2.8 (4.98) 11.6 (13.9) 5.00 (10.67) 5.00 (10.7) 197 (385) Baffin Is. 1.10 (1.30) 1.91 (2.17) 3.99 (5.60) 21.0 (35.8) 21.0 (35.8) 92.5 (150) Banks Is. 1.15 (1.34) 5.60 (6.2) 15.9 (18.0) 377 (438) Bathurst Is. 0.28 (0.59) 3.20 (3.95) 16.3 (15.5) 7.00 (9.17) 7.00 (9.17) 162 (271) Bylot Is. 1.16 (1.44) 6.10 (5.23) 0.90 (1.03) 10.0 (14.9) 10.9 (14.9) 300 (333) Coats Is. 6.11 (5.97) 7.92 (9.20) 12.3 (46.6) 12.2 (46.6) Cornwallis Is. 0.40 (0.42) 1.80 (2.26) 19.1 (16.8) 8.50 (7.98) 8.59 (7.98) 80.0 (164) Devon Is. 0.35 (0.51) 1.72 (2.64) 17.2 (16.3) 7.10 (8.52) 7.19 (8.52) 83.5 (116) Ellef Ringnes Is. 1.18 (1.42) 1.90 (2.08) 2.40 (6.22) 5.00 (17.3) 5.00 (17.3) 147 (148) Ellesmere Is. 1.20 (1.97) 3.75 (6.61) 24.1 (23.8) 12.0 (22.2) 12.0 (22.2) 298 (473) Melville Is. 0.20 (0.41) 4.30 (5.28) 6.70 (11.9) 10.0 (11.0) 10.0 (11.04) 224 (327) Northwest Territories 0.40 (0.81) 10.4 (19.4) 3.95 (9.08) 9.00 (27.7) 9.00 (27.7) 348 (489) Nunavut 0.30 (0.48) 3.80 (4.99) 4.10 (5.10) 33.0 (37.0) 33.0 (37.0) 218 (366) Prince Charles Is. 3.79 (3.56) 9.07 (8.60) 15.8 (17.4) 15.8 (17.4) 390 (390) Prince of Wales Is. 0.48 (0.45) 4.60 (3.94) 24.7 (24.4) 14.0 (18.6) 14.0 (18.6) 399 (403) Prince Patrick Is. 0.17 (0.41) 6.90 (6.71) 7.60 (9.43) 26.0 (35.3) 26.0 (35.3) 490 (515) Somerset Is. 0.14 (0.22) 0.78 (1.17) 9.35 (9.91) 8.00 (11.9) 8.00 (11.9) 100 (117) Southampton Is. 0.61 (1.16) 5.33 (5.62) 20.3 (19.8) 235 (207) Victoria Is. 0.80 (0.92) 2.20 (2.71) 21.3 (20.5) 13.0 (17.57) 13.0 (17.6) 230 (258) Yukon 2.42 (3.84) 10.6 (12.5) 19.3 (27.9) 6.50 (11.5) 6.50 (11.5) 520 (640)

Table 2.6 continued

-1 -1 -1 -1 -1 -1 µg·L µg·L µg·L µg·L µg·L µg·L Region TN* TP* Al Ba Fe Mn Axel Heiberg Is. 188 (385.52) 3.15 (4.66) 10 (68.94) 9.6 (18.72) 5 (31.43) 0.4 (40.28) Baffin Is. 89.17 (104.55) 8 (11.72) 8.53 (217.55) 2.51 (3.67) 27.35 (123.71) 4.44 (9.6) Banks Is. 425 (502.69) 9.55 (13.37) 10 (136.91) 14.1 (19.14) 101 (396.22) 14.2 (27.32) Bathurst Is. 476.8 (526.27) 6.47 (8.31) 20 (75.24) 34.5 (47.58) 64.5 (203.52) 3 (5.16) Bylot Is. 255 (303.29) 7.8 (9.33) Coats Is. 438.34 (516.9) 23.69 (24.76) 6.4 (7.82) 3.31 (3.37) 19.6 (21.33) 8.19 (8.59) Cornwallis Is. 73.11 (63.9) 3.8 (25.69) 6.5 (11.93) 6.2 (17.35) 10 (20.69) 1.05 (1.35) Devon Is. 133.5 (278.01) 6.06 (15.01) 10 (35.29) 2.63 (5.08) 21 (44.08) 1.06 (1.77) Ellef Ringnes Is. 232 (294.52) 11.3 (23.98) 180 (399.2) 13.9 (16.35) 216 (680.32) 18 (205.19) Ellesmere Is. 477 (679.84) 6.34 (7.56) 20 (179.45) 5.1 (11.32) 73.5 (295.7) 5.5 (11.3) Melville Is. 191.5 (191.5) 11.2 (15.39) 55 (329.08) 5.4 (8.39) 150.5 (468.5) 5.45 (7.55) Northwest Territories 403 (530.32) 7.8 (10.29) 27.29 (43.31) 5 (15.09) 39.45 (129.63) 44 (3843.76) Nunavut 315 (402.7) 5.8 (7.82) 20.7 (57.43) 6.11 (7.62) 109 (222.63) 6.66 (263.59) Prince Charles Is. 432.21 (401.81) 14.75 (15.76) 5.18 (7.41) 1.27 (1.27) 11.78 (13.2) 9.3 (7.22) Prince of Wales Is. 174 (305.33) 4.25 (5.18) 40 (39) 13 (15.34) 59 (52.4) 2.1 (1.94) Prince Patrick Is. 591 (616.34) 9.6 (12.45) 30 (117.89) 12.7 (15.82) 267 (724.43) 5.6 (18.49) Somerset Is. 110.69 (144.16) 4.09 (5.36) 5.44 (6.17) 3.7 (3.39) 16 (24.71) 1.7 (2.69) Southampton Is. 605.9 (715.56) 4.7 (22.63) 12.35 (20.96) 12.35 (18.39) Victoria Is. 221 (258.53) 4.1 (5.57) 8.9 (10.56) 6.2 (7.13) 23.95 (31.55) 1.47 (2.05) Yukon 540 (650.45) 10.57 (13.13) 30.5 (30.07) 32 (30.23) 36.65 (107.08) 13.8 (20.55)

Table 2.6 continued

-1 µg·L Ratio Ratio Ratio Region Sr TN:TP Na:Cl Na:K Axel Heiberg Is. 48.2 (198) 42.5 (55.1) 0.53 (1.64) 1.23 (2.68) Baffin Is. 12.2 (13.5) 65.0 (118) 0.72 (1.60) 0.30 (0.97) Banks Is. 28.1 (38.1) 39.1 (48.9) 0.59 (0.88) 0.00 (0.44) Bathurst Is. 41.5 (86.5) 40.1 (48.9) 0.60 (1.23) 0.17 (1.14) Bylot Is. 115 (126) 0.58 (1.02) 1.27 (1.56) Coats Is. 26.0 (28.9) 171 (198) 0.57 (0.78) 32.0 (57.1) Cornwallis Is. 44.4 (56.2) 51.8 (65.5) 0.67 (1.49) 0.53 (2.67) Devon Is. 20.5 (104) 39.2 (46.5) 0.65 (1.74) 0.00 (7.47) Ellef Ringnes Is. 70.7 (126) 24.9 (23.8) 0.77 (1.04) 5.21 (7.43) Ellesmere Is. 78.3 (126) 40.4 (47.5) 0.61 (1.99) 0.60 (1.72) Melville Is. 16.0 (49.9) 46.1 (54.7) 0.66 (1.14) 0.00 (0.58) Northwest Territories 12.0 (92.1) 55.8 (65.1) 0.66 (1.27) 0.52 (0.73) Nunavut 19.5 (25.4) 53.6 (62.8) 0.64 (0.98) 0.27 (1.39) Prince Charles Is. 12.7 (15.3) 59.9 (59.9) 0.63 (0.63) 1.73 (2.54) Prince of Wales Is. 55.0 (48.0) 49.7 (37.1) 0.83 (1.24) 0.00 (0.00) Prince Patrick Is. 32.3 (48.3) 21.1 (22.8) 0.62 (0.81) 0.00 (1.88) Somerset Is. 28.0 (35.3) 44.1 (43.0) 0.62 (1.08) 0.00 (0.72) Southampton Is. 55.6 (67.4) 0.55 (1.51) 9.11 (9.57) Victoria Is. 16.1 (18.3) 45.3 (47.4) 0.70 (1.28) 0.00 (3.51) Yukon 174 (194) 37.6 (49.7) 0.59 (1.10) 1.18 (1.92)

Table 2.7 Significant differences calculated with Kruskal–Wallis rank sum test and Dunn’s post hoc test with Bonferroni adjustment for chemical variables among the geology types, and are indicated by * = p values <0.05, ** = p-value <0.01, and *** = p- value < 0.001

Igneous Igneous Igneous Sedimentary Sedimentary Supracrustal Sedimentary Supracrustal Unclassified Supracrustal Unclassified Unclassified pH *** *** * Cond *** *** *** Ca *** *** ** * K *** *** * Mg *** *** *** Na *** *** Cl *** **

SO4 *** ***

SiO2 *** DOC ** DIC *** ** * POC *

NH3* *** TKN* *** * TN* *** TP* *** *** Al ** Ba *** *** * Fe Mn ** Sr *** *** * ***

Table 2.8 Significant differences calculated with Kruskal–Wallis rank sum test and Dunn’s post hoc test with Bonferroni adjustment for chemical variables among the ecoregion types and are indicated by * = p values <0.05, ** = p-value <0.01, and *** = p-value < 0.001. AC = Arctic Cordillera, TU = Tundra, TA = Taiga, and NWF = Northwestern Forested Mountains.

AC AC AC NWF NWF TA NWF TA TU TA TU TU pH *** *** *** *** Cond *** *** *** *** Ca *** * *** *** *** K *** *** *** *** Mg *** *** *** *** *** Na ** *** *** ** Cl * * ***

SO4 *** *** ***

SiO2 * ** ** DOC *** *** * *** *** DIC ** *** * *** POC *** *** * ***

NH3* * TKN* *** *** *** *** TN* *** *** *** *** *** TP* ** * *** Al * Ba *** * * *** *** 55

Fe * Mn * *** *** Sr *** * ** *** ***

The median value for conductivity was 97.4 µS·cm–1, but conductivity values varied widely from 1.46 to 13,200 µS·cm–1 (CV = 262.98%) (Table 2.2). The majority of sites (75.5%, n = 933) had values below 200 µS·cm–1. Median conductivity values per region was lowest for Bylot Is. = 15.9 µS·cm–1, Northwest Territories = 48.0 µS·cm–1, and Baffin Is. = 50.8 µS·cm–1, while higher values were found for Yukon = 305 µS·cm–1,

Ellef Ringnes Is. = 228 µS·cm–1, and Southampton Is. = 222 µS·cm–1 (Table 2.6). Among geology types, conductivity value was higher for Sedimentary =132.0 µS·cm–1,

Supracrustal = 98.2 µS·cm–1, Unclassified = 87.0 µS·cm–1, and Igneous = 40.0 µS·cm–1

(Table 2.4). Similar to pH, conductivity was significantly different (p < 0.001) between

Igneous and the rest of the geology types (Figure 2.5 Left top; Table 2.7). Conductivity was more significantly correlated with Ca (rs = 0.89), Mg (rs = 0.88), and DIC (rs = 0.80), than of Na (rs = 0.69) and Cl (rs = 0.60) (all with p < 0.01; Figure 2.4) suggesting that geology had a greater impact on conductivity than sea-salt aerosols.

2.4.3 Major Cations and Anions

Median concentration of cations was Ca = 14.6 mg·L–1, K = 0.57 mg·L–1, Mg =

3.70 mg·L–1 and Na = 1.65 mg·L–1 (Table 2.2) and are within the lower range of inland freshwater systems in Canada (McNeely et al., 1979). Sedimentary geology had the higher median Ca concentration (20.8 mg·L–1), while Unclassified had the higher Mg concentration (5.90 mg·L–1) (Table 2.4). For both K and Na concentrations, higher medians were both (K = 1.17 mg·L–1, Na = 2.77 mg·L–1) found in the Supracrustal

geology (Table 2.4). Lower median values for all cations were found in the Igneous geology; Ca = 4.11 mg·L–1, K = 0.4 mg·L–1, Mg = 0.9 mg·L–1, and Na = 0.63 mg·L–1

(Table 2.4). Significant differences (p < 0.001) were found between Igneous with

Sedimentary and Supracrustal geology for all cation species (Table 2.7).

–1 –1 For anions, median concentrations were 2.05 mg·L for Cl, 3.10 mg·L for SO4

– (Table 2.2). Although studies have not reported bicarbonate (HCO3 ) concentrations, DIC concentrations can be used to represent bicarbonate, median concentration for DIC was

10.9 mg·L–1 (Table 2.2). Median anion values were also within the lower range of inland freshwater systems in Canada (McNeely et al., 1979). Higher median anion concentrations were consistently found for Sedimentary geology; 2.93 mg·L–1 for Cl, 3.90

–1 –1 mg·L for SO4, and 15.9 mg·L for DIC, while lower medians were consistently found

–1 –1 –1 for Igneous geology; 0.90 mg·L for Cl, 1.99 mg·L for SO4, and 3.7 mg·L for DIC

(Table 2.4). Similarly, to cations, significant differences among anions were primarily found between Igneous and Sedimentary geology (Figure 2.5 Left-bottom; Table 2.7).

Highly correlated rs values were found for DIC with Ca and Mg, Ca–DIC rs =

0.87, Mg–DIC rs = 0.81 (p < 0.01; Figure 2.4), supporting the dominate geological coverage of Sedimentary type geology in the study region. Correlation between Na and Cl

(rs = 0.85; p < 0.01, Figure 2.4) and elevation (rs = – 0.62; p < 0.01; Figure 2.4) suggests a sea-salt influence. Generally, median concentrations were Ca>Mg>Na>Na for cations and DIC> SO4>Cl for anions. Other median ion concentrations were; Nitrate Nitrite

–1 –1 (NO3+NO2) = 0.01 µg·L (n = 458), ammonia (NH3) = 0.0.1 µg·L (n = 744), and fluoride (F) = 0.02 µg·L–1 (n = 89) (Supplementary information Table A1).

2.4.4 Phosphorous, Nitrogen and Carbon

Among sites (n = 1248), Total Phosphorous (TP) range greatly from = 0.04 to

761.10 µg·L–1, with a median of 6.0 µg·L–1 (Table 2.2). A large portion of sites were classified as oligotrophic (45.6%, n = 569), with TP concentrations between 4–10 µg·L–1

(Canadian Council of Ministers of the Environment, 2004). This is followed by ultra- oligotrophic (24.8%, n = 310), mesotrophic (20.9%, n = 261), meso-eutrophic (5.5%, n =

69), eutrophic (2.5%, n = 31), and hyper-eutrophic (0.7%, n = 8). Median TP concentrations per region were lowest on Axel Heiberg Is. (3.15 µg·L–1, n = 47),

Cornwallis Is. (3.80 µg·L–1, n = 44), and Somerset Is (4.09 µg·L–1, n = 9), while higher median TP concentrations can be found on Coats Is. (23.7 µg·L–1, n = 10), Prince

Charles Is., (14.8 µg·L–1, n = 5) and Ellef Ringnes Is. (11.3 µg·L–1, n = 25) (Table 2.6).

Among geology types significant difference (p < 0.001) were found between Igneous

(median = 6.0 µg·L–1, n = 332) and Sedimentary (median = 7.4 µg·L–1, n = 825), and between Igneous and Supracrustal (median = 7.7 µg·L–1, n = 77) (Table 2.4). Between ecoregions, significant differences were found for AC (median = 6.0 µg·L–1, n = 23) –

NWF (median = 10.8 µg·L–1, n = 56; p < 0.01), NWF-TA (median = 7.85 µg·L–1, n =

118; p < 0.05), and NWF-TU (median = 6.85 µg·L–1, n = 1050; p < 0.001) (Table 2.5,

Table 2.8).

Median TN concentrations was median = 312 µg·L–1 for all observations (n =

864) (Table 2.2). The region of Cornwallis Is. had the lowest mean TN concentration

(73.1 µg·L–1, n = 4), followed by Baffin Is. (89.1 µg·L–1, n = 88), and Somerset Is. (110

µg·L–1, n = 8) (Table 2.6). Higher TN concentrations were found on Southampton Is.

(605 µg·L–1, n = 32), Prince Patrick Is. (591 µg·L–1, n = 35), and the Yukon (540 µg·L–1, 58

n = 22) (Table 2.6). Significant differences in TN concentration were primary between igneous (211 µg·L–1, n = 241) and sedimentary geology (376 µg·L–1, n = 590) (Table 2.6; p < 0.001) (Table 2.4, Table 2.7). Amongst ecoregions, significant differences (all with p

< 0.001) occurred between AC (75.00, n = 9) with NWF (613 µg·L–1, n = 16), TA (438, n

= 76) and TU (286, n = 763) and TU with NWF (613, n = 16) and TA (Figure 2.5 Bottom right; Table 2.5, Table 2.8). Generally, most sites (85.0%, n = 733) were found to be P- limited (TN:TP > 17; Sakamoto, 1966), while 11.7% (n = 101) are suggested to be N- limited systems (TN:TP < 14; Downing and McCauley, 1992).

Dissolved Organic Carbon (DOC) concentrations were almost ten-times higher than Particulate Organic Carbon (POC) concentrations; median = 3.50 mg·L–1 and 0.41 mg·L–1, respectively (Table 2.2). Concentrations of DOC ranged from 0.02 to 332 mg·L–1

(Table 2.2), reflecting coverage of different ecoregions the study region (Figure 2.1

Bottom). Among regions, higher median concentrations were found for Yukon (11.60, n

= 59), Northwest Territories (10.35, n = 118), and Prince Patrick Is. (6.90 mg·L–1, n =

35), with the lowest concentrations found for Somerset Is. (0.78 mg·L–1, n = 9), Devon Is.

(1.72 mg·L–1, n = 64) and Cornwallis Is. (1.80 mg·L–1, n = 43) (Table 2.6). Similarly to

TN, DOC was found to be significantly different (p < 0.01) primary between igneous

(2.90 mg·L–1, n = 241) and sedimentary geology (3.80 mg·L–1, n = 590) (Table 2.7), while ecoregions differ between AC (1.50 mg·L–1, n = 21) with NWF (10.6 mg·L–1, n =

53; p < 0.001), TA (12.90 mg·L–1, n = 84; p < 0.001) and TU (3.15 mg·L–1, n = 987; p <

0.05) and TU with NWF and TA (both with p < 0.001) (Figure 2.5 Top right, Table 2.5;

Table 2.8)

Strong correlation was found for DOC–TN (rs = 0.81, p < 0.01), and DOC–TKN

(rs = 0.88, p < 0.01), while weaker correlations were found between POC with TP (rs =

0.51, p < 0.01), POC–TKN (rs = 0.50, p < 0.01), TP–TN (rs = 0.47, p < 0.01) (Figure 2.4) suggest similar allochthonous (outside of the aquatic system) sources of nutrients from the catchment. There was a weak correlation between DOC–K (rs = 0.53, p < 0.01), TN–

K (rs = 0.57, p < 0.01) and TKN–K (rs = 0.52, p < 0.01), which may suggest that inputs of nitrogen were associated with vegetation (Figure 2.4).

2.4.5 Trace metals

More than 40 trace metal species were report (n = 42; Supplementary information

Table A1), although many were below detection limits or had limited observations across studies. Concentrations of Al (n = 872), Ba (n = 814), Fe (n = 1015), Mn (n = 875), and

Sr (n = 819) (Table 2.2.) were reported most often, along with Zn, Cu, and Li, which had observations >500 sites (Supplementary information Table A1). Of the subset of trace metal species, higher median concentrations of Al, Ba, Fe, Mn, and Sr were: 17.0 µg·L-1

(n = 872), 6.45 µg·L-1 (n = 814), 48.00 µg·L-1 (n = 1015), 5.30 µg·L-1 (n = 875), and

23.9 µg·L-1 (n = 819), respectively (Table 2.2). For regions of the Canadian Arctic, higher median values were found for Al on Ellef Ringnes Is. (180 µg·L-1, n = 25), Ba in

Bathurst Is. (34.5 µg·L-1, n = 64), Fe on Prince Patrick Is. (267 µg·L-1, n = 35), Mn in

Northwest Territories (44.0 µg·L-1, n = 109), and Sr in Yukon (174 µg·L-1, n = 33)

(Table 2.6). The lowest median values were found for Al (5.18 µg·L-1, n = 4) and Ba

(1.27 µg·L-1, n = 4) on Prince Charles Is., Fe (5.00 µg·L-1, n = 45) and Mn (0.40 µg·L-1, n = 45) on Axel Heiberg Is., and Sr in Northwest Territories (12.00 µg·L-1, n = 61)

(Table 2.6). 60

For geology type, significant differences were found between Igneous-

Supracrustal for Al and Mn, i.e., [Al] Igneous = 12.8 µg·L-1, n = 225 [Al] Supracrustal =

29.0 µg·L-1, n = 37 (p < 0.001); [Mn] Igneous = 6.57 µg·L-1, n = 236, [Mn] Supracrustal

= 7.35 µg·L-1, n = 42 (p < 0.01) (Table 2.4, Table 2.7). For Ba, differences were found between Igneous (4.00 µg·L-1, n = 215) with Sedimentary (7.89 µg·L-1, n = 551) and

Supracrustal (18.7 µg·L-1, n = 39) (both with p < 0.001), and Sedimentary with

Supracrustal (p < 0.05) (Table 2.4, Table 2.7). For Sr, differences were found between

Igneous (14.5 µg·L-1, n = 215) with sedimentary (30.1 µg·L-1, n = 556; p < 0.001), supracrustal (116 µg·L-1, n = 39; p < 0.001), unclassified (28.0 µg·L-1, n = 9; p < 0.05), and sedimentary and supracrustal (p < 0.001) (Table 2.4, Table 2.7). No significant differences were found for Fe across all geology types. Strong correlations were found between Sr–Cond (rs = 0.80) and Sr–Ca (rs = 0.78) (all p < 0.01; Figure 2.4). A weak correlation was found between Fe–Al (rs = 0.52; p < 0.01) and Ba-Sr (rs = 0.66; p < 0.01)

(Figure 2.4).

Amongst ecoregions, Al and Fe had only one significant difference; Al = TA

(30.0 µg·L-1, n = 16) – TU (13.8 µg·L-1, n = 771) (p < 0.05) and Fe = AC (100 µg·L-1, n

= 16) – NWF (32.5 µg·L-1, n = 56) (p < 0.05) (Table 2.5, Table 2.8). Manganese was significantly different between NWF (12.9 µg·L-1, n = 55) – TU (4.43 µg·L-1, n = 729) and TA (61.0 µg·L-1, n = 81) – TU (both p < 0.001), than AC (8.34 µg·L-1, n = 10) – TU

(p < 0.05) (Table 2.5, Table 2.8). For Ba and Sr, significant difference among ecoregions were found: AC ([Ba = 1.46, n = 9], [Sr = 3.68, n = 9]) – NWF ([Ba = 32.0, n = 33], [Sr

= 174, n = 33]) (p < 0.001), AC – TA ([Ba = 5.00, n = 53], [Sr = 19.0, n = 53) (Ba = p <

0.05, Sr = p < 0.01), NWF – TA (p < 0.001), and NWF – TU([Ba = 6.37, n = 719], [Sr =

23.0, n = 724) (p < 0.001) (Table 2.5, Table 2.8).

2.4.6 Drivers and relationships of water chemistry

A PCA was used to determine key drivers and relationship among water chemistry variables, and the variability within the dataset (Figure 2.6). The PCA was limited to 17 variables that were common across 613 sites. Components one and two of the PCA explained a total of 49.3% (PCA 1 = 33.8% and PCA 2 = 15.5%) of the variation (Figure

2.6). Eigenvalues (λ) were 6.44 for Axis 1, 2.58 for Axis 2, 2.08 for Axis 3, 1.41 for Axis

4, and 1.15 for Axis 5. Axis 3, 4, and 5 accounted for smaller portions of the variation

(12.2%, 8.3%, and 6.7%, respectively), and were not examined furthered, but PC loadings can be found in the supplementary information (Table A2). Variables that influenced

Axis 1 were (in descending order) were Cond, Mg, Na, Ca, and Cl. This suggests that the weathering of carbonate materials, such as those found on sedimentary geology, and the input of sea-salts are drivers of conductivity and major ions. Variables associated with

Axis 2 were: Al, Fe, and TP, suggesting weathering of non-carbonate soils. Variables of

Cond, Mg, and Ca were clustered with DIC and pH (Figure 2.6) and were highly correlated with another (p < 0.01; Figure 2.4), which suggest weathering of carbonate geology. Again, variables of Na and Cl were correlated with one another (p < 0.01;

Figure 2.4). They were weakly negatively correlated with elevation and positively correlated with Cond, Ca, and Mg (p < 0.01; Figure 2.4), which suggest the input of sea– salt aerosols. Also, Al, Fe, and TP were correlated with one another (p < 0.01; Figure

2.4).

Figure 2.6 Principal component analysis (PCA) of 27 physical and chemical variables, and 613 sites from across the Canadian Arctic. Sites are depicted in four geological type from Harrison et al. (2011); Igneous (red), Sedimentary (Green), Supracrustal (Blue), and Unclassified (purple).

2.5 Discussion

This study synthesizes observations of water chemistry from 1300 Arctic lakes and ponds spanning a period of 37 years. It is recognized (AMAP, 2006; Adrian et al.,

2009; Bégin et al., 2017) that there is a need to establish further baseline studies within known spatial gaps, such as northern mainland Northwest Territories, Boothia Peninsula,

parts of Baffin Is., Prince of Wales Is. Such knowledge is especially important if anthropogenic activities such as shipping and their emissions are expected to increase in future, potentially impacting Arctic freshwater bodies (Pizzolato et al., 2016). Equally, temporal studies such as Roberts et al. (2017) and Lougheed et al. (2011) are needed, as changes such as the prolonging of the growing season (Rouse et al., 1997) and reduction of ice-cover days (Surdu et al., 2016) have occurred within the Arctic region and are expected to impact physical, biological, and chemical processes in Arctic aquatic ecosystems.

Water chemistry of Arctic lakes and ponds were primarily driven by a conductivity/cation and a metal nutrient gradient (Figure 2.6). The conductivity/cation gradient has been previously reported by other limnological studies (Pienitz et al., 1997;

Michelutti et al., 2002a, b; Antoniades et al., 2003a, b; Lim and Douglas, 2003; Rühland et al., 2003; Lim et al., 2005; Mallory et al., 2006; Hamilton et al., 2001; Dranga et al.,

2017). This was of expected as most sites are situated over sedimentary type geology

(66.5%, n = 864; Figure 2.2 Top; Table 2.4). Our metal (Al, Fe) and phosphorus gradient is similar to other studies that have reported a mixture gradient of nutrient (POC, DOC,

TNU, PON, TPF, TP) and metal species (Al, Fe, Zn, and Mn) (Pienitz et al., 1997;

Michelutti et al., 2002; Antoniades et al., 2003a). Other studies have also reported physical-climatic conditions, i.e., depth and temperature, as a main driver of water chemistry (Pienitz et al., 1997; Rühland et al., 2003; Dranga et al., 2017), which may be explain through higher biogeochemical cycling with higher temperatures and/or different cycling between ponds and lakes. Although such variables (depth and temperatures) were not explored in the PCA (Figure 2.6), depth was explored though correlations (Figure

2.4), which produced similar results that indicate more dilution (less concentration of ions, nutrients, and metals) among deeper systems.

It is well established that geology can influence surface water pH (Michelutti et al., 2002a, b; Antoniades et al., 2003a, b; Michelutti et al., 2003; Lim et al., 2005; Keatley

2007; Keatley et al., 2007; Westover et al., 2009; Hadley et al., 2013). Lakes situated on sedimentary geology tend to be more alkaline (median pH = 8.06, n = 826) and those situated on igneous geology tend to be more acidic or neutral (median pH = 7.31, n =

328) (Table 2.4). Micheluitti et al. (2010) reported a mean pH of 8.1 (n = 407; range =

3.6–9.0) for Canadian High Arctic sites, which is much more alkaline than the results of this study, due to limited sampling and location on mid-western Arctic, where geology is predominately sedimentary. The clustering of pH and conductivity with Ca, Mg, and DIC

(Figure 2.6) suggests that pH and conductivity is largely influence by the weathering of carbonate rich sedimentary geology, which are composed of limestone (CaCO3), dolostone (CaMg(CO3)2 mineral (Harrison et al., 2011). These variables (Ca, Mg, and

DIC) were correlated (p < 0.01; rs > 0.55) with pH (Figure 2.4).

The igneous geology type was the most significantly different from the other geological types and was generally associated with lower concentrations of base cations, nutrients, and metals (Figure 2.5 Left middle; Table 2.4, Table 2.7). This is largely attributed to the high quartz content lithology that are commonly found in igneous geology, e.g., granite and rhyolite > 69% SiO2, Trachyte ~ 63% SiO2, and gabbro and basalt 45–52% SiO2 (Hodgon, 2003; Harrison et al., 2011), which provides limited buffering capacity against inputs of acidity.

Although most sites follow the regional pattern of geologically driven pH values, localized geology can have a significant impact on water chemistry. For examples,

Antoniades et al., (2003b) assessed the pH of 25 sites within a 7.5 km radius, which resulted in a range of pH values from acidic 5.1 to more alkaline values of 7.9. This suggest that lakes and ponds systems are greatly influenced by localized geology and can vary spatially. Certain geology may contain higher concentration of sulfur, such as pyrite

(FeS2) containing shale, e.g. Smoking Hills, NWT, which when oxidized, produces sulphate. When sulphate compounds are mixed with runoff, it produces sulphuric acid that can leach into nearby aquatic systems. Several sites with pH < 4.0 were found to be influenced by the oxidation of with sulfate soils (Johannesson and Lyons, 1995;

Michelutti et al., 2002; Antoniades et al., 2003a), which resulted in higher sulphate concentrations in the water (Havas and Hutchinson, 1983; Michelutti et al., 2002;

Antoniades et al., 2003a). Two sites surveyed in this study (KM_6, and KM_7) had pH <

4.0, with higher SO4 concentrations (>50 mg/L). These characteristics are similar to sites that are impacted by the oxidation of sulphur rich soils (Michelutti et al., 2002) and are attributed to iron sulphide minerals commonly found on the Meta Incognita Peninsula

(Hodgson, 2005).

The influence of sea-salt aerosols has long been observed to elevate concentration of Na and Cl among Arctic lakes and ponds, especially coastal sites (Pienitz et al., 1997a;

Lim et al., 2001 Michelutti et al., 2002a; Antoniades et al., 2003b; Mallory et al., 2006;

Cӧte et al., 2010; Hadley et al., 2013). Ratios between observed Na and Cl can be used to assessed the inputs from marine aerosols, with ratios values lower than 0.85 (Moller,

1989) suggestive of marine aerosols influence. Of the 1300 sites, 1251 had observations

for both Na and Cl, where median Na:Cl ratios (0.64) were lower than of 0.85 (Moller,

1989), which suggests a marine aerosols influence. Approximately 66.8% (n = 836) of the

1251 sites had ratios ≤ 0.85, while 33.2% (n = 415) have ratios > 0.85, which suggest that there was some contribution from weathering of geology. In general, within polar conditions, the concentrations of sea-salt components (primary Na and Cl) decrease exponentially from the coast to 200 km, where concentrations remain stable until 1000 km inland (Suzuki et al., 2002). In this study, most sites were situated <100 m.a.s.l

(40.3%, n = 524) and within <200 km (88.6%, n = 1153) from the coast (Figure 2.3), as such, the majority were influenced by sea-salts, as evidenced by the strong relationship between Na and Cl (rs ≥ 0.85, p < 0.01, Figure 2.4), and to some weak extend (but noted) distance to coast (Figure 2.4). In addition, Na and Cl were clustered together (Figure 2.6) and were negatively correlated with distance to coast (DistC) and elevation (Elev) (Figure

2.4). Although, Na is largely attributed to sea-salt aerosols, other sources can be from geological processes, especially those on shale (from clay minerals) and from historical marine beds (McNeely et al., 1979; Cerline et al., 1989).

The climatic characteristics of the Arctic are a major obstacle for primary productivity. Concentrations of K are attributed to the leaching of plant litter in the watershed (Cornwell, 1992), with higher concentration associated with more developed catchments, i.e., the present of vegetation and the absent of bare ground (Michelutti et al.,

2002; Keatley et al., 2007). Ratios of Na:K in waters can indicate the development of vegetation in the watershed, with typical ratios for lakes range from 1 to 5 (McNeely et al., 1979; Pienitz et al., 1997a, b; Rühland and Smol, 1998; Antoniades et al., 2003a) with higher ratios (18.4; Antoniades et al., 2003b) attributed to less developed catchments.

However, as discussed above, inputs of Na can be greatly attributed from the input of sea- salts, which may distort Na:K ratios toward a less developed catchment value.

To overcome this, Na and K concentrations were corrected for sea-salt, similar to that for critical loads of acidity approach (Henriksen and Posch, 2001). For median sea- salt correct concentrations for Ca, Mg, Na, K, and SO4, per region, ecoregion, and geology type, please see supplementary information (Table A3). Of the sites with both observations of Na and K (n = 1184), 92.5% (n = 1095) have Na:K ratios (sea-salt corrected; median = 3.1) within the natural range, i.e. Na:K = 1-5 (McNeely et al., 1979).

By correcting for sea-salt ratios, base cation concentrations were lowered than those reported by other studies (Pienitz et al., 1997a, b; Rühland and Smol, 1998; Antoniades et al., 2003a). For example, highest mean ratio found on Ellef Ringne Is. by Antoniades et al., (2003b) was 18.4, this value decreased (but was still out of the natural range) to 7.43

(Table 2.6) when corrected for sea-salt. Surprisingly, median Na:K ratios for Coats Is.

(31.95) and Southampton Is. (9.11) had ratios >5 (Table 2.6) suggesting that catchments were bare or had little vegetation. However, the works of Gould et al., (2002), Gould et al., (2003), Walker et al., (2005), Mallery et al., (2006), Wiken et al., (2011), indicate strong presence of Arctic willow (Salix Arctica) and Mountain-avens (Dryas integrifolia) on both Coats Is. and Southampton Is., which may suggest that sea-salt correction may not account for all sea-salt input in these coastal sites.

Generally, most of Arctic lakes and ponds in this study were nutrient poor.

Phosphorous is the limiting factor for primary productivity in freshwater systems. The majority of sites were oligotrophic (4.0–10.0 µg·L–1; CCME, 2004). Phosphorous can enter the aquatic system through external loading such as that from the decay of 68

vegetation matter (Antoniades et al., 2003b), mammal and avian feces (Lim et al., 2001;

Mallory et al., 2006) and runoff over phosphorus geology (Hamilton et al., 2001). The cluster of Al, Fe, and TP (Figure 2.6) and the weak correlation between TP with Fe (rs =

0.44) and POC (rs = 0.50) (p < 0.01; Figure 2.4) are suggestive of phosphorus inputs from geological allochthonous source such as strengite (FePO4·2H2O), and carbonatite, or sedimentary phosphorites found in shale or chert (Konhauser et al., 1994; Hamilton et al.,

2001; Antoniades, et al., 2003a; Harrison et al., 2011). Large Arctic avian colonies of

Greater Snow Geese (Chen caerulescens atlanticus), Lesser Snow Geese (Chen caerulescens caerulescens) and Ross's goose (Chen rossii), have been known to cause lake catchment degradation (Alisauskas et al., 2006; Hines et al., 2010) and enhance phosphorus concentration within surface waters (Mallory et al., 2006; Brimble et al.,

2009; Côte et al., 2010; Michelutti et al., 2010). However, these were found to be a small percentage of sites and only cause 39 sites (of 1247) to be classified as eutrophic or hyper-eutrophic. Although these mechanisms suggest high inputs of phosphorus loading, phosphorus availability are often limited. Since most sites (with available data) are shallow (85.4%, n = 581; Figure 2.3 Bottom) and small (58.0%, n = 509, at ≤10 ha;

Hamilton et al., 2001), they are oxic environments that bind phosphorus to iron (III) compounds in lake sediment (Mortimer, 1941; Søndergaard et al., 2003). Whalen &

Cornwell (1985) indicated that most phosphorus enters polar lakes through streams and runoff and is removed though sedimentation and burial.

Ratios of TN to TP can determine if an aquatic system is either N or P limited.

Median TN:TP ratio for all sites (with observations of TN and TP) was 48:1, with most sites (85.0%, n = 733) found to be P-limited. Ratios indicating N-limited systems (TN:TP

< 14; Downing and McCauley, 1992) accounted for 11.7% (n = 101) of all sites with observations of TN and TP.

When using TN concentration to determine trophic status, 54.9% (n = 474) are considered to oligotrophic (TN < 350 µg·L–1; Nürnberg, 1996). Past studies of Arctic aquatic systems (Alexander et al., 1989; Ditmar and Kattner, 2003) suggest that inputs of nitrogen are primarily the result of nitrogen fixation from algae and cyanobacteria

(Whalen and Cornwell, 1985; Alexander et al., 1989). However, sites with elevated TN concentration were attributed to the input of feces from large bird colonies (Mallory et al.

2006; Brimble et al., 2009; Keatley et al., 2009; Michelutti et al., 2010). However, these cases are very specific and were not found to be common across our results, i.e., majority of sites (45.6%, n = 569) are oligotrophic. This large input of nitrogen (from feces) can enhance catchment development as nitrogen is more limiting than phosphorus in tundra ecosystems (Elser et al., 2007). For example, Bazely and Jefferies (1985) reported that the increase in biomass of Creeping goose grass (Puccinellia phryganodes) and Hoppner's sedge (Carex subspathacea) were significant (p < 0.01; mean of geese site = 199 g·m–3 vs. non-geese sites = 122 g·m–3) and were attributed to the effect of >5000 pairs of Lesser

Snow Geese near Churchill, Manitoba, Hudson Bay.

Carbon across Arctic aquatic systems are predominately found in the form of

DOC. Median DOC across the Canadian high Arctic have been reported to be 3.0 mg·L–1

(n = 404) (Michillutti et al., 2010), which is slightly lower than that found in this study

(3.50 mg·L–1, n = 1135; Table 2.2). This can be attributed to the inclusion of lower latitude studies, e.g., Moser et al. (1998); Wilson and Gajewski (2002); Rühland et al.,

(2003), in the NWF and TA ecoregions (Figure 2.2 Bottom; Table 2.1). Vegetation and 70

soils within catchments provide an allochthonous source of organic carbon in the form of

DOC and POC, via terrestrial runoff. Neff et al., (2016) reported that the input of DOC is vastly terrestrial, with most occurring during the spring melt. They also indicated that sources change from recent organic matter, i.e., vegetation litter and surface soil horizons, during the spring and older stored carbon during the late summer. Other studies have reported higher DOC concentrations with lusher vegetated catchments (Lim et al., 2001,

2005; Wilson and Gajewski, 2002; Antoniades et al., 2003a; Rühland et al., 2003).

Overall, weak correlations between DOC with K (rs = 0.54), and POC with Fe (rs = 0.47) may support both origins, i.e., vegetation litter and soil. Lusher vegetated catchments can contribute to elevated DOC concentrations, but they are also associated with more developed organic soils.

Climate models have suggested a higher rate of precipitation in the Arctic (7.5–

18.1% greater) with larger portions occurring as rain (Kattsov et al., 2005; AMAP, 2017).

Increasing air temperature (thawing of permafrost) and (wet) precipitation are factors that can increase the transportation (via runoff) of solutes and nutrients into aquatic environments, thus changing the water chemistry of aquatic systems. In a recent study,

Robert et al., (2017) reported an increase of +500% and +340% in sulphate concentrations (from 5 to 17 mg·L-1 and from 3 to 15 mg·L-1) from 2006 to 2016 in two high Arctic lakes. In addition, concentrations of Ca (~50%), Mg (~75%), K (~25 to 75%), and Na (~75 to 100%) also increased between the year 2003 and 2015 (Robert et al.,

2017). Similar results were also found by Thienpont et al. (2013), where disturbed lakes

(n = 5) had higher ionic concentrations when compared to reference lakes (n = 5; no disturbance from thaw slumping); these included less diluted [Disturbed = 460.3 µS·cm-1,

Non-disturbed = 88.1 µS·cm-1], higher pH [Disturbed = 8.04, Reference = 6.95],

Alkalinity [Disturbed = 93.0 mg·L-1, Reference = 23.2 mg·L-1], Ca [Disturbed = 93.0

-1 -1 -1 mg·L , Reference = 9.90 mg·L ], and SO4 [Disturbed = 93.0 mg·L , Reference = 12.2 mg·L-1]. Increased streamflow from groundwater are also associated with increased nutrient loading into aquatic systems (Walvoord and Strieg, 2007). These large changes in water chemistry from enhance catchment processes (owning from climate change) further press the need for more long-term monitoring sites in the Canadian Arctic

2.6 Conclusion

Few studies have provided an overview of Arctic water chemistry at a regional scale (Hamilton et al., 2001; Medeiros et al., 2012; Dranga et al., 2017). In general, bedrock geology dictates a large portion of water chemistry variables, where significant differences where primarily found between sites on sedimentary and igneous geology.

However, there can be hydrochemical characteristics overlap among geology type, e.g., where sites with more acidic pH can occur on sedimentary geology. One rational to this is that the scale of the geological layer (1:5,000,000) by Harrison et al. (2011) was not able to capture the localized variation. However, in general, those on sedimentary tended to be alkaline, and have a higher concentration of major ions, nutrients, and trace metals than those on igneous geology. However, across all geology, most chemical sources are of terrestrial allochthonous sources. Localized characteristics such as proximity to the coast, minerology (Pyrite or carbonate minerals) and biological communities (avian colonies or vegetation) can greatly impact pH, concentrations of metals, and nutrient inputs.

Changes within the cryosphere from climate change are expected to change the

Arctic landscape and will ultimately change the water chemistry of Arctic lakes and ponds. Current Arctic limnological studies are limited due to weather and logistical conditions, which results in spatial and temporal knowledge gaps, and a grab-what-you- can system of sampling. More limnological studies are needed to fill in known spatial gaps such as those on Baffin Is, Prince of Wales Is., southwestern Victoria Is., and northern mainland of Yukon and Northwest Territories (other than the Mackenzie basin).

This will further our knowledge of baseline limnological characteristic of Arctic lakes and ponds, which can then be used to assess potential impacts from increased anthropogenic activity, such as those from increased shipping (Pizzolato et al., 2016). In addition, and equally as important, are the need for temporal studies such as those of Robert et al.

(2017), or to some extent re-sampling of past sites such as Lougheed et al. (2011), to help understand the impacts of climate change on Arctic aquatic ecosystems.

Acknowledgments

This study was funded by the Natural Science and Engineering Research Council

(NSERC) grant awarded to Julian Aherne, as well as the McLean foundation, Northern

Studies Training Programme (NSTP) award to Tanner Liang. We are thankful for the financial support from Environment and Climate Change Canada (ECCC), and logistical support from the Nunavut Research Institute (NRI) and the Polar Continental Shelf

Program (PCSP). We thank the many authors of the various publication for making the data available. Thanks to Scott Fleming and Dr. Peter Lafleur for their assistance in the field, and to Hazel Cathcart for her GIS and data mining assistance.

References Adrian, R., O’Reilly, C.M., Zagarese, H., Baines, S.B., Hessen, D.O., Keller, W., Livingstone, D.M., Sommaruga, R., Straile, D., Van Donk, E., Weyhenmeyer, G.A., & Winder, M. (2009). Lakes as sentinels of climate change. Limnology and Oceanography. 54: 2283–2297. doi:10.4319/lo.2009.54.6_part_2.2283. Alexander, V., Whalen, S.C. & Klingensmith, K.M. (1989). Nitrogen cycling in Arctic lakes and ponds. Pages 165-172 in High Latitude Limnology. Springer, Dordrecht. doi: 10.1007/978-94-009-2603-5_12 Aliabadi, A.A., Staebler, R.M. & Sharma, S. (2015). Air quality monitoring in communities of the Canadian Arctic during the high shipping season with a focus on local and marine pollution. Atmospheric Chemistry and Physics. 15: 2651-2673. doi: 10.5194/acp-15-2651-2015 Alisauskas, R. T., J. Charlwood, & D. K. Kellett. (2006). Vegetation correlates of nesting history and density by Ross’s and lesser snow geese at Karrak Lake, Nunavut. Arctic 59:201–210.doi: 10.14430/arctic342 Arctic Monitoring and Assessment Programme. (1998). AMAP - Assessment Report Arctic Pollution issues. Arct. Pollut. Issues: 859. doi:10.4324/9781315064451. Arctic Monitoring and Assessment Programme. (2017). Snow, Water, Ice and Permafrost in the Arctic SWIPA. Arctic Monitoring and Assessment Programme AMAP, Oslo, Norway. 269 p Antoniades, D., Douglas, M.S. V., & Smol, J.P. (2003a). Comparative physical and chemical limnology of two Canadian High Arctic regions: Alert (Ellesmere Is., NU) and Mould Bay (Prince Patrick Is., NWT). Arch. für Hydrobiol. 158: 485–516. doi:10.1127/0003-9136/2003/0158-0485. Antoniades, D., Douglas, M.S.V., & Smol, J.P. (2003b). The Physical and Chemical Limnology of 24 Ponds and One Lake from Isachsen, Ellef Ringnes Is., Canadian High Arctic. Int. Rev. Hydrobiol. 88: 519–538. doi:10.1002/iroh.200310665. Babaluk, J.A., Heuring, L.G., Reist, J.D., & Billeck, B.N. (1999). Selected chemical analyses of water from lakes in Ellesmere Is. National Park Reserve, Northwest Territories. Can. Data Rep. Fish. Aquat. Sci. 1050 iv: 7. Babaluk, J.A., Gantner, N., Michaud, W., Muir, D.C.G., Power, N., Reist, J.D., Sinnatamby, R., & Wang, X. (2009). Chemical Analysis of Water from Lakes and Streams in Quttinirpaaq National Park, Nunavut. Can. Data Rep. Fish. Aquat. Sci. 1217 Bazely, D.R., & Jefferies, R.L. (1985). Goose Faeces: A source of nitrogen for plant growth in a grazed salt marsh. Journal of Applied Ecology. 22:693-703. doi: 10.2307/2403222 Bégin, P.N., Lebedeva, L., Tashyreva, D., Velazquez, D. & Blaen, P.J. (2017). Future 74

priorities for Arctic freshwater science from the perspective of early-career researchers. Arctic Science, 3:661-671. Bouchard, G., Gajewski, K., & Hamilton, P.B. (2004). Freshwater diatom biogeography in the Canadian Arctic Archipelago. J. Biogeogr. 31: 1955–1973. doi:10.1111/j.1365-2699.2004.01143.x. Brimble, S.K., Blais, J.M., Kimpe, L.E., Mallory, M.L., Keatley, B.E., Douglas, M.S. V, & Smol, J.P. (2009). Bioenrichment of trace elements in a series of ponds near a northern fulmar (Fulmarus glacialis) colony at Cape Vera, Devon Is.. Can. J. Fish. Aquat. Sci. 66: 949–958. doi:10.1139/F09-053. Bunbury, J., & Gajewski, K. (2005). Quantitative analysis of freshwater ostracode assemblages in southwestern Yukon Territory, Canada. Hydrobiologia. 545: 117– 128. doi:10.1007/s10750-005-2746-0. Bunbury, J., and Gajewski, K. (2009). Biogeography of freshwater ostracodes in the canadian arctic archipelago. Arctic. 62: 324–332. Canadian Council of Ministers of the Environment. (2004). Canadian water quality guidelines for the protection of aquatic life: Phosphorus: Canadian Guidance Framework for the Management of Freshwater Systems. In: Canadian environmental quality guidelines, 2004, Canadian Council of Ministers of the Environment, Winnipeg. Commission for Environmental Cooperation. (1997). Ecological regions of North America: toward a common perspective. Cerling, T.E., Pederson, B.L. and Von Damm, K.L. (1989). Sodium-calcium ion exchange in the weathering of shales: Implications for global weathering budgets. Geology. 17:552–554.doi: 10.1130/0091-7613(1989)017<0552:scieit>2.3.co;2 Corbett, J.J., Lack, D.A., Winebrake, J.J., Harder, S., Silberman, J.A., & Gold, M. (2010). Arctic shipping emissions inventories and future scenarios. Atmospheric Chemistry and Physics. 10:9689–9704.doi: 10.5194/acp-10-9689-2010 Côté, G., Pienitz, R., Velle, G., & Wang, X. (2010). Impact of geese on the limnology of lakes and ponds from Bylot Is. (Nunavut, Canada). Int. Rev. Hydrobiol. 95: 105– 129. doi:10.1002/iroh.200911151. Dranga, S.A., Hayles, S. & Gajewski, K. (2017). Synthesis of limnological data from lakes and ponds across Arctic and Boreal Canada. Arctic Science. doi:10.1139/as- 2017-0039. Dawson, J., D.A., Winebrake, J.J., Harder, S., Silberman, J.A. & Gold, M. (2017). Shipping Trends in Nunavut from 1990-2015. Ottawa. Devlin, J.E. (2010). Applications Of Paleolimnology In Ecosystem Monitoring For Sirmilik National Park: Developing Indicators Of Ecological Integrity. University of Toronto.

Devlin, J.E. & Finkelstein, S.A. (2011). Local physiographic controls on the responses of Arctic lakes to climate warming in Sirmilik National Park, Nunavut, Canada. Journal of Paleolimnology. 45: 23–39.doi:10.1007/s10933-010-9477-6 Dittmar, T. & Kattner, G. (2003). The biogeochemistry of the river and shelf ecosystem of the Arctic Ocean: a review. Marine Chemistry. 83:103-120.doi: 10.1016/s0304- 4203(03)00105-1 Downing, J.A. & McCauley, E. (1992). The nitrogen:phosphorus relationship in lakes. Limnology and Oceanography. 37:936–945.doi: 10.4319/lo.1992.37.5.0936 Eckhardt, S., Hermansen, O., Grythe, H., Fiebig, M., Stebel, K., Cassiani, M., Bäcklund, A. and Stohl, A., 2013. The influence of cruise ship emissions on air pollution in Svalbard–a harbinger of a more polluted Arctic?. Atmospheric Chemistry and Physics, 13:8401-8409.doi: 10.5194/acp-13-8401-2013 Elser, J.J., Bracken, M.E., Cleland, E.E., Gruner, D.S., Harpole, W.S., Hillebrand, H., Ngai, J.T., Seabloom, E.W., Shurin, J.B., & Smith, J.E. (2007). Global analysis of nitrogen and phosphorus limitation of primary producers in freshwater, marine and terrestrial ecosystems. Ecology letters. 10:1135-1142.doi: 10.1111/j.1461- 0248.2007.01113.x Environmental Canada. (1994a). Environmental Canada Manual of Analytical Methods: Major Ions and Nutrients, Vol 1. National Laboratory for Environmental Testing, Canadian Centre for Inland Waters, Burlington, Ontario, Canada. Environmental Canada. (1994b). Environmental Canada Manual of Analytical Methods: Trace Metals, Vol 2. National Laboratory for Environmental Testing, Canadian Centre for Inland Waters, Burlington, Ontario, Canada. Fulton, R.J. (1989). Quaternary Geology of Canada and Greenland. Geological Survey of Canada, Geology of Canada Series.1–839 Gould, W.A., Edlund, S., Zoltai, S., Raynolds, M., Walker, D.A., & Maier, H. (2002). Canadian Arctic vegetation mapping. International Journal of Remote Sensing. 23:4597-4609.doi: 10.1080/01431160110113962 Gould, W.A., Rayolds, M., & Walker, D.A. (2003). Vegetation, plant biomass, and net primary productivity patterns in the Canadian Arctic. Journal of Geophysical Research. 108:1–14. doi: 10.1029/2001jd000948 Hadley, K.R. (2007). Assessing Thule Inuit Impacts on High Arctic Lakes and Ponds: A Paleolimnological Approach. Master of Science Thesis. Queen’s University. Hadley, K.R., Douglas, M.S. V, Lim, D., & Smol, J.P. (2013). Diatom assemblages and limnological variables from 40 lakes and ponds on Bathurst Is. and neighboring high Arctic Island. Int. Rev. Hydrobiol. 98: 44–59. doi:10.1002/iroh.201201449. Haley, S., Klick, M., Szymoniak, N., & Crow, A. (2011). Observing trends and assessing 76

data for Arctic mining. Polar Geography 34:37–61. doi: 10.1080/1088937x.2011.584449 Hamilton, P.B., Gajewski, K., Atkinson, D.E., & Lean, D.R.S. (2001). Physical and chemical limnology of 204 lakes from the Canadian Arctic Archipelago. Hydrobiologia.457: 133–148. doi:10.1023/A:1012275316543. Harrison, J.C., St–Onge, M.R., Petrov, O.V., Strelnikov, S.I., Lopatin, B.G., Wilson, F.H., Tella, S., Paul, D., Lynds, T., Shokalsky, S.P., Hults, C.K., Bergman, S., Jepsen, H.F. & Solli, A. 2011. Geological map of the Arctic / Carte géologique de l'Arctique; Geological Survey of Canada, Map 2159A, scale 1:5 000 000 Havas, M. & Hutchinson, T.C. (1983). The Smoking Hills: natural acidification of an aquatic ecosystem. Nature.30:23–27.doi:10.1038/301023a0. Hines, J.E., P.B. Latour, & C.S. Machtans. 2010. The effects on lowland habitat, breeding shorebirds and songbirds in the Banks Is. Migratory Bird Sancturay Number 1 by the growing colony of Lesser Snow Geese (Chen caerulescens caerulescens). Canadian Wildlife Service Occasional Paper No. 118. Environment Canada, Ottawa Hodgson, D.A. (2005). Quaternary geology of Western Meta Incognita Peninsula and Iqaluit area, Baffin Is., Nunavut. Bull. Geol. Surv. Canada: 1–72. ICP Waters Programme Centre. (2010). ICP Waters Programme Manual 2010. Norwegian Institute for Water Research, Oslo. ISBN 978-82-577-5809-7 Johannesson, K.H. & Lyons, W.B. (1995). Rare-earth element geochemistry of Colour Lake, an acidic freshwater lake on Axel Heiberg Is., Northwest Territories, Canada. Chemical Geology, 119:209-223.doi: 10.1016/0009-2541(94)00099-t Kattsov, V.M., & Källén. (2005). Future Climate Change: Modeling and Scenarios for the Arctic. In Arctic Climate Impact Assessment. Cambridge University Press, p1042 Keatley, B.E. (2007a). Environmental Change in Three Distinct Ecosystem Types, Canadian High Arctic. Doctor of Philosophy Thesis, Department of Biology, Queen’s University. Kingston, Ontario, Canada. Keatley, B.E., Douglas, M.S. V, & Smol, J.P. (2007b). Limnological Characteristics of a High Arctic Oasis and Comparisons Ellesmere Is. across Northern. Arctic. 60: 294– 308. Keatley, B.E., Douglas, M.S., Blais, J.M., Mallory, M.L. & Smol, J.P. (2009). Impacts of seabird-derived nutrients on water quality and diatom assemblages from Cape Vera, Devon Is., Canadian High Arctic. Hydrobiologia. 621:191-205.doi: 10.1007/s10750-008-9670-z Konhauser, K.O., Fyfe, W.S., Schultze-Lam, S., Ferris, F.G. & Beveridge, T.J. (1994). Iron phosphate precipitation by epilithic microbial biofilms in Arctic Canada. Can. J. Earth Sci. 31:1320-1324.doi: 10.1139/e94-114

Law, K. S., & Stohl, A. (2007). Arctic air pollution: Origins and impacts. Science. 315:1537–1540.doi: 10.1126/science.1137695 Lim, D.S.S., Douglas, M.S.V., Smol, J.P., & Lean, D.R.S. (2001). Physical and chemical limnological characteristics of 38 lakes and ponds on Bathurst Is., Nunavut, Canadian high arctic. Int. Rev. Hydrobiol. 86: 1–22. doi:10.1002/1522- 2632(200101)86:1<1::AID-IROH1>3.0.CO;2-E. Lim, D.S.S., & Douglas, M.S.V. (2003). and Ponds in the Haughton of 22 Lakes Characteristics Limnological Arctic Canadian of Devon Island, Nunavut, High Region. Arctic, Antarctic, and Alpine Research. 35: 509–519. Lim, D.S.S., Douglas, M.S.V. & Smol, J.P. (2005). Limnology of 46 lakes and ponds on Banks Is., N.W.T., Canadian Arctic Archipelago. Hydrobiologia. 545: 11–32. doi:10.1007/s10750-005-1824-7. Lougheed, V.L., Butler, M. G., McEwen, D. C., & Hobbie, J. E. (2011). Changes in Tundra Pond Limnology: Re-sampling Alaskan Ponds After 40 Years. AMBIO, 40:589–599. Available at: http://dx.doi.org/10.1007/s13280-011-0165-1. Mallory, M.L., Fontaine, A.J., Smith, P.A., Robertson, M.O.W., & Gilchrist, H.G. (2006). Water chemistry of ponds on Southampton Is., Nunavut, Canada: Effects of habitat and ornithogenic inputs. Arch. Fur Hydrobiol. 166: 411–432. doi:10.1127/0003- 9136/2006/0166-0411. Maxwell, J.B. (1981). Climatic regions of the Canadian Arctic Islands. Arctic. 34: 225– 240. McNeely, R.N., Dwyer, L., & Neimanis, V.P. (1979). Water quality sourcebook : a guide to water quality parameters. Environment Canada, Inland Waters Directorate, Water Quality Branch, Ottawa, pp.89 Medeiros, A.S., Biastoch, R.G., Luszczek, C.E., Wang, X.A., Muir, D.C.G., & Quinlan, R. (2012). Patterns in the limnology of lakes and ponds across multiple local and regional environmental gradients in the eastern canadian arctic. Inland Waters. 2: 59–76. doi:10.5268/IW-2.2.427. Meier, W.N., Stroeve, J. and Fetterer, F. (2007). Whither Arctic sea ice? A clear signal of decline regionally, seasonally and extending beyond the satellite record. Annals of Glaciology, 46:428–434.doi: 10.3189/172756407782871170 Michelutti, Neal, Douglas, Marianne, Muir, Derek, Wang, Xiaowa,& Smol, J. (2002a). Limnological Characteristics of 38 Lakes and Ponds on Axel Heiberg Is., High Arctic Canada. Int. Rev. Hydrobiol. 87: 385–399. Michelutti, N., Douglas, M.S. V, Lean, D.R.S., & Smol, J.P. (2002b). Physical and chemical limnology of 34 ultra-oligotrophic lakes and ponds near Wynniatt Bay, Victoria Is., Arctic Canada. Hydrobiologia. 482: 1–13. doi:10.1023/A:1021201704844.

Michelutti, N., Douglas, M. S., & Smol, J. P. (2003). Diatom response to recent climatic change in a high arctic lake (Char Lake, Cornwallis Is., Nunavut). Global and Planetary Change, 38:257–271.doi: 10.1016/s0921-8181(02)00260-6 Michelutti, N., Wolfe, A.P., Vinebrooke, R.D., Rivard, B., and Briner, J.P. (2005). Recent primary production increased in Arctic lakes. Geophysical Research Letters. 32:1–4. doi: 10.1029/2005gl023693 Michelutti, N., Douglas, M.S. V, & Smol, J.P. 2007. Evaluating diatom community composition in the absence of marked limnological gradients in the high Arctic: A surface sediment calibration set from Cornwallis Is. (Nunavut, Canada). Polar Biol. 30: 1459–1473. doi:10.1007/s00300-007-0307-x. Michelutti, N., Blais, J.M., Mallory, M.L., Brash, J., Thienpont, J., Kimpe, L.E., Douglas, M.S. V., &Smol, J.P. (2010). Trophic position influences the efficacy of seabirds as metal biovectors. Proc. Natl. Acad. Sci. 107: 10543–10548. doi:10.1073/pnas.1001333107. Mortimer, C.H. (1941). The exchange of dissolved substances between mud and water in lakes. Journal of Ecology, 29:280-329.doi: 10.2307/2256691 Moser, K. a, Smol, J.P., & MacDonald, G.M. (1998). Ecology and distribution of diatoms from boreal lakes in Wood Buffalo National Park, northern Alberta and the Northwest Territories, Canada. Hydrobiologia.23:25–43.doi: Neary, B.P. & Dillon, P.J. (1988). Effects of sulphur deposition on lake-water chemistry in Ontario, Canada. Nature, 333:340.doi: 10.1038/333340a0 Nürnberg, G.K. (1996). Trophic state of clear and colored, soft-and hardwater lakes with special consideration of nutrients, anoxia, phytoplankton and fish. Lake and Reservoir Management. 12(4): 432-447, doi: 10.1080/07438149609354283 Osterkamp, T.E. & Romanovsky, V.E. (1999). Evidence for warming and thawing of discontinuous permafrost in Alaska. Permafrost and Periglacial Processes, 10: 17- 37.doi 10.1002/(sici)1099-1530(199901/03)10:1<17::aid-ppp303>3.0.co;2-4: Overland, J.E., Hanna, E., Hanssen-Bauer, I., Kim, S.J., Walsh, J.E., Wang, M., Bhatt, U.S., & Thoman, R.L. (2017). Surfae Air Temperature [in Arctic report Card 2017], ftp://ftp.oar.noaa.gov/arctic/documents/ArcticReportCard_full_report2017.pdf Paltan, H., Dash, J., & Edwards, M. (2015). A refined mapping of Arctic lakes using Landsat imagery. Int. J. Remote Sens. 36: 5970–5982. Taylor & Francis. doi:10.1080/01431161.2015.1110263. Payette, S., Delwaide, A., Caccianiga, M. & Beauchemin, M. (2004). Accelerated thawing of subArctic peatland permafrost over the last 50 years. Geophysical Research Letters. 31: doi: 10.1029/2004gl020358 Peters, G.P., Nilssen, T.B., Lindholt, L., Eide, M.S., Glomsrød, S., Eide, L.I. & Fuglestvedt, J.S., (2011). Future emissions from shipping and petroleum activities

in the Arctic. Atmospheric Chemistry and Physics, 11:.5305–5320. doi: 10.5194/acp-11-5305-2011 Pienitz, R., Smol, J.P., & Lean, D.R.S. (1997a). Physical and chemical limnology of 59 lakes located between the southern Yukon and the Tuktoyaktuk Peninsula, Northwest Territories (Canada). Can. J. Fish. Aquat. Sci. 54: 330–346. doi:10.1139/f96-274. Pienitz, R, Smol, J.P., & Lean, D.R.S. (1997b). Physical and chemical limnology of 24 lakes located between Yellowknife and Contwoyto Lake, Northwest Territories (Canada). Ca. J. Fish. Aquat. Sci. 54.47–358. doi: 10.1139/f96-275 Pienitz, R., Douglas, M.S.V., & Smol, J.P. (2004). Long-term Environmental Change in Arctic and Antarctic Lakes. Springer, Springer Netherlands.pp. 540 Pienitz R., Doran, P.T., & Lamoureux, S.F. (2008). Origin and geomorphology of lakes in the polar regions, in Polar Lakes and Rivers Page 25—41. In Limnology of Arctic and AntArctic Aquatic Ecosystems. Oxford University Press, Oxford, pp. 25–41. doi: 10.1093/acprof:oso/9780199213887.003.0002 Pizzolato, L., Howell, S.E.L., Dawson, J., Laliberté, F., & Copland, L. (2016). The influence of declining sea ice on shipping activity in the Canadian Arctic. Geophys. Res. Lett. 43: 12,146-12,154. doi:10.1002/2016GL071489. Plug, L.J., Walls, C., & Scott, B.M. (2008). Tundra lake changes from 1978 to 2001 on the Tuktoyaktuk Peninsula, western Canadian Arctic. Geophysical Research Letters. 35:1–5.doi: 10.1029/2007gl032303 Polyakov, I.V., Walsh, J.E. & Kwok, R., (2012). Recent changes of Arctic multiyear sea ice coverage and the likely causes. Bulletin of the American Meteorological Society, 93:145–151. doi: 10.1175/bams-d-11-00070.1 Prowse, T.D., Wrona, F.J., Reist, J.D., Gibson, J.J., Hobbie, J.E., Lévesque, L.M.J., & Vincent, W.F. (2006). Climate change effects on hydroecology of arctic freshwater ecosystems. Ambio. 35: 347–58. doi:10.1579/0044-7447(2006)35. Rautio, M., Dufresne, F., Laurion, I., Bonilla, S., Vincent, W.F., & Christoffersen, K.S. (2011). Shallow freshwater ecosystems of the circumpolar Arctic. Écoscience. 18: 204–222. doi:10.2980/18-3-3463. Reist, J.D., Wrona, F.J., Prowse, T.D., Power, M., Dempson, J.B., Beamish, R.J., King, J.R., Carmichael, T.J. and Sawatzky, C.D. 2006. General effects of climate change on Arctic fishes and fish populations. AMBIO: A Journal of the Human Environment, 35:370–380.doi: 10.1579/0044- 7447(2006)35[370:GEOCCO]2.0.CO;2 Roberts, K.E., Lamoureux, S.F., Kyser, T.K., Muir, D.C.G., Lafrenière, M.J., Iqaluk, D., Pieńkowski, A.J., & Normandeau, A. (2017). Climate and permafrost effects on the chemistry and ecosystems of High Arctic Lakes. Sci. Rep. 7: 1–8.

doi:10.1038/s41598-017-13658-9. Romanovsky, V.E., Drozdov, D.S., Oberman, N.G., Malkova, G.V., Kholodov, A.L., Marchenko, S.S., Moskalenko, N.G., Sergeev, D.O., Ukraintseva, N.G., Abramov, A.A. and Gilichinsky, D.A. (2010). Thermal state of permafrost in Russia. Permafrost and Periglacial Processes, 21:136–155.doi: 10.1002/ppp.683 Rothrock, D.A., Yu, Y. & Maykut, G.A. (1999). Thinning of the Arctic sea‐ice cover. Geophysical Research Letters. 26:3469–3472.doi: 10.1029/1999GL010863 Rouse, W.R., Douglas, M.S., Hecky, R.E., Hershey, A.E., Kling, G.W., Lesack, L., Marsh, P., McDonald, M., Nicholson, B.J., Roulet, N.T. & Smol, J.P. (1997). Effects of climate change on the freshwaters of Arctic and subarctic North America. Hydrological Processes, 11:873–902.doi: 10.1002/(sici)1099- 1085(19970630)11:8<873::aid-hyp510>3.0.co;2-6 Rühland, K.M. & Smol, J.P. (1998). Limnological Characteristics of 70 Lakes Spanning Arctic Treeline from Coronation Gulf to Great Slave Lake in the Central Northwest Territories, Canada. Int. Rev. Hydrobiol. 141: 137–141.doi: 10.1002/iroh.19980830302 Rühland, K.M., Smol, J.P., Wang, X., & Muir, D.C.G. (2003). Limnological characteristics of 56 lakes in the Central Canadian Arctic Treeline Region. Journal of Limnology. 62: 9–27. doi:10.4081/jlimnol.2003.9. Sakamoto, M. (1966). Primary production by phytoplankton community in some Japanese lakes and its dependence on lake depth. Fundamental and Applied Limnology/Archiv für Hydrobiologie. 62: 1–28. Smith, L. C., Sheng, Y., MacDonald, G. M., & Hinzman, L. D. (2005). Disappearing arctic lakes. Science. 308:1429–1429. doi: 10.1126/science.1108142 Søndergaard, M., Jensen, J.P. and Jeppesen, E. (2003). Role of sediment and internal loading of phosphorus in shallow lakes. Hydrobiologia, 506:135-145.doi: 10.1023/B:HYDR.0000008611.12704.dd Stewart, K.A., and Lamoureux, S.F. 2011. Connections between river runoff and limnological conditions in adjacent high arctic lakes: Cape Bounty, Melville Is., Nunavut. Arctic 64: 169–182.doi: /10.14430/arctic4097 Stonehouse, B. (1989). Polar ecology. Blackie, Glasgow, 222p.doi: 10.1007/978-1-4757- 1260-5 Surdu, C.M., Duguay, C.R., and Prieto, D.F. (2016). Evidence of recent changes in the ice regime of lakes in the Canadian high Arctic from spaceborne satellite observation. The Cryosphere. 10:941–960. doi: 10.5194/tc-10-941-2016 Tarnocai, C. (2003). Arctic Permafrost Soils Pages 3-17 in Permafrost soils. (Eds. Margesin, R.) Springer, Berlin, vol 16 doi: 10.1007/978-3-540-69371-0_1

Trettin, H.P. ed. (1991). Geology of the Innuitian orogen and Arctic platform of Canada and Greenland. Geological Society of America. Geology of Canada Series no. pp.3– 569. doi: 10.4095/133959 Thienpont, J.R., Rühland, K.M., Pisaric, M.F.J., Kokelj, S. V., Kimpe, L.E., Blais, J.M., & Smol, J.P. (2013). Biological responses to permafrost thaw slumping in Canadian Arctic lakes. Freshw. Biol. 58: 337–353. doi:10.1111/fwb.12061. Vincent, W.F., Laurion, I., Pienitz, R., & Walter Anthony, K.M. (2012). Climate Impacts on Arctic Lake Ecosystems. Page 496 In Climatic Change and Global Warming of Inland Waters: Impacts and Mitigation for Ecosystems and Societies (Eds. Goldman, C.R., Kumagai, M., and Robarts, R.D.). John Wiley & Sons. doi: 10.1002/9781118470596.ch2 Walker, D.A., Raynolds, M.K., Daniëls, F.J.A., Einarsson, E., Elvebakk, A., Gould, W.A., Katenin, A.E., Kholod, S.S., Markon, C.J., Melnikov, E.S., Moskalenko, N.G., Talbot, S.S., Yurtsev, B.A., & CAVM Team. (2005). The Circumpolar Arctic Vegetation Map. Journal of Vegetation Science.16:267-282. doi: 10.1111/j.1654- 1103.2005.tb02365.x Walvoord, M.A. & Striegl, R.G. (2007). Increased groundwater to stream discharge from permafrost thawing in the Yukon River basin: Potential impacts on lateral export of carbon and nitrogen. Geophysical Research Letters, 34:1–6. doi: 10.1029/2007gl030216 Westover, K.S., Moser, K.A., Porinchu, D.F., MacDonald, G.M., & Wang, X. (2009). Physical and chemical limnology of a 61-lake transect across mainland Nunavut and southeastern Victoria Is., Central Canadian Arctic. Fundam. Appl. Limnol. / Arch. für Hydrobiol. 175: 93–112. doi:10.1127/1863-9135/2009/0175-0093. Wetzel, R.G. (2001). Salinity Of Inland Waters. Pages 169–186 in Limnology. Elsevier. doi:10.1016/B978-0-08-057439-4.50014-9. Wiken, Ed, Francisco J.N., & Griffith, G. (2011). North American Terrestrial Ecoregions – Level III. Commission for Environmental Cooperation, Montreal, Canada Whalen, S.C., and Cornwell, J.C. (1985). Nitrogen, Phosphorus, and Organic Carbon Cycling in an Arctic Lake. Can. J. Fish. Aquat. Sci. 42: 797–808. doi:10.1139/f85- 102. Wilson, S.E., and Gajewski, K. (2002). Surface-sediment diatom assemblages and water chemistry from 42 subarctic lakes in the southwestern Yukon and northern British Columbia, Canada. Ecoscience. 9: 256–270. doi:10.1080/11956860.2002.11682712.

3.0 Critical Loads of Acidity and Exceedances for 1138 Lakes and Ponds in the Canadian Arctic

3.1 Abstract

Sulphur emissions associated with increased anthropogenic activity, such as resource extraction and marine shipping, may lead to the acidification of aquatic freshwater systems in the Arctic. In the current study, acid sensitivity (based on the critical loads

(CL(A)) approach) of 1138 lakes and pond in the Canadian Arctic was quantified using the Steady-State Water Chemistry (SSWC) model. Acidification risk was estimated under modelled sulphur deposition for the year 2010 (using the GEM-MACH models: with two scenarios: with and without shipping). In general, different levels of ecosystem protection based on ANClimit values (for Brown Trout and Arctic Char) produced similar CL(A) values, with those for Arctic Char having the most sensitive CL(A). Overall, surface water CL(A) values for the Canadian Arctic were low (median = 35.8 meq·m―2·yr―1, mean = 96.2 meq·m―2·yr―1), with approximately 40% (n = 455) of sites estimated to be sensitive to acidification (CL(A) < 40 meq·m-2·yr-1). Higher CL(A) values were found in the Yukon (435 meq·m―2·yr―1, n = 40), on Ellesmere Is. (262 meq·m―2·yr―1, n = 143), and on Southampton Is. (252 meq·m―2·yr―1, n = 35), while lower CL(A) values were found on Melville Is. (5.5 meq·m―2·yr―1, n = 48), Banks Is. (18.4 meq·m―2·yr―1, n =

45) and Bylot Is. (20.4 meq·m―2·yr―1, n = 36). Under modelled deposition for 2010, both with shipping and without shipping scenarios, approximately 12% (n = 136–142) of all sites were exceeded. Which suggest that impacts from shipping are very small.

Key Words: Critical loads, Arctic surface waters, steady-state water chemistry model, sulphur deposition

3.2 Introduction

Elevated air temperatures owing to anthropogenic climate change have reduced sea ice extent (Perovich et al., 2016), diminished sea ice thickness (Kwok and Rothrock,

2009; Tilliong et al., 20156) and reduced the amount of multi-year ice (Tilliong et al.,

20156) in the Arctic. The presence of multiyear ice is the most significant obstacle to marine traffic in the Canadian Arctic; Pizzolato et al. (2016) observed increased traffic

(by 43–150 transits) in regions where multiyear ice concentration was at a low of 2/10th

(amount of ice coverage per unit of ocean). This reduction in sea ice (both extent and thickness) has opened up new opportunities for resource extraction, tourism, fishing, and shipping within the Canadian Arctic. Current observations suggest that shipping activity

(in km travelled) increased (during the period of 1990–2015) in most regions of the

Canadian Arctic by 250–550 km (2.5–5.5 transits) per year (Pizzolato et al., 2016). In the territory of Nunavut alone, 25 of 28 communities have observed increased shipping traffic since 1999 (Dawson et al., 2017). The utilization of the North-West Passage

(NWP) was attributed to higher shipping traffic in Cambridge Bay, Nunavut, while mining related activity was attributed to higher ship traffic in Pond Inlet (in addition to increased tourism), Chesterfield Inlet, and Baker Lake (Dawson et al., 2017).

Increased shipping in the Arctic has been shown to lead to increased ship source emissions of acidifying compounds such as sulphur dioxide (SO2). Corbett et al. (2010) estimated that cargo and container shipping emitted 34,000 and 40,000 mt of SOx, respectively, which represented 23.3% and 27.4% of total ship emissions (the rest are attributed to other types of shipping; Fishing vessels, Tankers, Offshore Service Vessels,

Barges, etc.) in the Arctic for the year 2004. It is predicted that the growth of cargo 85

shipping along the NWP will increase by 3.0 to 4.8% under business-as-usual (BAU) to high-growth (HG) scenarios from 2004–2020 and will continue to grow by up to 3.8% and 5.4% between 2004–2050 (Corbett et al., 2010). Concentrations of SO2 in ambient air within Arctic coastal towns (Cape Dorset and Resolute, Nunavut) have been found to be elevated by 10–18% when ships are in transit (Aliabadi et al., 2015) and by 15% when ships are in port (Eckhardt et al., 2013).

There is growing concern that elevated ship-source emissions, especially SO2, in the Arctic will impact surrounding Arctic ecosystems. Lentic aquatic systems, i.e., lakes and ponds, are a major feature of the Canadian Arctic landscape, and are estimated to cover an area of 190,000 km2 (Paltan et al., 2015), with upwards of 80% of the total land cover in some regions (Pienitz et al., 2008). These freshwater systems are often small

(<10 ha) and shallow (<12 m depth) lakes or ponds (Hamilton et al., 2001, Rautio et al.,

2011; Paltan et al., 2015; Chapter 2). Most are hydrologically isolated and recharged primarily by precipitation and runoff (Stewart and Lamoureux, 2011), which results in concentrations of major ions existing within the lower range of natural freshwater systems in Canada (McNeely et al., 1979; Chapter 2). Past research has shown that many of the lakes and ponds across the western Canadian Arctic have a pH of 8.0 to 8.5 (Hamilton et al., 2001; Michelutti et al., 2010) owing to carbonate rich bedrock. In contrast, lakes and ponds located on the central to eastern mainland (Northwest Territories and Nunavut) and in the eastern perimeter of the Arctic Archipelago (Baffin Island) have pH values that range from 5.8 to 8.0 (Rühland and Smol, 1998; Hamilton et al., 2001; Joynt III and

Wolfe, 2001; Chapter 2), which reflects the acid-sensitive geology of the Pre-Cambrian

Shield. As such, lakes and ponds in the Canadian Arctic may be sensitive to acidic deposition depending on their surrounding bedrock.

The impacts of acidic deposition have had been well established since the early

1970’s, when the acidification of aquatic systems resulted in the loss of fish populations in Scandinavia and North America (Beamish, 1974; Leivestad and Muniz, 1976;

Hesthagen et al., 1990; Schindler et al., 1985). To address this concern, effects-based emissions reductions based on the critical loads (CL) concept were adopted into air pollutant control policies. A CL is defined as a “quantitative estimate of exposure to one or more pollutants below which significant harmful effects on specific sensitive elements of the environment do not occur according to present knowledge” (Nilsson and Grennfelt,

1988). A CL of acidity (CL(A)) is the long-term maximum amount of acidic deposition that an ecosystem can tolerate; the lower the CL(A) the more sensitive to acidic deposition. Risk of damage to the selected biota is based on the excess of acidic deposition (is greater than) than the CL(A). Impacts of acidification to aquatic systems can results in loss of lower trophic communities (starvation), with long-term stress leading to fish unable to reproduce (Schindler et al., 1985).

This threshold (effects-based) concept has been widely used to reinforce emissions control polices by linking deposition of atmospheric pollutants to ecosystem health. In Europe, CL underpin the United Nation Economic Council for Europe (UN-

ECE) Convention of Long-Range Transboundary Air Pollution (LRTAP) and its protocols (Sliggers and Kakebeeke 2004) and the European Unions (EU) National

Emission Ceiling Directive (Hettleingh et al., 2013). In North America, the Memorandum of Intent on Transboundary Air pollution (US-Canada, 1983) and The Canada-Wide Acid 87

Rain Strategy for Post-2000 (CCME, 1998) have reduced sulphur emissions by 65% since

1985 (CCME, 2013). Within Canada, CL(A) has been used to address ecosystem health concerns from resources extraction activities, such as the Athabasca Oil Sand (Aherne,

2008; Cathcart et al., 2016) and Liquefied Natural Gas Projects in British Columbia

(Williston et al., 2016). However, few studies in Canada have determined CL(A) for

Arctic ecosystems (AMAP 1998; Forsius et al., 2001). For aquatic systems, AMAP

(1998) reported a median CL(A) value of 127 meq·m−2·yr−1 for 251 lakes, with 8% of lakes (mostly on Baffin Island and central mainland) with values ≤10 meq·m−2·yr−1. For terrestrial systems, a median CL(A) value of 73.5 meq·m−2·yr−1 was reported by Forsius et al. (2001). They attributed low CL(A) to low weathering rates (owing to polar climatic conditions) of sandy soils on acid parent material, i.e., Pre-Cambrian Shield.

The objective of this study is to (1) to determine CL(A) for 1138 lakes and ponds, and (2) to estimate exceedance under modelled 2010 sulphur deposition from the Global

Environmental Multi-scale - Modelling Air quality and CHemistry model with under two scenarios; with shipping and without shipping.

3.3 Methods

3.3.1 Study Area

The Canadian region of the Monitoring and Assessment Program (Stonehouse,

1989; AMAP 1998) was used to define the study area. This region covers all land mass above 60⁰N; an area of approximately 4.0 x106 km2 (Figure 3.1), which includes the regions of Yukon, Northwest Territories and Nunavut, and parts of northern Quebec and

Labrador (AMAP, 1998). It is estimated that there are 1.49 million lakes and ponds

within the Canadian Arctic region (>60⁰N; Paltan et al., 2015). Climate across the region is characterised by low temperatures (–20 to –35⁰C in January) and precipitation (<100–

200 mm annually, with 20–50% as snow or ice), extended long winters, short summers, and extreme seasonally light exposure (24 hrs of darkness in the winter, and 24 hrs of light during the summer; Maxwell, 1981). Much of the Canadian Arctic overlies either the Arctic Platform, which consists of sedimentary geology, such as shale, siltstone, sandstone, limestone, and dolomite. The eastern perimeter and much of mainland

Northwest Territories and Nunavut are situated on the Pre-Cambrian Shield, which is composed of gabbro, gneisses, granitic, and volcanic rocks (Fulton, 1989). Arctic soils are poorly developed with chemical properties similar to that of their parent materials.

There are two major shipping routes within the Canadian Arctic; the NWP and the

Hudson Strait. The NWP connects the Atlantic and the Pacific Ocean through a labyrinth of straits and islands of the Canadian Arctic Archipelago (Figure 3.1 Top). The Hudson

Strait (between Southern Baffin Island and Northern Quebec and Labrador) connects the

Atlantic Ocean to Hudson Bay (Figure 3.1 Top). Both maritime corridors are generally open from July to mid-October (AMSA, 2009).

Figure 3.1 Location names of islands and regions within the Canadian Arctic (top). Locations of all sites (bottom) from both the database (red dots) and from the lake surveys (yellow dots), with the AMAP boundary depicted in red. 3.3.2 Arctic water chemistry database

Hydrochemical data for lakes and ponds in the Canadian Arctic were collated from studies published during the period 1979 – 2010 (Chapter 2). Data points were filtered to include only the most recent water chemistry, i.e., if a lake had two observations, then the most recent was chosen. Further, only water bodies with chemical observations that included Ca, Mg, K, Na, Cl, and SO4 were selected (n = 1138) for CL calculations (Table 3.1; Figure 3.1 Bottom) as they are the primary variables used to determine CL(A) (Henriksen et al., 1992). Water chemistry quality controls (ionic balance check; ICP Waters Programme Centre, 2010) were performed on the data set (See section 2.3.2 in Chapter 2). This data subset included the lake surveys on Baffin Is. (n =

80), Coats Is. (n = 10), Prince Charles Is. (n = 4), and Northwest Territories (n = 6) from

Chapter 2.

Table 3.1 Number of lakes, and average values for runoff and selected chemical variables for regions of the Canadian Arctic.

No. of Runoff pH COND Ca† Mg† K† Na† Cl† SO4† Region sites m·yr-1 µS·cm-1 mg·L-1 mg·L-1 mg·L-1 mg·L-1 mg·L-1 mg·L-1 Axel Heiberg Is. 47 0.03 5.03 402 32.1 14.0 4.22 53.6 80.6 62.9 Baffin Is. 125 0.26 1.06 88.1 8.79 1.59 0.68 3.44 6.84 8.49 Banks Is. 45 0.01 7.90 182 19.3 12.6 1.38 15.9 29.4 15.0 Bathurst Is. 67 0.04 7.85 127 22.9 4.52 0.55 4.50 7.69 5.05 Bylot Is. 36 0.10 1.56 37.0 3.33 1.64 0.68 2.18 2.37 2.12 Coats Is. 10 0.15 ― 112 22.6 6.62 0.62 13.1 33.7 4.00 Cornwallis Is. 44 0.05 0.74 116 23.9 5.63 0.60 8.81 11.4 5.00 Devon Is. 40 0.02 7.98 121 25.1 6.01 0.26 1.85 3.82 13.9 Ellef Ringnes Is. 25 0.03 6.21 405 49.1 35.6 4.26 71.6 47.4 303 Ellesmere Is. 143 0.10 2.16 338 44.3 15.6 3.25 29.4 31.3 74.5 King William Is. 4 0.02 7.27 155 17.0 9.70 0.47 3.32 6.95 2.10 Melville Is. 48 0.01 7.80 152 15.2 8.26 1.50 18.5 35.4 12.8 Northwest Territories 96 0.07 6.95 93.5 12.7 4.30 1.39 2.36 2.43 12.4 Nunavut 183 0.08 2.26 77.7 6.54 2.45 0.80 5.84 11.1 2.73 Prince Charles Is. 5 0.10 0.10 98.6 17.3 3.20 0.60 5.35 7.47 2.57 Prince of Wales Is. 5 0.02 8.18 114 27.0 10.6 0.48 1.54 3.03 9.52 Prince Patrick Is. 35 0.03 7.67 115 13.2 5.15 1.06 12.9 25.0 7.70 Somerset Is. 13 0.06 7.81 80.9 13.4 3.97 0.24 1.21 2.18 3.46 Southampton Is. 35 0.14 7.82 261 29.9 6.98 1.38 20.2 8.78 41.7 Victoria Is. 89 0.02 7.74 155 21.2 10.8 0.45 1.74 4.04 4.41 Yukon 40 0.14 8.25 407 39.1 28.5 4.19 6.66 3.36 56.7 Note: COND = conductivity, DIC = dissolve inorganic carbon, DOC = dissolved organic carbon, TN = Total Nitrogen, TP = Total Phosphorous † The influence of sea-salt aerosols has not been removed.

Table 3.1 continue DOC DIC TN TP Region mg·L-1 mg·L-1 µg·L-1 µg·L-1 Axel Heiberg Is. 4.98 13.9 0.38 0 Baffin Is. 2.20 5.73 0.10 0.06 Banks Is. 6.20 18.0 0.50 0.01 Bathurst Is. 3.77 15.5 0.52 0.01 Bylot Is. 1.31 1.03 0.03 0.01 Coats Is. 5.97 9.2 0.52 0.01 Cornwallis Is. 2.20 16.8 0.06 0 Devon Is. 1.86 16.3 0.13 0 Ellef Ringnes Is. 2.08 6.22 0.29 0.02 Ellesmere Is. 5.89 21.9 0.56 0.01 King William Is. 0 ― ― 0.01 Melville Is. 5.27 12.1 0.19 0.01 Northwest Territories 11.4 9.85 0.64 0.02 Nunavut 3.98 5.28 0.41 0.01 Prince Charles Is. 3.56 8.6 0.4 0.02 Prince of Wales Is. 3.94 24.4 0.29 0 Prince Patrick Is. 6.71 9.43 0.62 0.01 Somerset Is. 0.81 9.92 0.12 0 Southampton Is. 5.83 19.8 0.72 0.02 Victoria Is. 2.09 20.3 0.25 0.01 Yukon 10.4 ― ― 0.01

3.3.3 Steady-State Water Chemistry Model

Critical loads of acidity, CL(A), were calculated with the Steady-State Water

Chemistry (SSWC) Model (Henriksen et al., 1992; Henriksen and Posch, 2001):

[1] CL(Ac) = ([BC ∗]0 − [ANClimit]) · Q

where [BC ∗]0 is the pre-acidification sea-salt corrected catchment-average net base cation flux, ANClimit is a threshold value for the 95% probability that damage will not occur to a specified indicator biota, such as fish, and Q is the catchment runoff. Soil weathering within catchments produces base cations that leach into lakes through runoff.

Long-term meteorological data (New et al., 2000) in conjunction with the MeHyd model

(a meteo-hydrological model; Bonten et al., 2016) were used to generate Q, where the nearest neighbour function was performed to calculate the average runoff values for each site (in m·yr-1).

Observed water chemistry data (through limnological lake surveys, as mentioned above)

4 base cations (calcium [Ca], magnesium [Mg], potassium [K], and sodium [Na]) and sulphate [SO4] are required to determine CL for each site. The input of ions through sea- salt aerosols, were corrected using individual ion ratios to chloride (Cl) concentration following Lyman and Fleming (1940). Sea-salt corrected concentrations were calculated as follows:

[2] [Ca∗] = [Ca] − 0.037 · [Cl]

[3] [K∗] = [K] − 0.018 · [Cl]

[4] [Mg∗] = [Mg] − 0.195 · [Cl]

[5] [Na∗] = [Na] − 0.858 · [Cl]

∗ [6] [SO4] = [SO4] − 0.103 · [Cl]

[7] [Cl∗] = 0

Sulphate concentrations are not only derived from atmospheric deposition, but also from geological sources. Higher concentrations of 푆푂4 from sedimentary type geology have been found in lakes across the Arctic (Pienitz et al., 1997; Antoniades et al., 2003; Lim and Douglas, 2003; Lim et al., 2005). These higher concentrations of sulphate can reduce the ability of surface waters to buffer incoming acidity; however, they are typically associated with the weathering of base cations. Following Lien et al. (1995) and

Závodský et al. (1995), geological sulphate [SO4]geo was determined by subtracting SO4

∗ deposition from the current observation of lake sulphate [SO4]t:

S [8] [SO ] = [SO∗ ] − dep,t 4 geo 4 t Q

where Sdep,t is anthropogenic SO4 deposition (for the year 2000 in the current study;

Lamarque et al., 2013), divided by Q (runoff; see above), to estimate the concentration from deposition alone. Higher SO4 conditions will affect base cation concentrations, and thus geological sulphate and sea-salt corrected base cation concentrations are calculated as:

∗† ∗ [9] [BC ]t = [BC ]t − [SO4]geo

where [BC*†]t is the current observed base cation concentration corrected both for sea- salts and base cations released in association with the release of geological sulphate. A similar approach has been used for lakes in Saskatchewan (Whitefield et al., 2016).

Thus, the pre-industrial concentration of non-anthropogenic, non-marine base cation

∗ concentration [BC]0 needed to calculated CL(A) (see equation 1) is estimated as:

S −S [10] [BC∗] = [BC∗†] − F · dep,t dep,0 0 t Q

where Sdep is the acidic deposition in the Canadian Arctic, with the subscripts t and 0 referring to present and pre-acidification concentrations, respectively, determined using data from the Atmospheric Chemistry and Climate Model Intercomparison Project

(ACCMIP) for the year 2000 (to represent present deposition) and 1850 (to represent pre- acidification concentrations) (Lamarque et al., 2013).

The leaching of base cations owing to acidic deposition was estimated using the F-factor

(Henriksen and Posch, 2001), which also used the modified geological sulphate and sea- salt corrected base cation concentration [BC*†]t to account for the influence of geological

SO4:

π Q[BC∗†] [11] F = sin ( t) 2 S where S represents the base cation concentration when all incoming acidity is neutralized in the catchment. This value was set at 400 meq·m―2·yr―1based from sites in Norway

∗† ―2 ―1 (Brakke et al., 1990). Lake or pond observations with [BC]t > than 400 meq·m ·yr were given an F-factor of 1 (n = 54).

The ANClimit is set to protect sensitive indicator biota and is generally based on empirical studies between fish or invertebrate populations, and water chemistry (Lien et al., 1996).

Specific ANClimit values have been recommended for a variety of fish species, but a

-1 general ANClimit of 20 µeq·L is widely used to protect fish, aquatic invertebrates, and benthic organisms (Lien et al., 1996). Organic acids (TOC and DOC) can act as strong acid anions reducing the acid neutralizing (buffering) capacity of a lake to incoming acidity. To accommodate this, Lydersen et al. (2004) proposed an organic acid adjusted

ANClimit (ANCoaa.limit) and found that prediction of fish population status improved under

-1 ANCoaa.limit values. Observations of DOC (expressed in mg·L ) were used to calculate an

-1 ANClimit based on ANCoaa.limit values of 8 µeq·L for Brown Trout (Salmo trutta), and 11

µeq·L-1 for Arctic Char (Salvelinus alpinus) (Lydersen et al., 2004). A similar approach was used in northern Saskatchewan (Scott et al., 2010; Cathcart et al., 2016) and northern

Manitoba (Jefferies et al., 2010).

[12] 퐴푁퐶푙푖푚푖푡 = 퐴푁퐶표푎푎.푙푖푚푖푡 + (10.2/3) × 퐷푂퐶

This equation assumes that the particulate carbon fraction is <10% (DOC is used instead of TOC) that 1/3 of organic acids are deprotonate (proton donor) at pH >4.6 and that they carry a charge density of 10.2 µeq·mgC-1 (Lydersen et al., 2004). In the current study,

-1 lakes without DOC values (n = 146), were given an ANClimit of 20 µeq·L .

Sites were classified on their acid sensitivity and are based on the classification classes of

Dupont et al. (2005). Classes are as followed: Highly Sensitive (<20 meq·m―2·yr―1), Sensitive

(20–40 meq·m―2·yr―1), Moderately Sensitive (40–60 meq·m―2·yr―1), Sensitivity (60–100 meq·m―2·yr―1), and Very Low Sensitivity (> 100 meq·m―2·yr―1).

3.3.4 Exceedances

Total sulphur deposition (푆푑푒푝) values for the year 2010, were obtained from two scenarios from Environment and Climate Change Canada’s (ECCC) Global

Environmental Multi-scale - Modelling Air quality and CHemistry version 2 (GEM-

MACHv2) Model. The GEM-MACH is an atmospheric chemistry and meteorology process model that is able to simulate chemical transport at a regional level. The gas- phase chemistry within GEM-MACH includes 47 gas-phase species and 114 reactions that are represent though nine aerosol chemicals: sulfate (SO4), nitrate (NO3), ammonium

(NH4), elemental carbon (EC), primary organic matter (POA), secondary organic matter

(SOA), crustal material (CM), sea-salt, and particle-bound water (Gong et al., 2018). The model is primary used for forecasting data for Canada’s Air Quality Health Index, and for supporting air quality policies such as the United Nation’s Convention on Long-range

Transboundary Air Pollution and air quality agreements between the Government of

Canada and the Government of the United States of America.

The first scenario was obtained from Moran et al. (2016) at a 10 km × 10 km resolution across most of North America. The second scenario was obtained by Gong et al. (2018) and was produced by a modified GEM-MACH model, that encompassed the entire domain of the Canadian Arctic, and Arctic marine shipping activity (and emissions) at a 15 km × 15 km resolution. Shipping activity in the second scenario was obtained from the Canadian Coast Guard (CCG), which included all commercial marine vessels within the Canadian Arctic Domain. While shipping emissions were obtained through

SNC-Lavalin Environment’s (2012) Marine Emission Inventory Tool, baseline emissions inventory were obtained from the CCG for the year 2010. The domain coverage in Gong 98

et al. (2018) includes all regions of the Canadian Arctic, while those from Moran et al.,

(2016) had spatial coverage up to ~80.5⁰N. Sites outside of this domain (n = 121) had the nearest neighbor interpolation of the closest sites and was given the average deposition value of the three nearest neighbor.

Sites outside of the Moran et al. (2016) domain (n = 102) were given Sdep values of the nearest grid. Values of Sdep for each study site were extracted from both models and were used to calculate Ex:

[13] 퐸푥 = 푆푑푒푝 − 퐶퐿(퐴푐)

3.4 Results

3.4.1 Critical Loads of Acidity

The ANClimit was influenced by site specific DOC concentrations, which ranged from 0.01 (Baffin Is.; Chapter 2) to 80.1 mg·L-1 (Northwest Territories; Rühland et al.,

-1 2003), with a mean DOC concentration of 4.7 mg L . Mean DOC adjusted ANClimit values were 25.6 µeq·L-1 (Brown Trout) and 28.2 µeq·L-1 (Arctic Char) and were similar

-1 to the widely used ANClimit of 20 µeq·L (Lydersen et al. 2004). This resulted in similar

CL(A) values (Table 3.2; Figure 3.2) among geological classes (Table 3.3) and a similar proportion of acid sensitivity classes (Table 3.4) among the ANClimit values. Overall, estimated CL(A) were low for the Canadian Arctic, with the most sensitive CL(A) values

―2 ―1 ―2 ―1 for Arctic Char ANClimit (mean = 96.2 meq·m ·yr , median = 35.8 meq·m ·yr )

(Table 3.2). Given the importance and vast distribution of Arctic char in the Canadian

Arctic (Scott and Crossman, 1973) and their importance role for Arctic communities

(provides economic and subsistence opportunities) the determination of CL(A) in this

study will primarily focus on the ANClimit for Arctic char (see supporting information for other ANClimits). Large portions of sites were classified as Highly Sensitive (40%, n =

455) and Very Low Sensitivity (27.6%, n = 314) (Table 3.4). The remaining sites were classified as follow: 12.0% as Sensitive, 9.3% as Moderately Sensitive, and 11.2% as Low

Sensitivity (Table 3.4).

Table 3.2 Selected statistics (mean, standard deviation, minimum, maximum, median, 5th th percentile, and 95 percentile) of CL(A) for the three ANClimits. Where ANClimit= 8 µeq·L–1 for Brown Trout, 11 µeq·L–1 for Arctic Char, and 20 µeq·L–1 for Ecosystem. Percentile th th ANClimit Count Mean SD Min Max Median 5 95 Brown Trout 1138 96.5 193 0.00 4393 36.4 0.4 384 Arctic Char 1138 96.2 193 0.00 4393 35.8 0.4 384 Ecosystem 1138 96.9 194 0.00 4394 35.8 0.4 390

Figure 3.2 Cumulative frequency distribution for critical loads of acidity for the three –1 –1 ANClimits values. Where ANClimit= 8 µeq·L for Brown Trout, 11 µeq·L for Arctic Char, and 20 µeq·L–1 for Ecosystem. Table 3.3 Site count and mean CL(A) (median) values in meq·m–2·yr–1 per geological –1 type and among the three ANClimits values. Where ANClimit= 8 µeq·L for Brown Trout, 11 µeq·L–1 for Arctic Char, and 20 µeq·L–1 for Ecosystem. Site Brown Arctic Geology type Ecosystem Count Trout Char Igneous 313 48.5 (22.1) 48.1 (21.7) 48.6 (21.7) Sedimentary 738 108 (25.6) 107 (45.5) 108 46.5) Supracrustal 75 170 (58/6) 170 (58.6) 171 (58.6) Unclassified 12 206 (73.0) 206 (72.8) 206 (72.8)

100

Table 3.4 Percentage (and count) of sites within different acid sensitivity classes (meq·m– 2 –1 –1 ·yr ) among the three ANClimits values. Where ANClimit= 8 µeq·L for Brown Trout, 11 µeq·L–1 for Arctic Char, and 20 µeq·L–1 for Ecosystem. Acid sensitivity Brown Trout Arctic Char Ecosystem <20 40% (455) 40.0% (455) 39.8% (453) 20–40 11.8% (134) 12.0% (136) 11.8% (134) 40–60 9.2% (105) 9.3% (106) 9.4% (107) 60–100 11.3% (129) 11.2% (127) 11.2% (128) >100 27.7% (315) 27.6% (314) 27.8% (316)

Regions with a larger percentage of sites classified as Highly Sensitive are

Melville Is. (97.9%, n = 47), King William Is. (75%, n = 3) and Ellef Ringnes Is. (72%, n

= 18), while regions with more percentage of sites classified as Very Low Sensitivity are

Coats Is. (100%, n = 10), Southampton Is. (91.4, n = 32) and the Yukon (85.0%, n = 43)

(Table 3.5). Larger mean CL(A) were found for the Yukon (435 meq·m―2·yr―1, n = 40),

Ellesmere Is. (262 meq·m―2·yr―1, n = 143) and Southampton Is. (252 meq·m―2·yr―1, n =

35), while the lowest CL(A) values were found for Melville Is. (5.53 meq·m―2·yr―1, n =

48), Banks Is. (18.4 meq·m―2·yr―1, n = 45) and Bylot Is. (20.42 meq·m―2·yr―1, n = 36)

(Table 3.6). A cluster of Highly Sensitive sites are situated in Northwest Territories,

Nunavut, Melville Is., parts of Banks Is., and scattered amongst Victoria Is., Baffin Is.,

Bathurst Is., Axel Heiberg Is., and Ellesmere Is (Figure 3.3). In contrast, clusters of Very

Low Sensitivity can be seen in Southern Southampton Is., Yukon Territory, Cornwallis Is., northern Axel Heiberg, and northern Ellesmere Is. (Figure 3.3).

101

Table 3.5 Percentage (and number) of sites (under Arctic Char ANClimit) within different acid sensitivity classes (values are expressed in meq·m―2·yr―1) among different regions of the Canadian Arctic. Region <20 20–40 40–60 60–100 >100 Axel Heiberg Is. 55.3% (26) 10.6% (5) 12.8% (6) 12.8% (6) 8.5% (4) Baffin Is. 18.4% (23) 11.2% (14) 13.6% (17) 19.2% (24) 37.6% (47) Banks Is. 68.9% (31) 20% (9) 4.4% (2) 6.7% (3) 0% (0) Bathurst Is. 19.4% (13) 11.9% (8) 20.9% (14) 29.9% (20) 17.9% (12) Bylot Is. 69.4% (25) 16.7% (6) 8.3% (3) 2.8% (1) 2.8% (1) Coats Is. 0% (0) 0% (0) 0% (0) 0% (0) 100% (10) Cornwallis Is. 6.8% (3) 4.5% (2) 9.1% (4) 50% (22) 29.5% (13) Crozier Is. 0% (0) 0% (0) 100% (2) 0% (0) 0% (0) Devon Is. 42.5% (17) 37.5% (15) 5% (2) 5% (2) 10% (4) Ellef Ringnes Is. 72% (18) 12% (3) 0% (0) 4% (1) 12% (3) Ellesmere Is. 16.1% (23) 6.3% (9) 4.9% (7) 3.5% (5) 69.2% (99) King William Is. 75% (3) 0% (0) 0% (0) 0% (0) 25% (1) Little Cornwallis Is. 100% (1) 0% (0) 0% (0) 0% (0) 0% (0) Melville Is. 97.9% (47) 2.1% (1) 0% (0) 0% (0) 0% (0) Northwest Territories 61.5% (59) 9.4% (9) 5.2% (5) 9.4% (9) 14.6% (14) Nunavut 50.3% (92) 20.2% (37) 14.2% (26) 11.5% (21) 3.8% (7) Prince Charles Is. 0% (0) 0% (0) 20% (1) 20% (1) 60% (3) Prince of Wales Is. 0% (0) 40% (2) 40% (2) 20% (1) 0% (0) Prince Patrick Is. 45.7% (16) 34.3% (12) 20% (7) 0% (0) 0% (0) Somerset Is. 15.4% (2) 23.1% (3) 46.2% (6) 7.7% (1) 7.7% (1) Southampton Is. 2.9% (1) 2.9% (1) 0% (0) 2.9% (1) 91.4% (32) Victoria Is. 59.6% (53) 0% (0) 1.1% (1) 6.7% (6) 32.6% (29) Yukon 5% (2) 0% (0) 2.5% (1) 7.5% (3) 85% (34)

102

–1 Figure 3.3 Critical loads of Acidity for 1138 sites using an ANClimit for Arctic Char (11 µeq·L ).

103

―2 ―1 Table 3.6 Number of sites mean CL(A) (calculated with Arctic Char ANClimit; meq·m ·yr ), mean sulphur deposition (meq·m–2·yr–1) from both model scenarios, and exceedances (percentage of total sites and number of sites under both models) for each region of the Canadian Arctic. No. of CL(A) Deposition Model Exceedances With Without With Without Region sites Arctic Char shipping shipping shipping shipping Axel Heiberg Is. 47 43.9 1.72 1.59 9% (4) 6% (3) Baffin Is. 125 99.2 2.27 3.38 9% (11) 7% (9) Banks Is. 45 18.4 1.37 1.56 9% (4) 9% (4) Bathurst Is. 67 61.8 2.85 1.63 10% (7) 9% (6) Bylot Is. 36 20.4 2.32 1.27 8% (3) 8% (3) Coats Is. 10 228 1.39 3.15 0% (0) 0% (0) Cornwallis Is. 44 86.5 2.05 1.49 5% (2) 0% (0) Devon Is. 40 32.5 2.49 1.25 3% (1) 0% (0) Ellef Ringnes Is. 25 32.8 4.40 2.31 60% (15) 48% (12) Ellesmere Is. 143 262 4.20 0.63 3% (4) 0% (0) King William Is. 4 30.4 3.39 1.90 25% (1) 0% (0) Melville Is. 48 5.50 2.93 1.66 44% (21) 38% (18) Northwest Territories 96 46.3 2.28 3.46 24% (23) 26% (25) Nunavut 183 30.1 2.99 3.01 16% (29) 17% (31) Prince Charles Is. 5 103 3.97 1.63 0% (0) 0% (0) Prince of Wales Is. 5 43.3 3.15 1.55 0% (0) 0% (0) Prince Patrick Is. 35 24.8 4.86 1.70 3% (1) 0% (0) Somerset Is. 13 48.5 3.71 1.62 8% (1) 8% (1) Southampton Is. 35 252 1.16 3.06 0% (0) 0% (0) Victoria Is. 89 53.8 1.93 2.09 13% (12) 24% (21) Yukon 40 435 2.56 1.17 5% (2) 5% (2) 12.5% (142) 12.0% (136)

104

3.4.2 Sulphur deposition and Exceedances

Modelled sulphur deposition across the Arctic for 2010 from both with shipping

(Gong et al., 2018) and non-shipping (Moran et al., 2016) scenarios were similar, with a mean value of 2.74 meq·m―2·yr―1 (median = 2.12 meq·m―2·yr―1) with shipping, and

2.13 meq·m―2·yr―1 (median = 2.12 meq·m―2·yr―1) the without shipping (Table 3.7).

However, the mean deposition differed among regions, with higher values for with shipping for Prince Patrick = 4.86 meq·m―2·yr―1, Ellef Ringnes Is. = 4.40 meq·m―2·yr―1, and Ellesmere Is.= 4.20 meq·m―2·yr―1 (Table 3.6). Higher values for without shipping are found on Northwest Territories = 3.46 meq·m―2·yr―1, Baffin Is.

3.38 meq·m―2·yr―1 and Coats Is. = 3.15 meq·m―2·yr―1 (Table 3.6).

The total number of exceeded sites (with CL(A) produced with the Arctic Char

ANClimit) was 142 (12.5%) under with shipping, and 136 (12%) under without shipping deposition scenarios (Table 3.6). The percentage of exceeded sites was consistent under the other ANClimit values (see Supporting Information Table B3, B4, and Figure B3, B4,

B5, B6). The percentage of exceeded sites was consistent among regions for both deposition models; Ellef Ringnes Is. (With shipping = 60%, n = 15; without shipping =

48%, n = 12), Melville Is. (With shipping = 44%, n = 21; without shipping) = 38%, n =

18), and Northwest Territories With shipping = 24%, n = 23; without shipping = 26%, n =

25) (Table 3.6). A similar spatial pattern of exceeded was observed for both with shipping

(Figure 3.4), and without shipping (Figure 3.5), with major clusters found on the central mainland (between Northwest Territories and Nunavut), Melville Is., parts of Bathurst Is.,

Axel Heiberg Is., Ellef Ringnes Is., Ellesmere Is., and scattered amongst Victoria Is.,

105

Baffin Is., Banks, and Cornwallis Is. Exceedances values with shipping (Figure 3.4), were elevated in a few sites; in Northwest Territories (n = 2) and Baffin Is. (n = 4) (Figure 3.5).

Table 3.7 Selected statistics (Mean, standard deviation (SD), minimum, maximum, median, 5th percentile, and 95th percentile) of total sulphur deposition values (meq·m―2·yr―1) for both scenarios.

Percentile th th Sdep Models Count Mean SD Min Max Median 5 95 With shipping 1138 2.74 1.6 0.75 8.03 2.12 1.02 5.23 Without shipping 1138 2.13 1.18 0.07 5.89 1.91 0.2 4.25

106

Figure 3.4 Critical loads exceedances under 2010 sulphur deposition with shipping scenario.

107

Figure 3.5 Critical loads exceedances under 2010 sulphur deposition with shipping scenario.

108

3.5 Discussion

The SSWC model (Henriksen et al., 1992; Henriksen and Posch, 2001) is well- established and widely used (Fennoscandia, Henriksen et al., 1998; Ontario, Henriksen et al., 2002; Eastern Canada, Dupont et al., 2005; Manitoba, Jeffries et al., 2010;

Saskatchewan, Cathcart et al., 2016) for assessing the acid-sensitivity of aquatic ecosystems. The model is based on the assumption of an equilibrium environment (steady state) though simple titration, i.e., incoming acidity must be neutralized by base cations weathered from the surrounding catchment. In the current study, the application of the

SSWC model followed that of Lydersen et al. (2004) to account for the influence of organic acids, and that of Lien et al. (1995) and Závodský et al. (1995) to account for the influence of geological sulphate. In general, the acid sensitivity of lakes in the Canadian

Arctic resulted in lower critical loads (median = 35.8 meq·m―2·yr―1) when compared to those found by AMAP (2006) for lakes in the Canadian Arctic region (median = 127 meq·m―2·yr―1). Factors such as more sites (this study = 1138 vs. AMAP (2006) = 251) and the coverage of surface waters from regions with acid-sensitive geology, i.e., Baffin

Is, may contribute to the lower CL(A) summary values. A small portion of sites (3.5%, n

= 40) had a CL(A) value of 0 meq·m―2·yr―1, which indicates that these systems have very limited buffering capacity against anthropogenic acidic deposition. Although these systems are natural, it is most likely that they are unproductive, and support limited aquatic biota (see Hershey et al., 2005).

109

Countries with similar Arctic climatic conditions on the Fennoscandia Peninsula, had lower mean values when compared to this study; Finland (63.0 meq·m―2·yr―1, n =

1450), Norway (56.0 meq·m―2·yr―1, n = 4018), and Sweden (61.0 meq·m―2·yr―1, n =

1005) (Henriksen et al., 1998), or was similar in percentages; CL(A) <20 meq·m―2·yr―1 in Kola peninsula = 50% (n ≥ 185) (Moiseenko, 1994), vs this study 40.0% (n = 455).

Other regions of the Arctic, Svalbard and Iceland, had less sensitive values when compared to this study: [This study] 27.6% of site with ≥ 100 meq·m―2·yr―1, Svalbard

67% at ≥ 100 meq·m―2·yr―1 (Lien et al., 1995); [This study] 50th percentile = 35.8 meq·m―2·yr―1; Iceland: 693 meq·m―2·yr―1 (European Environment Agency, 2010), which can be attributed to the lack of coverage of this study.

Critical loads of acidity are determined by catchment processes, such as the weathering of bedrock to produce base cations, which enter the lakes/ponds via runoff

(Henriksen et al., 1992; AMAP 1998; Forsius et al., 2001). Therefore, the water chemistry of surface waters on the Pre-Cambrian shield (igneous bedrock: granite, granodiorite, quartz, monzonite, quartz diorite, diorite) are more acid sensitive than those on geology with higher carbonate content (limestone, dolostone, shale, evaporites, chalk; carbonate reefs) (AMAP 1998; Forsius et al., 2001). For example, sites at similar latitudes but overlying different geology, such as those on mainland Northwest Territories (mean latitude: 63.477; n = 96) and Southampton Is. (mean latitude: 63.971; n = 35), have very different hydrochemical characteristics. In general, surface waters situated mainly on the

Pre-Cambrian shield have acid sensitive hydrochemical characteristics, e.g., more dilute

(93.5 µS·cm-1) with low concentrations of base cations (mean Ca = 12.7 mg·L-1, Mg =

4.30 mg·L-1) and DIC (mean = 9.00 mg·L-1) (Moser et al, 1993; Rühland and Smol,

110

1998; Rühland et al., 2003; Chapter 2; Table 3.1), while those over sedimentary limestone geology tend to have insensitive hydrochemical characteristics, e.g., more concentrated

(261 µS·cm-1) with higher concentrations of base cations (mean Ca = 29.9 mg·L-1 Mg =

7.0 mg·L-1) and DIC (mean = 19.0 mg·L-1) (Mallory et al., 2006; Hamilton et al., 2010;

Table 3.1). This resulted in higher mean critical loads of acidity for Southampton Is. (252 meq·m―2·yr―1) compared with Northwest Territories (46.3 meq·m―2·yr―1) (Table 3.6).

In the Canadian Arctic, the mean and median CL(A) for surface waters underlain by igneous geology were n = 48.1 meq·m―2·yr―1and 21.7 meq·m―2·yr―1 (n = 313), respectively (Table 3.3) and were comparable to other regions underlain by igneous geology across Canada [northern Saskatchewan = 12.3–32.5 meq·m―2·yr―1 (Jeffries et al., 2010; Scott et al., 2010; Whitfield et al., 2016; Cathcart et al., 2016), northern

Manitoba = 17.6 meq·m―2·yr―1 (Jeffries et al., 2010), eastern Ontario = 65.1–102.8 meq·m―2·yr―1 (Henriksen et al., 2002), Québec = 80.2 meq·m―2·yr―1, Nova Scotia =

37.5 meq·m―2·yr―1, New Brunswick = 111 meq·m―2·yr―1, and Newfoundland Labrador

= 58.5 meq·m―2·yr―1 (Dupont et al., 2005; Whitfield et al., 2006)]. Higher mean and median CL(A) values were primarily associated sedimentary geology (mean = 107.4 meq·m―2·yr―1; median = 45.4 meq·m―2·yr―1; n = 738). These values were much lower than values found in other regions underlain by sedimentary geology [Southern Québec =

436.7 meq·m―2·yr―1 (Dupont et al., 2005), and Southern Ontario: 424 meq·m―2·yr―1

(Henriksen et al., 2002)] in Canada. Similar segregation of CL(A) values driven by geology were found in other Arctic regions, with lower CL(A) values on weathering resistant igneous geology (Moiseenko, 1994; Henriksen et al., 1998), such as those of the of the Fennoscandia Shield (Lahtinen 2012), and higher values found on sedimentary

111

rock, such as Svalbard (Lien et al., 1995), and in Iceland (European Environment Agency,

2010).

Climate strongly mediates geochemical weathering; low temperatures and low precipitation are common climatic conditions of the Arctic (Maxwell, 1981), which reduces (a) the rate of biogeochemical processes (such as the weathering of bedrock to produce base cations), and (b) runoff (reduces the flux of base cations into lakes). An example is the different runoff values between the Arctic sedimentary region (mean =

0.09 m·yr―1, median = 0.06 m·yr―1) and that of other sedimentary regions of Canada;

Ontario = 0.4 – 0.5 m·yr―1 (Henriksen et al., 2002) and Québec = 0.4 – 0.9 m·yr―1

(Dupont et al., 2005). Which produced much higher CL(A) values (Ontario: 424 meq·m―2·yr―1 (Henriksen et al., 2002) and Québec = 437 meq·m―2·yr―1 (Dupont et al.,

2005) than that of the sedimentary region of the Canadian Arctic (mean = 107.4 meq·m―2·yr―1). This can be attributed to the increased biogeochemical process associated with more wet and warmer climatic conditions commonly found in southern

Canada. While mean runoff for igneous (0.138 m·yr―1) was higher for than that of sedimentary geology (0.06 m·yr―1), it did not result in higher critical loads values for igneous (Table 3.3). This suggest that CL(A) are primarily affected by geological sensitivity to acidity, e.g., limestone = low sensitivity, igneous = high sensitivity; the production of base cations, and secondly, runoff, where higher catchment runoff allows for higher input of base cations (and deposition loads).

The SSWC model assumes that N deposition is retained in the catchment and does not contribute to acidification (Henriksen and Posch, 2001). As such, N deposition was excluded from both critical loads and exceedance calculations. This is supported by the 112

low concentration of nitrate and nitrite (mean = 0.02 mg·L―1; Chapter 2) in Arctic surface waters. Deposition of S predicted by the Moran et al. (2016) model (Without shipping) was generally higher in lower latitude sites than the Gong et al. (2018) model

(With shipping) (Table 3.6; supporting information Figure B7). Generally, both models produced similar S deposition values (With shipping = 2.74 meq·m―2·yr―1, without shipping = 2.13 meq·m―2·yr―1; Table 3.7), resulting in similar exceedance percentage of

12.5% (n = 142) of sites for the with shipping, and 12.0% (n = 136) for the without shipping scenarios (Table 3.6). It is difficult to compare the exceedance percentage of other studies (Lien et al., 1995; Henriksen et al., 2002; Dupont et al., 2005; Scott et al.,

2010; Cathcart et al., 2010), as emissions of sulphur (and deposition) in Canada have been reduced greatly since 1990 (3066.5 kt) when compared to the 2010 emissions of

1371.4 kt (Environment and Climate Change Canada, 2017).

Decreasing trends in atmospheric sulphur concentrations observed at Arctic monitoring stations during the period of 1991–2000 have been attributed to the reduction of long-range transboundary air pollution policies among European and North American countries (Hole et al., 2009). However, there is growing concern regarding more localized sources of sulphur concentrations. As marine shipping is expected to increase in the

Arctic region, there is a growing concern regarding the impacts of ship-source S (Corbett et al., 2010). To regulate ship emissions, Canada have adopted the United Nations

International Maritime Organization (IMO)’s 1973 International Convention for the

Prevention of Pollution from Ships (MARPOL). Currently, under Tier II amendments, S limits in fuels are regulated to 3.5% (3,500 mg·kg-1) for regions outside the Emission

Control Areas (ECAs) and will be expected to be capped to 0.5% (500 mg·kg-1) by 2020

113

(International Maritime Organization, n.d). Regions under ECAs, S limits in fuels are regulated to 0.1% (100 mg·kg) for emission control areas (International Maritime

Organization, n.d). Currently, the Arctic is not under an ECA, but the establishment of one is currently being discussed.

Currently (for the year 2010) it is estimated that shipping contributes to 20–100% of ambient SO2 concentrations within shipping routes in the Arctic (Gong et al., 2018).

Under business-as-usual scenario (BAU) for the year 2030, it is expected that shipping will continue to contribute to higher ambient SO2 concentrations (~32% higher than

2010), with greater contributions (as much as 57%) in regions along shipping routes

(Gong et al., 2018), assuming that the IMO sulphur content of 0.5% is established (After the 1st of January, 2020; International Maritime Organization, n.d).) In addition, under

BAU, total sulphur deposition will increase, with values up to 20% along the coast of

Baffin Bay (Gong et al., 2018). For deposition, current estimates of ship contribution are

<5.0% in the Canadian Arctic (Gong et al., 2018). Similarly, to ambient SO2 concentrations, sulphur deposition is expected to increase up to 20% under BAU scenario for 2030. Approximately 29.6% of all Baffin Island sites are considered to be Highly

Sensitive or Sensitive to acidic deposition (Figure 3.3), with 7–9% of sites exceeded under with and without shipping scenarios (Table 3.6). Sites in between the coast of

Nunavut and Victoria Is. are also at risk, as this area is a major channel for the NWP.

Another route for the NWP is along the southern coast of Melville Is., where, it was found to have the greatest proportion of Highly Sensitive (<20 meq·m―2·yr―1) (Table

3.5) and high amounts of exceeded (Table 3.6) sites. Although it was found that modelled shipping emissions contributed to <5% of sulphur deposition to Arctic systems, the

114

establishment of an ECA within the Arctic would greatly protect these sites, especially for those with very low CL(A) values and/or those situated along major shipping routes

(Figure 3.4 and 3.5). Under an ECA for the year 2030, shipping contribution to ambient

SO2 concentrations and total sulphur deposition was found to drop to under 2010 baseline values (Gong et al., 2018).

3.6 Conclusion

The SSWC model has been widely used across Canada to determine the sensitivity of surface waters to acidification and to support emission control policies.

Generally, results indicate that the Canadian Arctic is sensitive (i.e., <40 meq·m―2·yr―1) to acidic deposition (median CL(A) = 35.8 meq·m―2·yr―1) when compared with previous studies in the Canadian Arctic (AMAP, 2006). In the current study, 40% of sites (n =

455) were classified as very sensitive (< 20 meq·m―2·yr―1) to acidic deposition. The

CL(A) differed among regions of the Canadian Arctic, primarily to the two dominant geological types. Surface waters situated on igneous geology tend to have lower CL(A) values, i.e., median = 21.7 meq·m―2·yr―1 (mean = 48.1 meq·m―2·yr―1), while those on sedimentary geology, tend to have higher values, i.e., median = 45.4 meq·m―2·yr―1

(mean = 107.4 meq·m―2·yr―1). Exceedance of CL(A) under modelled S deposition in

2010 was approximately 12% of all sites (n = 136–142), with those on Melville Is., mainland Nunavut, and Baffin Is. most at risk due to their location near the NWP. The establishment of an ECA would help to protect these sites from acidification, as it is expected that shipping will increase in the Canadian Arctic. In this study, the long-term steady-state condition was estimated using current available hydrochemical data to identify sites that are at risk of acidification. However, this does not provide a timeline to 115

when damage is expected to occur. Thus, calculated exceedances are not associated with immediate ecosystem damage, and damage may occur over an extended period of time.

Additional lake surveys among expected ship-transit routes, should be implemented to further assess acid sensitive regions (as described in this study) and to support the establishment an ECA or emissions reduction policies.

Acknowledgments

This study was funded by the Natural Science and Engineering Research Council

(NSERC) grant awarded to Julian Aherne, as well as the McLean foundation, Northern

Program (PCSP). Thanks to Max Posch for reviewing the modified SSWC formulation, and to Scott Fleming and Dr. Peter Lafleur for their assistance in the field, and to Hazel

Cathcart for her GIS and data mining assistance.

116

References Aherne, J. (2008). Calculating Critical Loads of Acid Deposition for Forest Soils in Alberta: Critical Load, Exceedance and Limitations. Final Report. Canadian Council of Ministers of the Environment. 1-19 Aliabadi, A.A., Staebler, R.M., & Sharma, S. (2015). Air quality monitoring in communities of the Canadian Arctic during the high shipping season with a focus on local and marine pollution. Atmospheric chemistry and physics 15(5),2651– 2673.DOI: 10.5194/acp-15-2651-2015 Arctic Monitoring and Assessment Programme. (1998). AMAP Assessment Report: Arctic Pollution Issues, Arctic Monitoring and Assessment Programme AMAP, Oslo, Norway. 859 p. Arctic Monitoring and Assessment Programme. (2006). AMAP Assessment Report: Acidifying Pollutants, Arctic Haze, and Acidification in the Arctic, Arctic Monitoring and Assessment Programme AMAP, Oslo, Norway. 859p Arctic Marine Shipping Assessment. (2009). Arctic Marine Shipping Assessment 2009 Report. Arctic Council, second printing. 194p Antoniades, D., Douglas, M. S., & Smol, J. P. (2003). The physical and chemical limnology of 24 ponds and one lake from Isachsen, Ellef Ringnes Island, Canadian High Arctic. International Review of Hydrobiology, 88(5), 519-538.DOI: 10.1002/iroh.200310665 Bonten, L.T., Reinds, G.J. and Posch, M. (2016). A model to calculate effects of atmospheric deposition on soil acidification, eutrophication and carbon sequestration. Environmental Modelling & Software, 79, 75-84. DOI: 10.1016/j.envsoft.2016.01.009 Beamish, R. J. (1974). Loss of fish populations from unexploited remote lakes in Ontario, Canada as a consequence of atmospheric fallout of acid. Water Research, 8(1), 85– 95.DOI: 10.1016/0043-1354(74)90012-8 Brakke, D., A. Henriksen & S. Norton. (1990). A variable F-factor to explain changes in base cation concentrations as a function of strong acid deposition. Internationale Vereinigung für theoretische und angewandte Limnologie: Verhandlungen, 24(1), 146-149.DOI: 10.1080/03680770.1989.11898707 British Columbia Ministry of Environment. (2016). Prince Rupert Airshed Study: Summary. British Columbia Ministry of Environment, Environmental Protection Division, Regional Operations Branch. Available Online: https://www2.gov.bc.ca/assets/gov/environment/air-land-water/air/reports-pub/pr- airshed-study-report-summ.pdf

117

Cathcart, H., Aherne, J., Jeffries, D. S. & Scott, K. A. (2016). Critical loads of acidity for 90,000 lakes in northern Saskatchewan: A novel approach for mapping regional sensitivity to acidic deposition. Atmospheric Environment, 146, 290–299.DOI: 10.1016/j.atmosenv.2016.08.048 Canadian Council of Ministers of the Environment. (1998). The Canada-wide Acid Rain Strategy for Post-2000: Strategy and Support Documents. Federal/Provincial/Territorial Ministers of Energy and Environment. Halifax, Nova Scotia, p 20 Canadian Council of Ministers of the Environment. (2013). 2010 – 2011 Progress Report on The Canada-wide Acid Rain Strategy for Post-2000. Canadian Council of Ministers of the Environment, Winnipeg, Manitoba, p 41 Corbett, J.J., Lack, D.A., Winebrake, J.J., Harder, S., Silberman, J.A., & Gold, M. (2010). Arctic shipping emissions inventories and future scenarios. Atmosphere Chemistry Physics. 10(19), 9689–9704. DOI: 10.5194/acp-10-9689-2010 Dawson, J., Copland, L., Mussells, O., & Carter, N. (2017). Shipping Trends in Nunavut 1990-2015: A report prepared for the Nunavut General Monitoring Program. Ottawa, Canada and Iqaluit, Nunavut.68p Dupont, J., Clair, T. A., Gagnon, C., Jeffries, D. S., Kahl, J. S., Nelson, S. J., & Peckenham, J. M. (2005). Estimation of critical loads of acidity for lakes in northeastern United States and eastern Canada. Environmental monitoring and assessment, 109(1–3), 275–92.DOI: 10.1007/s10661-005-6286-x Eckhardt, S., Hermansen, O., Grythe, H., Fiebig, M., Stebel, K., Cassiani, M., Baecklund, A. & Stohl, A. (2013). The influence of cruise ship emissions on air pollution in Svalbard – a harbinger of a more polluted Arctic?. Atmospheric Chemistry and Physics. 13(16), 8401–8409.DOI: 10.5194/acp-13-8401-2013 Environment and Climate Change Canada (2017) Canadian Environmental Sustainability Indicators: Air pollutant emissions. Consulted on Month day, year. Available at: www.ec.gc.ca/indicateurs-indicators/default.asp?lang=en&n=E79F4C12-1. European Environment Agency EEA. (2010). Freshwater – State and impacts Iceland, SOER Common environmental theme from Iceland. Copenhagen, Demark. Available Online: https://www.eea.europa.eu/soer/countries/is/freshwater-state-and- impacts-iceland/#ecological-quality Forsius, M., Posch, M., Aherne, J., Reinds, G. J., Christensen, J., & Hole, L. (2010). Assessing the Impacts of Long-Range Sulfur and Nitrogen Deposition on Arctic and Sub-Arctic Ecosystems. Ambio, 39(2), 136–147.DOI: 10.1007/s13280-010- 0022-7

118

Fulton, R.J. (1989). Quaternary Geology of Canada and Greenland. Geological Survey of Canada, Geology of Canada Series no. 839p Gong, W., Beagley, S.R., Cousineau, S., Sassi, M., Munoz-alpizar, R., Ménard, S., Racine, J., Zhang, J., Chen, J., Morrison, H., Sharma, S., Huang, L., Bellavance, P., Ly, J., Izdebski, P., Lyons, L., & Holt, R. (2018). Assessing the impact of shipping emissions on air pollution in the Canadian Arctic and northern regions: current and future modelled scenarios. Atmospheric Chemistry and Physics. Hamilton, P.B., Gajewski, K., Atkinson, D.E., & Lean, D.R.S. (2001). Physical and Chemical limnology of 204 lakes from the Canadian Arctic Archipelago. Hydrobiogia. 457: 133–148 Henriksen, A., Kämäri, J., Posch, M., and Wilander, A. (1992). Critical loads of acidity: Nordic surface waters. Ambio, 356–363. Henriksen, A., Skjelvåle, B.L., Mannio, J., Wilander, A., Harriman, R., Curtis, C., Jensen, J.P., Fjeld, E. & Moiseenko, T. (1998). Northern European lake survey, 1995: Finland, Norway, Sweden, Denmark, Russian Kola, Russian Karelia, Scotland and Wales. Ambio, 80–91 Henriksen, A., and Posch, M. (2001). Steady-state models for calculating critical loads of acidity for surface waters. Water, Air, and Soil Pollution: Focus. 7(1-2),375–98 Henriksen, A., Dillon, P. J., & Aherne, J. (2002). Critical loads of acidity for surface waters in south-central Ontario, Canada: regional application of the Steady-State Water Chemistry SSWC model. Canadian Journal of Fisheries and Aquatic Sciences. 59(8), 1287–1295. Hershey, A.E., Beaty, S., Fortino, K., Keyse, M., Mou, P.P., O'BRIEN, W.J., Ulseth, A.J., Gettel, G.A., Lienesch, P.W., Luecke, C., & McDonald, M.E. (2006). Effect of landscape factors on fish distribution in arctic Alaskan lakes. Freshwater Biology, 51(1):39-55. Hesthagen, T., Sevaldrud, I. H., & Berger, H. M. (1999). Assessment of damage to fish population in Norwegian lakes due to acidification. Ambio, 28(2),112–117. Hettelingh, J.-P., Posch, M., Velders, J. M., Ruyssenaars, P., Adam, M., de Leeuw, F., Lükewille, A., Maas, R., Sliggers, J., & Slootweg, J. (2013). Assessing interim objectives for acidification, eutrophication and ground-level ozone of the EU National Emissions Ceilings Directive with 2001 and 2012 knowledge. Atmospheric Environment, 75, 129–140. Hole, L.R., Christensen, J.H., Ruoho-Airola, T., Tørseth, K., Ginzburg, V., & Glowacki, P. (2009). Past and future trends in concentrations of sulphur and nitrogen compounds in the Arctic. Atmospheric Environment, 43(4), 928-939.

119

International Maritime Organization. (n.d). Sulphur oxides SOx and Particulate Matter PM – Regulation 14. International Maritime Organization IMO. Available Online: http://www.imo.org/en/OurWork/Environment/PollutionPrevention/AirPollution/Pa ges/Sulphur-oxides-SOx-%E2%80%93-Regulation-14.aspx Jeffries, D. S., Semkin, R. G., Gibson, J. J., & Wong, I. (2010). Recently surveyed lakes in northern Manitoba and Saskatchewan, Canada: characteristics and critical loads of acidity. Journal of Limnology, 69(1s),45–55. Joynt III, E.H. & Wolfe, A.P. (2001). Paleoenvironmental inference models from sediment diatom assemblages in Baffin Island lakes Nunavut, Canada and reconstruction of summer water temperature. Canadian Journal of Fisheries and Aquatic Sciences. 58(6), 1222–1243. Kwok, R., & Rothrock, D.A. (2009). Decline in Arctic sea ice thickness from submarine and ICESat records: 1958–2008.Geophysical Research Letters.36(15),1–5 Lahtinen, R. (2012). Main geological features of Fennoscandia. Geological Survey of Finland, Special Paper, 53, 13-18. Lamarque, J.-F. and Dentener, F. and McConnell, J. and Ro, C.-U. and Shaw, M. and Vet, R. and Bergmann, D. and Cameron-Smith, P. and Dalsoren, S. and Doherty, R. and Faluvegi, G. and Ghan, S. J. and Josse, B. and Lee, Y. H. and MacKenzie, I. A. and Plummer, D. and Shindell, D. T. and Skeie, R. B. and Stevenson, D. S. and Strode, S. and Zeng, G. and Curran, M. and Dahl-Jensen, D. and Das, S. and Fritzsche, D. and Nolan, M. (2013). Multi-model mean nitrogen and sulfur deposition from the Atmospheric Chemistry and Climate Model Intercomparison Project ACCMIP: evaluation of historical and projected future changes. Atmospheric Chemistry and Physics. 1316:7997-8018. Doi: 10.5194/acp-13-7997-2013. Leivestad, H., & Muniz, I. P. (1976). Fish kill at low pH in a Norwegian river. Nature. 259(5542), 391–392. Lien, L., Henriksen, A., & Traaen, T. S. (1995). Critical loads of acidity to surface waters: Svalbard. Science of the Total Environment, 160, 703–713. Lien, L., Raddum, G. G., Fjellheim, A., & Henriksen, A. (1996). A critical limit for acid neutralizing capacity in Norwegian surface waters, based on new analyses of fish and invertebrate responses. Science of the Total Environment, 177(1–3), 173–193. Lim, S.S.D., & Douglas, M.S.V. (2003). Limnological Characteristics of 22 Lakes and Ponds in the Haughton Crater Regions of Devon Island, Nunavut, Canadian High Arctic. Arctic, Antarctic, and Alpine Research. 35(4),509–519 Lim, S.S.D., Douglas, M.S.V., & Smol, P.J. (2005). Limnology of 46 lakes and ponds on Banks Island, N.W.T., Canadian Arctic Archipelago. Hydrobiologia. 545(1), 11–32

120

Lydersen, E., Larssen, T., & Fjeld, E. (2004). The influence of total organic carbon (TOC) on the relationship between acid neutralizing capacity (ANC) and fish status in Norwegian lakes. Science of the Total Environment. 326: 63–69. DOI:10.1016/j.scitotenv.2003.12.005. Lyman, J., & Fleming, R. H. (1940). Composition of sea water. Journal of Marine Research. 3:134–146. Mallory, M.L., Fontaine, A.J., Smith, P.A., Robertson, M.O.W., & Gilchrist, H.G. (2006). Water chemistry of ponds on Southampton Is., Nunavut, Canada: Effects of habitat and ornithogenic inputs. Arch. Fur Hydrobiol. 166: 411–432. doi:10.1127/0003- 9136/2006/0166-0411. Maxwell, J.B. (1981). Climatic regions of the Canadian Arctic islands. Arctic, 225–240. McNeely, R.N., Dwyer, L., & Neimanis, V.P. (1979). Water quality sourcebook: a guide to water quality parameters. Environment Canada, Inland Waters Directorate, Water Quality Branch, Ottawa, 89p Michelutti, N., Blais, J.M., Mallory, M.L., Brash, J., Thienpont, J., Kimpe, L.E., Douglas, M.S. & Smol, J.P. (2010). Trophic position influences the efficacy of seabirds as metal biovectors. Proceedings of the National Academy of Sciences, 107(23):10543–10548. Moiseenko, T. (1994). Acidification and critical loads in surface waters: Kola, Northern Russia. Ambio, 418–424. Moran, M.D., Lupu, A., Zhang, Z., Savic-Jovicic, V., & Gravel, S. (2016). A comprehensive performance evaluation of the next generation of the Canadian operational regional air quality deterministic prediction system. Proc 35th International Technical Meeting on Air Pollution modelling and its Application. 3-7 Oct., Chania, Crete Moser, K. A., Smol, J. P., Lean, D. R. S., & MacDonald, G. M. (1998). Physical and chemical limnology of northern boreal lakes, Wood Buffalo National Park, northern Alberta and the Northwest Territories, Canada. Hydrobiologia, 377(1–3), 25–43. New, M., Hulme, M., Jones, P.D. (2000). Global Monthly Climatology for the Twentieth Century Data set Oak Ridge National Laboratory Distributed Active Archive Center 2000 Retrieved from: http://daac.ornl.gov Nilsson, J., & Grennfelt, P. (1988). Critical Loads for Sulphur and Nitrogen: Report from a Workshop Held at Stockloster, Sweden, 19e24 March 1988. Nordic Council of Ministers. Paltan, Homero., Dash, J., & Edwards, M. (2015). A refined mapping of Arctic lakes using Landsat imagery. International Journal of Remote Sensing, 36(23), 5970– 5982 121

Perovich, D., Meier, W., Tschudi, M., Farrell, S., Gerland, S., Hendricks, S., Krumpen, T., & Haas, S. (2016). Sea Ice [In Arctic Report Card 2016], http://www.arctic.noaa.gov/Report-Card Pienitz, R., Smol, J.P., & Lean, D.R. (1997). Physical and chemical limnology of 59 lakes located between the southern Yukon and the Tuktoyaktuk Peninsula, Northwest Territories Canada. Canadian Journal of Fisheries and Aquatic Sciences, 54(2), .330–346. Pienitz R., Doran, P.T., & Lamoureux, S.F. (2008). Origin and geomorphology of lakes in the polar regions, in Polar Lakes and Rivers—Limnology of Arctic and Antarctic Aquatic Ecosystems eds. W.F. Vincent and J. Laybourn-Parry, Oxford University Press, Oxford, p 25–41. Pizzolato, L., Howell, S. E., Dawson, J., Laliberté, F., & Copland, L. (2016). The influence of declining sea ice on shipping activity in the Canadian Arctic. Geophysical Research Letters, 43(23). Rautio, M., Dufresne, F., Laurion, I., Bonilla, S., Vincent, W.F., & Christoffersen, K.S. (2011). Shallow freshwater ecosystems of the circumpolar Arctic. Ecoscience, 18(3), 204–222. Rühland, K., & Smol, J.P. (1998). Limnological characteristics of 70 lakes spanning Arctic treeline from Coronation Gulf to Great Slave Lake in the central Northwest Territories, Canada. International Review of Hydrobiology, 83(3),183–203. Rühland, K. M., Xiaowa, W., & Derek, C. G. (2003). Limnological characteristics of 56 lakes in the central Canadian arctic treeline region. Journal of Limnology, 62(1), 9- 27. Schindler, D.W., Mills, K.H., Malley, D.F., Findlay, D.L., Shearer, J.A., Davies, I.J., Turner, M.A., Linsey, G.A., & Cruikshank, D.R. (1985). Long-term ecosystem stress: the effects of years of experimental acidification on a small lake. Science, 22(8),1395–1402. Scott, W.B. & E.J., Crossman. (1973). Freshwater fishes of Canada. Fisheries Research Board of Canada, Bulletin 184, 1-966. Scott, K. A., B. Wissel, J. J. Gibson, & J. S. Birks. (2010). Chemical characteristics and acid sensitivity of boreal headwater lakes in northwest Saskatchewan. Journal of Limnology, 69(1s),33–44. Sliggers, J., & Kakebeeke, W. (2004). Clearing the air: 25 years of the convention on long-range transboundary air pollution. Geneva: United Nations, Economic Commission for Europe SNC-Lavalin Environment, 2012. (2010). National marine emission inventory for Canada. Final report prepared for 5 Environment Canada, 351 St. Joseph 122

Boulevard, Place Vincent Massey, Gatineau, Quebec. https://projects.eao.gov.bc.ca/api/document/58bf02ee7b72fa001cf1f270/fetch. Stewart, K.A., & Lamoureux, S.F. (2011). Connections between River Runoff and Limnological Conditions in Adjacent High Arctic Lakes: Cape Bounty, Melville Island, Nunavut. Arctic.64(2),169–182 Stonehouse, B. (1989). Polar ecology. Blackie, Glasgow, p222 Tilling, R.L., Ridout, A., Shepherd, A., and Wingham, D. (2015). Increased Arctic sea ice volume after anomalously low melting in 2013. Nature Geoscience. 8(8),643–646 US-Canada. (1983). Memorandum of intent on transboundary air pollution. Report of the impact assessment working group I, Section 3 Aquatic effects. Whitfield, C.J., Aherne, J., Watmough, S.A., Dillon, P.J., & Clair, T.A. (2006). Recovery from acidification in Nova Scotia: temporal trends and critical loads for 20 headwater lakes. Canadian Journal of Fisheries and Aquatic Sciences, 63(7), 1504- 1514. Whitfield, C. J., Mowat, A. C., Scott, K. A., & Watmough, S. A. (2016). A modified approach for estimating the aquatic critical load of acid deposition in northern Saskatchewan, Canada. Atmospheric Environment, 14(6), 300–310. Williston, P., Aherne, J., Watmough, S., Marmorek, D., Hall, A., de la Cueva Bueno, P., Murray, C., Henolson, A. and Laurence, J.A., (2016). Critical levels and loads and the regulation of industrial emissions in northwest British Columbia, Canada. Atmospheric Environment, 146, 311-323. Závodský D, Babiaková G, Mitosinková M, Pukanèíková K, Roneák P, Bodis D, Rapant S, Mindás J, Skvarenina J, Cambel B, Rehák S, Wathne BM, Henriksen A, Sverdrup H, Tørseth K, Semb A, & Aamlid D. (1995). Mapping critical level/loads for the Slovak Republic. Acid Rain Research Report 37/1995. Norwegian Institute for Water Research NIVA, Oslo, Norway

123

4.0 Conclusion

4.1 Study Conclusion

Given the growing pressures from climate change, lakes and ponds are effective indicators of change as they encompass characteristics from the surrounding landscape. In addition, they are common features of the Canadian Arctic landscape.

One direct effect of warmer temperatures (as a result of climate change) is the reduction of sea-ice (Environment and Climate Change Canada, 2016; Perovich et al.,

2017), which can lead to the consequent rise of shipping activity within the Canadian

Arctic (Pizzolato et al., 2016; Melia et al., 2016). It is estimated that shipping has contributed between 20 –100% of ambient SO2 concentrations and < 5% to sulphur deposition for the year 2010, however contributions are expected to increase greatly by the year 2030 under a business-as-usual scenario (Gong et al., 2017). The objective of this thesis is to the characterise the chemistry of lakes and ponds in the Canadian Arctic, and to assess the current (2010) impacts of atmospheric sulphur deposition on lake acidity. A total of 1300 sites and 26 common physical and chemical parameters (see Chapter 2) were collated from published sources and recent lake surveys on Baffin Is. (n = 80), Coats

Is. (n = 10), Prince Charles Is. (n = 4), and eastern Northwest Territories (n = 6).

Relationships and drivers of water chemistry characteristics were evaluated by spearman’s correlation matrix and by principal component analysis. In addition, median and mean values were calculated for 20 regions of the Canadian Arctic (mainland territories and the Queen Elizabeth Islands), four ecoregions, and four geological types.

The SSWC model (see Chapter 3) was used to evaulate the acid sensitivity (through the

124

critical loads approach) for a subset of sites (n = 1138) using their associated water chemistry (from Chapter 2). Acid sensitivity was then compared to two senarios of modelled sulphur deposition scenario for the year 2010; with shipping and without shipping.

Chapter 2 results indicate that the majority of sites are shallow (85.4%; <10 m), tundra (84.8%) systems, over sedimentary geology (66.5%). Chemical characteristics of most sites are slightly alkaline (pH median = 7.9), slightly conductive (conductivity median = 97.4 µS·cm-1), oligotrophic (i.e., phosphorus limited with TP = 4–10 µg·L–1;

Canadian Council of Ministers of the Environment, 2004), and are often located at low elevation (66.5%; <200 m.s.a.l) close to the coast (73.5%; 0–50 km). The principal component analysis (of 17 variables and 613 sites) identified three major drivers of water chemistry: (1) sites closer to the coast influenced by sea-salt aerosols, (2) carbonate materials (derived from sedimentary geology) are an important factor for pH,

Conductivity, Ca, Mg, and DIC, and (3) the weathering of non-carbonate geology (Al and

Fe) is associated with inputs of phosphorus. This indicates that most lakes and ponds in the Arctic are primarily driven by the weathering of bedrock geology and inputs of sea- salt aerosols; as such, lakes have similar water chemistry characteristics, i.e., slightly alkaline, nutrient poor, and low metal concentrations. Although most lakes follow these trends, outliers are driven by localized drivers such as bird colonies, sulphate soils, and vegetation, which elevate certain parameters. For example, sites situated near large bird colonies can have elevated nutrient concentrations from the input of feces.

Critical loads of acidity [CL(A)] and exceedances (under modelled total sulphur deposition with and without shipping scenarios) of 1138 sites were assessed for the year 125

2010 (see Chapter 3). Generally, results indicate that most sites are classified as highly sensitive to acidic deposition (median = 35.8 meq·m―2·yr―1). Regions with the lowest value of CL(A) are: Melville Is. (n = 48; mean = 5.5 meq·m―2·yr―1), Banks Is. (n = 45; mean = 18.4 meq·m―2·yr―1), and Bylot Is. (n = 36; mean = 18.4 meq·m―2·yr―1).

Exceedance percentage for the entire region of the Arctic do not differ much between scenarios with shipping (12.5%; n = 142) and without shipping (12.0%; n = 136). Similar exceedances values between the two scenarios (with and without shipping) indicate that the impacts from current (2010) shipping emissions are relatively small. Regions with the highest exceedance percentage are: Ellef Ringnes Is. [n = 25; with shipping = 60% (n =

15); without shipping = 48% (n = 12)], Melville Is. [n = 48; with shipping = 44% (n =

21); without shipping = 38% (n = 18)], and Northwest Territories [n = 25; with shipping

= 60% (n = 15); without shipping = 48% (n = 12)].

Although most sites were classified as highly sensitive to acidic deposition (owing to low weathering rates), low deposition of sulphur throughout the Arctic (with and without the influence of shipping) render a small percentage of sites (~12%; n = 136–

142) as exceeded. Two factors are needed for sites to be exceeded: (1) low CL(A) values and (2) high deposition values (in excess of CL(A). Sites with low CL(A) and low deposition will not be exceeded as there is enough acid neutralization capacity to buffer incoming acidity. However, if shipping traffic continues to grow, more sites may be exceeded due to elevated emissions and deposition of sulphur.

4.2 Contribution to research

126

There have been many Arctic limnological studies in parts of the Canadian Arctic, e.g., Bathurst Is., Lim et al. (2001); Victoria Is., Michelutti et al. (2002); Ellesmere Is.,

Antoniades et al. (2003); Southampton Is., Mallory et al., (2006); nonetheless, there have been few attempts to define baseline water chemistry for the entire Canadian Arctic region (Hamilton et al., 2001; Dranga et al., 2017). Moreover, there has been limited spatial assessment in previous acid sensitivity assessments of Arctic aquatic sites (AMAP,

2006). This research contributes to known gaps in water chemistry and provides a more comprehensive assessment of the acid sensitivity of Arctic aquatic sites.

More specifically, this study provides new water chemistry data (from Baffin Is.,

Coats Is., Prince Charles Is., and Northwest Territories) of Arctic lakes and ponds and a regional assessment of the acid sensitivity of 1138 sites in the Canada Arctic. The database from this study (n=1300) can be expanded using existing water chemistry data, i.e., Geological Survey of Canada, Environment and Climate Change Canada’s Fresh

Water Quality Monitoring and Surveillance program (FWQM), Van Hove et al. (2006),

Namayandeh and Quinlan (2011), and from the works of future lake surveys. Use of such database can facilitate other regional assessments of Arctic lakes and ponds, such as the

Arctic Freshwater Biodiversity Monitoring Plan by the Freshwater Ecosystem Monitoring

Group (FEMG) of Conservation of Arctic Flora and Fauna (CAFF), International

Cooperative Programme for assessment and monitoring of the effects of air pollution on rivers and lakes (ICP-Waters), and Environment and Climate Change Canada’s FWQM.

Results from the assessment of 1138 sites can help better understand the potential impacts from elevated shipping emissions and has advanced the works of AMAP (2006), Aherne and Jeffries (2015) and Gong et al. (2018); combined these studies may support the

127

development of an Emission Control Area in the Arctic by the International Maritime

Organization.

4.3 Recommendations and further work

There is an urgent need to improve our knowledge of baseline chemistry for

Arctic lakes and ponds. The widespread adoption of water chemistry analysis for other research projects, i.e., paleolimnology, polar contaminants, biomonitoring, fish assessments, are essential for a better understanding of the Arctic as a whole; there is a greater need for more lakes surveys to address known spatial and temporal monitoring gaps, and more intensive lake surveys among high-risk areas, especially those along major shipping routes. Investigation into the deposition of acidic compounds through the use of precipitation collectors in high-risk areas are also necessary. Similar analysis of water chemistry and acid sensitivity undertaken in Chapter 2 and 3, would be useful to analyze new spatial and temporal trends of acid sensitivity of lakes and ponds in the

Canada Arctic. Given that most sites are sampled once and within the summer months, there is a need for an increase in the sampling frequency of sites for long-term monitoring and to account for seasonality bias. Greater investigation of long-term monitoring sites will further inform our understanding on chemical cycling, and concentration changes owning to climate.

128

References

Aherne, J. & Jeffries, D. (2015). Critical load assessments and dynamic model applications for lakes in North America. In Critical Loads and Dynamic Risk Assessments (pp. 485-503). Springer, Dordrecht. Antoniades, D., Douglas, M.S. V., & Smol, J.P. (2003). Comparative physical and chemical limnology of two Canadian High Arctic regions: Alert (Ellesmere Is., NU) and Mould Bay (Prince Patrick Is., NWT). Arch. für Hydrobiol. 158: 485–516. doi:10.1127/0003-9136/2003/0158-0485. AMAP. (2006). AMAP Assessment Report: Acidifying Pollutants, Arctic Haze, and Acidification in the Arctic, Arctic Monitoring and Assessment Programme AMAP, Oslo, Norway. Xii+859pp Dranga, S.A., Hayles, S. & Gajewski, K. (2017). Synthesis of limnological data from lakes and ponds across Arctic and Boreal Canada. Arctic Science. doi:10.1139/as- 2017-0039. Environment and Climate Change Canada. (2016). Canadian Environmental Sustainability Indicators: Sea Ice in Canada. Consulted on Month day, year. Available at: www.ec.gc.ca/indicateursindicators/default.asp?lang=en&n=1F76A73A-1.

Gong, W., Beagley, S.R., Cousineau, S., Sassi, M., Munoz-alpizar, R., Ménard, S., Racine, J., Zhang, J., Chen, J., Morrison, H., Sharma, S., Huang, L., Bellavance, P., Ly, J., Izdebski, P., Lyons, L., & Holt, R. (2018). Assessing the impact of shipping emissions on air pollution in the Canadian Arctic and northern regions : current and future modelled scenarios. Atmospheric Chemistry and Physics Hove, P. V., Belzile, C., Gibson, J. A., & Vincent, W. F. (2006). Coupled landscape-lake evolution in High Arctic Canada. Canadian Journal of Earth Sciences. 43(5), 533- 546. Lim, D.S.S., Douglas, M.S.V., Smol, J.P., & Lean, D.R.S. (2001). Physical and chemical limnological characteristics of 38 lakes and ponds on Bathurst Is., Nunavut, Canadian high arctic. Int. Rev. Hydrobiol. 86: 1–22. doi:10.1002/1522- 2632(200101)86:1<1::AID-IROH1>3.0.CO;2-E. Melia, N., Haines, K., & Hawkins, E. (2016). Sea ice decline and 21st century trans‐ Arctic shipping routes. Geophysical Research Letters, 43(18), 9720-9728. Michelutti, N., Douglas, M.S. V, Lean, D.R.S., & Smol, J.P. (2002). Physical and chemical limnology of 34 ultra-oligotrophic lakes and ponds near Wynniatt Bay, Victoria Is., Arctic Canada. Hydrobiologiam. 482: 1–13. doi:10.1023/A:1021201704844. Mallory, M.L., Fontaine, A.J., Smith, P.A., Robertson, M.O.W., & Gilchrist, H.G. (2006). 129

Water chemistry of ponds on Southampton Is., Nunavut, Canada: Effects of habitat and ornithogenic inputs. Arch. Fur Hydrobiol. 166: 411–432. doi:10.1127/0003- 9136/2006/0166-0411. Namayandeh, A., & Quinlan, R. (2011). Benthic macroinvertebrate communities in Arctic lakes and ponds of central Nunavut, Canada. Arctic, Antarctic, and Alpine Research, 43(3), 417-428.doi: 10.1657/1938-4246-43.3.417 Perovich, D., Meier, W., Tschudi, M., Farrell, S., Hendricks, S., Gerland, S., Hass, C., Krumpen, T., Polashenski, C., Ricker, R., & Webster, M. (2017). Sea Ice. In Arctic Report Card 2017. http://www.arctic.noaa.gov/Report-Card.

Pizzolato, L., Howell, S. E., Dawson, J., Laliberté, F., & Copland, L. (2016). The influence of declining sea ice on shipping activity in the Canadian Arctic. Geophysical Research Letters, 43(23).

130

Appendix A Supplementary information for Chapter 2 Field Collections

Surface water temperature (TEMP), pH, and conductivity (COND) were measured only on Baffin Is., and measurements were taken below the surface at an approximate depth of

0.15 m with a Hanna pH/Conductivity/TDS Meter (Model HI991300), while dissolved oxygen (DO) was measured with a handheld YB ProODO™ meter. Surface water were collected in a pre–cleaned 250 mL HDPE bottle (rinsed three times with lakes water) at a depth of 0.15 m below the surface. All samples are placed in a Ziploc bag during field transport and are kept refrigerated until analysis. Unfiltered water samples were used for the measurement of calcium (Ca), Iron (Fe), potassium (K), magnesium (Mg), manganese

(Mn), sodium (Na), and strontium (Sr) by inductively coupled plasma optical emission spectrometry (ICP–OES). Water samples filtered through 0.45 µm nylon filter for dissolve organic carbon (DOC), dissolve inorganic carbon (DIC), total dissolve nitrogen

– – – – (TDN) by Shimadzu TOCV; fluoride (F ), chloride (Cl ), nitrite (NO2 ), nitrate (NO3 ),

2– and sulphate (SO4 ) by Ion chromatography; total phosphorus (TP) by colourimetry

(Persulfate Digestion); ammonia (N–NH3) by continues flow; silver (Ag), aluminum (Al), arsenic (As), boron (B), barium (Ba), beryllium (Be), bismuth (Bi), cadmium (Cd), cobalt

(Co), chromium (Cr), copper (Cu), lithium (Li), molybdenum (Mo), nickel (Ni),

Rubidium (Rb), lead (Pb), sulfur (S), antimony (Sb), selenium (Se), silicon (Si), tin (Sn), thorium (Th), titanium (Ti), thallium (Tl), uranium (U), vanadium (V), zinc (Zn), and zirconium (Zr) by inductively coupled plasma mass spectrometry (ICP–MS).

131

Table A1. Descriptive statistics of 59 parameters including; count, mean, standard deviation (SD), minimum, maximum, and percentile (5th and 9th) values. Percentile Parameter Symbol Units Count Mean SD Min Max Median CV% 5th 95th Depth Depth m 680 5.54 5.71 0.03 49.00 4.00 103.13 0.50 16.81 Transparency Transp m 137 3.56 2.26 0.50 11.50 3.00 63.38 1.00 8.52 Surface Water Temp ⁰C 894 9.21 4.85 –0.40 25.00 8.02 52.68 2.07 18.14 Temperature Dissolve DO mg·L–1 216 11.72 2.44 1.10 20.00 12.56 20.85 7.88 13.95 Oxygen Alkalinity Alkalinity mg·L–1 217 47.83 65.27 –25.43 350.00 19.74 136.47 0.98 191.20 Soluble SRSi mg·L–1 16 0.64 0.50 0.20 1.86 0.48 77.43 0.24 1.69 Reactive Silicate Total Suspended TSS mg·L–1 20 25.62 86.05 0.11 384.00 1.00 335.90 0.23 92.35 Solids Particulate Organic PON mg·L–1 465 0.16 2.09 0.00 45.00 0.04 1314.21 0.01 0.19 Nitrogen –1 Nitrite NO2* µg·L 480 0.00 0.03 0.00 0.63 0.00 671.78 0.26 7.00 –1 Nitrate Nitrite NO3+NO2* µg·L 458 0.02 0.05 0.00 0.89 0.01 303.34 0.60 62.17 Particulate PP* µg·L–1 205 0.01 0.02 0.00 0.15 0.00 218.73 0.53 29.96 Phosphorous Particulate PN* µg·L–1 306 0.06 0.06 0.00 0.52 0.05 106.01 7.20 128.75 Nitrogen Total Dissolved TDN* µg·L–1 123 0.40 0.43 0.03 1.99 0.24 108.30 57.25 1425.00 Nitrogen Total Dissolved TDP* µg·L–1 22 0.00 0.00 0.00 0.01 0.00 110.09 1.00 8.56 Phosphorous Suspended SN µg·L–1 9 0.04 0.02 0.02 0.08 0.03 53.58 16.40 65.20 Nitrogen Suspended SC µg·L–1 9 0.30 0.13 0.20 0.57 0.25 42.92 204.00 530.00 Carbon 132

Dissolved Inorganic DIN µg·L–1 2 Nitrogen Dissolved Organic DON µg·L–1 2 Nitrogen Soluble Reactive SRP µg·L–1 675 0.15 0.50 0.00 4.80 0.00 325.05 0.16 1100.00 Phosphorus Chlorophyll–A Chla.a µg·L–1 684 0.18 0.57 0.00 6.90 0.00 316.73 0.05 1200.00 Chlorophyll–A Chla.u µg·L–1 496 0.00 0.00 0.00 0.02 0.00 151.00 0.05 2.62 Uncorrected Silver Ag µg·L–1 190 0.00 0.00 0.00 0.00 0.00 155.75 0.00 2.00 Arsenic As µg·L–1 224 0.00 0.00 0.00 0.02 0.00 290.50 0.03 1.41 Boron B µg·L–1 253 0.01 0.01 0.00 0.11 0.00 180.37 0.35 20.16 Beryllium Be µg·L–1 198 0.00 0.00 0.00 0.00 0.00 126.75 0.00 0.20 Bismuth Bi µg·L–1 57 0.00 0.00 0.00 0.01 0.00 245.91 0.00 4.59 Cadmium Cd µg·L–1 264 0.00 0.00 0.00 0.01 0.00 195.56 0.00 3.63 Cerium Ce µg·L–1 28 0.00 0.00 0.00 0.00 0.00 97.80 0.01 0.39 Cobalt Co µg·L–1 271 0.00 0.00 0.00 0.03 0.00 292.37 0.01 3.72 Chromium Cr µg·L–1 242 0.00 0.00 0.00 0.01 0.00 139.67 0.01 2.97 Cesium Cs µg·L–1 2 Copper Cu µg·L–1 592 0.00 0.00 0.00 0.03 0.00 123.32 0.22 8.00 Fluoride F µg·L–1 89 0.03 0.04 0.00 0.24 0.02 135.48 5.09 68.78 Gallium Ga µg·L–1 89 0.00 0.00 0.00 0.00 0.00 129.29 0.00 0.04 Lanthanum La µg·L–1 89 0.00 0.00 0.00 0.00 0.00 369.10 0.00 0.49 Lithium Li µg·L–1 548 0.00 0.02 0.00 0.34 0.00 389.78 0.05 13.65 Molybdenum Mo µg·L–1 364 0.00 0.06 0.00 1.19 0.00 1373.06 0.01 3.66 Neodymium Nb µg·L–1 2 Nickel Ni µg·L–1 424 0.01 0.09 0.00 1.35 0.00 670.26 0.11 9.00 Phosphorus P µg·L–1 5 0.06 0.10 0.00 0.24 0.02 163.82 4.56 197.04

133

Lead Pb µg·L–1 366 0.00 0.00 0.00 0.05 0.00 253.94 0.02 5.00 Platinum Pt µg·L–1 2 Rubidium Rb µg·L–1 246 0.00 0.00 0.00 0.02 0.00 161.44 0.13 7.16 Sulfur S µg·L–1 92 1.81 1.99 0.25 13.60 1.35 110.33 367.57 5625.00 Antimony Sb µg·L–1 147 0.00 0.00 0.00 0.01 0.00 445.70 0.00 0.53 Scandium Sc µg·L–1 2 Selenium Se µg·L–1 80 0.00 0.01 0.00 0.03 0.00 384.42 0.01 18 Silicon Si µg·L–1 134 1.77 2.36 0.02 21.5 1.36 133.77 114.21 4670 Tin Sn µg·L–1 60 0.00 0.00 0.00 0.01 0.00 387.64 0.00 1.97 Thorium Th µg·L–1 41 0.00 0.00 0.00 0.00 0.00 177.04 0.00 0.02 Titanium Ti µg·L–1 79 0.00 0.04 0.00 0.32 0.00 782.91 0.03 3.02 Thallium Tl µg·L–1 140 0.00 0.00 0.00 0.00 0.00 465.38 0.00 0.13 Uranium U µg·L–1 219 0.00 0.00 0.00 0.00 0.00 269.46 0.01 0.53 Vanadium V µg·L–1 287 0.00 0.00 0.00 0.02 0.00 194.04 0.02 2.97 Tungsten W µg·L–1 2 Yttrium Y µg·L–1 2 Zinc Zn µg·L–1 696 0.00 0.01 0.00 0.13 0.00 239.19 0.05 10.00 Zirconium Zr µg·L–1 58 0.00 0.00 0.00 0.00 0.00 146.61 0.00 0.55 Nitrogen Isotope d15N % 31 3.80 14.37 –28.55 23.84 7.40 377.83 –22.77 18.90 15

Table A2. List of PCA variables and their loadings for the five principal components with eigenvalues above 1. Variables PC1 PC2 PC3 PC4 PC5 Eigenvalues 6.445 2.576 1.080 1.406 1.154 Latitude -0.152 -0.149 -0.246 -0.130 0.610 Elev 0.169 -0.231 0.355 0.312 0.327 Distance 0.123 -0.062 0.539 0.217 0.042 pH -0.181 -0.257 0.111 -0.524 0.022 Cond -0.352 -0.153 -0.065 0.180 -0.018

134

Ca -0.313 -0.273 -0.056 0.128 0.133 K -0.295 0.157 0.113 0.257 -0.113 Mg -0.346 -0.147 0.022 0.145 0.099 Na -0.335 0.164 -0.107 0.033 -0.241 Cl -0.313 0.165 -0.106 -0.125 -0.270 DOC -0.185 0.127 0.502 -0.040 -0.117 DIC -0.280 -0.338 0.102 -0.174 0.087 TN -0.258 0.107 0.387 -0.235 0.083 TP -0.128 0.385 0.147 -0.042 0.059 Al -0.046 0.435 -0.121 0.140 0.362 Fe -0.076 0.421 0.070 -0.207 0.427 SO4 -0.243 0.008 -0.113 0.511 0.049

Table A3. Median (mean) sea-salt corrected concentrations for Ca, Mg, Na, K, and SO4, per region, ecoregion, and geology type. Group Ca Mg Na K SO4 Region Axel Heiberg Is. 16.1 (30.44) 4 (8.64) 0.8 (2.63) 0.6 (9.31) 8 (53.19) Baffin Is. 5.8 (8.56) 0.8 (1.18) 0.15 (0.56) 0.1 (0.2) 2.6 (7.86) Banks Is. 16.6 (18.58) 8.5 (10.61) 0.5 (0.79) 0 (0.74) 7.4 (11) Bathurst Is. 24.7 (22.74) 3.7 (4) 0.2 (0.39) 0.1 (0.47) 2.1 (4.39) Bylot Is. 1.65 (3.29) 0.9 (1.48) 0.5 (0.63) 0.55 (0.88) 0.6 (1.83) Coats Is. 18.8 (21.87) 4.8 (4.4) 0 (0.08) 0 (2.13) 0 (0.41) Cornwallis Is. 25.9 (23.64) 4.8 (4.86) 0.2 (0.33) 0 (2.71) 1.1 (3.45) Devon Is. 24.45 (25.04) 5.3 (5.75) 0.2 (0.21) 0 (0.12) 1.9 (13.37) Ellef Ringnes Is. 19.6 (48.06) 9.9 (32.48) 1.9 (3.31) 11.7 (45.33) 130 (296.63) Ellesmere Is. 30.85 (41.68) 6.95 (13.41) 0.8 (3.06) 0.5 (11.04) 5.6 (72.66) Melville Is. 5.95 (14.47) 2.85 (5.9) 0.65 (0.8) 0 (0.32) 0.6 (9.47) Northwest 5.1 (12.61) 1.1 (4.15) 0.7 (1.18) 0.3 (1.14) 2.4 (10.4) Territories Nunavut 3.85 (6.18) 0.9 (1.7) 0.4 (0.57) 0.1 (0.23) 0.9 (1.56)

135

Prince Charles Is. 17 (17.12) 3.1 (2.68) 0.5 (0.42) 0.8 (1.18) 0.8 (1.54) Prince of Wales Is. 25.9 (26.98) 10.1 (10.42) 0.4 (0.42) 0 (0) 6.2 (9.1) Prince Patrick Is. 10.8 (12.65) 2.9 (3.48) 0.6 (0.6) 0 (0.86) 3.2 (5.61) Somerset Is. 10 (13.42) 3.5 (3.82) 0.2 (0.2) 0 (0.12) 1.4 (3.15) Southampton Is. 32 (29.67) 6 (6.39) 0.5 (1.17) 5.1 (15) 13.65 (39.39) Victoria Is. 22.3 (21.24) 9.25 (10.6) 0.3 (0.38) 0 (0.04) 1.7 (3.91) Yukon 30 (32.97) 18.4 (28.29) 2.5 (3.71) 3.25 (7.28) 20.1 (67.44) Ecoregion AC 2.7 (20.08) 0.65 (6.66) 0.3 (2.24) 0.2 (2.3) 2.3 (70.4) NWF 30.5 (34.9) 18.4 (28.29) 2.95 (4.08) 3.45 (7.93) 24 (73.76) TA 5.25 (13.22) 1.5 (4.59) 0.7 (1.27) 0.4 (1.43) 2.75 (12.41) TU 15.1 (20.23) 3.2 (6.53) 0.4 (1.11) 0.1 (4.09) 2.1 (24.9) Geology Igneous 4 (7.62) 0.8 (1.59) 0.3 (0.58) 0.1 (0.37) 1.5 (4.74) Sedimentary 20.4 (25.59) 5 (8.86) 0.5 (1.52) 0.1 (5.61) 3 (36.67) Supracrustal 13.1 (17.12) 2.1 (11.09) 0.95 (1.79) 0.2 (2.61) 2.4 (11.09) unclassified 15.9 (20.19) 5.6 (21.35) 0.6 (2.26) 0.1 (1.7) 2.55 (54.81)

136

Figure A1. Water dissolved inorganic carbon (DIC) concentrations (mg·L–1) as a function of alkalinity (mg·L–1) in arctic lakes and ponds (n = 98) in the Canadian Arctic.

60000

y = 0.9082x + 237.68

50000 R² = 0.989 )

1 40000 -

30000 µmolc·L

Anions ( Anions 20000

10000

0 0 10000 20000 30000 40000 50000 60000 Cations (µmolc·L-1)

Figure A2. Ion balance of the major cations and anions for 224 lakes in the Canadian Arctic.

137

Appendix B Supporting Information for Chapter 3

Table B1. Percentage of sites (under Brown Trout ANClimit) within different acid sensitivity classes (values are expressed in meq·m–2·yr–1) among different regions of the Canadian Arctic. Region ≤20 20-40 40-60 60-100 >100 Axel Heiberg Is. 55.3% (n = 26) 10.6% (n = 5) 12.8% (n = 6) 12.8% (n = 6) 8.5% (n = 4) Baffin Is. 18.4% (n = 23) 10.4% (n = 13) 14.4% (n = 18) 19.2% (n = 24) 37.6% (n = 47) Banks Is. 68.9% (n = 31) 20% (n = 9) 4.4% (n = 2) 6.7% (n = 3) 0% (n = 0) Bathurst Is. 19.4% (n = 13) 11.9% (n = 8) 20.9% (n = 14) 29.9% (n = 20) 17.9% (n = 12) Bylot Is. 69.4% (n = 25) 16.7% (n = 6) 8.3% (n = 3) 2.8% (n = 1) 2.8% (n = 1) Coats Is. 0% (n = 0) 0% (n = 0) 0% (n = 0) 0% (n = 0) 100% (n = 10) Cornwallis Is. 6.8% (n = 3) 4.5% (n = 2) 9.1% (n = 4) 50% (n = 22) 29.5% (n = 13) Crozier Is. 0% (n = 0) 0% (n = 0) 100% (n = 2) 0% (n = 0) 0% (n = 0) Devon Is. 42.5% (n = 17) 37.5% (n = 15) 5% (n = 2) 5% (n = 2) 10% (n = 4) Ellef Ringnes Is. 72% (n = 18) 12% (n = 3) 0% (n = 0) 4% (n = 1) 12% (n = 3) Ellesmere Is. 16.1% (n = 23) 6.3% (n = 9) 4.9% (n = 7) 3.5% (n = 5) 69.2% (n = 99) King William Is. 75% (n = 3) 0% (n = 0) 0% (n = 0) 0% (n = 0) 25% (n = 1) Little Cornwallis Is. 100% (n = 1) 0% (n = 0) 0% (n = 0) 0% (n = 0) 0% (n = 0) Melville Is. 97.9% (n = 47) 2.1% (n = 1) 0% (n = 0) 0% (n = 0) 0% (n = 0) Northwest Territories 61.5% (n = 59) 9.4% (n = 9) 5.2% (n = 5) 9.4% (n = 9) 14.6% (n = 14) Nunavut 50.3% (n = 92) 19.7% (n = 36) 13.1% (n = 24) 12.6% (n = 23) 4.4% (n = 8) Prince Charles Is. 0% (n = 0) 0% (n = 0) 20% (n = 1) 20% (n = 1) 60% (n = 3) Prince of Wales Is. 0% (n = 0) 40% (n = 2) 40% (n = 2) 20% (n = 1) 0% (n = 0) Prince Patrick Is. 45.7% (n = 16) 34.3% (n = 12) 20% (n = 7) 0% (n = 0) 0% (n = 0) Somerset Is. 15.4% (n = 2) 23.1% (n = 3) 46.2% (n = 6) 7.7% (n = 1) 7.7% (n = 1) Southampton Is. 2.9% (n = 1) 2.9% (n = 1) 0% (n = 0) 2.9% (n = 1) 91.4% (n = 32) Victoria Is. 59.6% (n = 53) 0% (n = 0) 1.1% (n = 1) 6.7% (n = 6) 32.6% (n = 29) Yukon 5% (n = 2) 0% (n = 0) 2.5% (n = 1) 7.5% (n = 3) 85% (n = 34)

138

Table B2. Percentage of sites (under Ecosystem ANClimit) within different acid sensitivity classes (values are expressed in meq·m–2·yr–1) among different regions of the Canadian Arctic. Region ≤20 20-40 40-60 60-100 >100 Axel Heiberg Is. 55.3% (n = 26) 10.6% (n = 5) 12.8% (n = 6) 12.8% (n = 6) 8.5% (n = 4) Baffin Is. 19.2% (n = 24) 11.2% (n = 14) 12.8% (n = 16) 20% (n = 25) 36.8% (n = 46) Banks Is. 68.9% (n = 31) 20% (n = 9) 4.4% (n = 2) 6.7% (n = 3) 0% (n = 0) Bathurst Is. 19.4% (n = 13) 10.4% (n = 7) 22.4% (n = 15) 29.9% (n = 20) 17.9% (n = 12) Bylot Is. 69.4% (n = 25) 16.7% (n = 6) 8.3% (n = 3) 2.8% (n = 1) 2.8% (n = 1) Coats Is. 0% (n = 0) 0% (n = 0) 0% (n = 0) 0% (n = 0) 100% (n = 10) Cornwallis Is. 6.8% (n = 3) 4.5% (n = 2) 9.1% (n = 4) 50% (n = 22) 29.5% (n = 13) Crozier Is. 0% (n = 0) 0% (n = 0) 100% (n = 2) 0% (n = 0) 0% (n = 0) Devon Is. 42.5% (n = 17) 37.5% (n = 15) 5% (n = 2) 5% (n = 2) 10% (n = 4) Ellef Ringnes Is. 72% (n = 18) 12% (n = 3) 0% (n = 0) 4% (n = 1) 12% (n = 3) Ellesmere Is. 16.1% (n = 23) 6.3% (n = 9) 4.9% (n = 7) 2.8% (n = 4) 69.9% (n = 100) King William Is. 75% (n = 3) 0% (n = 0) 0% (n = 0) 0% (n = 0) 25% (n = 1) Little Cornwallis Is. 100% (n = 1) 0% (n = 0) 0% (n = 0) 0% (n = 0) 0% (n = 0) Melville Is. 97.9% (n = 47) 2.1% (n = 1) 0% (n = 0) 0% (n = 0) 0% (n = 0) Northwest Territories 59.4% (n = 57) 10.4% (n = 10) 6.3% (n = 6) 8.3% (n = 8) 15.6% (n = 15) Nunavut 49.7% (n = 91) 19.1% (n = 35) 14.2% (n = 26) 12.6% (n = 23) 4.4% (n = 8) Prince Charles Is. 0% (n = 0) 0% (n = 0) 20% (n = 1) 20% (n = 1) 60% (n = 3) Prince of Wales Is. 0% (n = 0) 40% (n = 2) 40% (n = 2) 20% (n = 1) 0% (n = 0) Prince Patrick Is. 42.9% (n = 15) 37.1% (n = 13) 20% (n = 7) 0% (n = 0) 0% (n = 0) Somerset Is. 23.1% (n = 3) 15.4% (n = 2) 46.2% (n = 6) 7.7% (n = 1) 7.7% (n = 1) Southampton Is. 2.9% (n = 1) 2.9% (n = 1) 0% (n = 0) 2.9% (n = 1) 91.4% (n = 32) Victoria Is. 59.6% (n = 53) 0% (n = 0) 1.1% (n = 1) 6.7% (n = 6) 32.6% (n = 29) Yukon 5% (n = 2) 0% (n = 0) 2.5% (n = 1) 7.5% (n = 3) 85% (n = 34)

139

–2 –1 Table B3. Number of sites, mean CL(A) (calculated with Brown Trout ANClimit value; meq·m ·yr ), mean sulphur deposition (meq·m–2·yr–1) from both model, and exceedances (percentage of total sites and number of sites under both models) for each region of the Canadian Arctic. No. of CL(A) Deposition Model Exceedances Brown Gong et Moran et al. Gong et Moran et al. Region sites Trout al. (2018) (2016) al. (2018) (2016) Axel Heiberg Is. 47 44.0 1.72 1.59 9% (4) 6% (3) Baffin Is. 125 100 2.27 3.38 6% (8) 6% (8) Banks Is. 45 18.4 1.37 1.56 9% (4) 9% (4) Bathurst Is. 67 61.9 2.85 1.63 10% (7) 9% (6) Bylot Is. 36 20.5 2.32 1.27 6% (2) 3% (1) Coats Is. 10 228 1.39 3.15 0% (0) 0% (0) Cornwallis Is. 44 86.6 2.05 1.49 5% (2) 0% (0) Devon Is. 40 32.5 2.49 1.25 3% (1) 0% (0) Ellef Ringnes Is. 25 32.9 4.40 2.31 56% (14) 48% (12) Ellesmere Is. 143 266 4.20 0.63 3% (4) 0% (0) King William Is. 4 30.4 3.39 1.90 25% (1) 0% (0) Melville Is. 48 5.55 2.93 1.66 44% (21) 38% (18) Northwest Territories 96 46.4 2.28 3.46 24% (23) 26% (25) Nunavut 183 30.2 2.99 3.01 16% (29) 17% (31) Prince Charles Is. 5 103 3.97 1.63 0% (0) 0% (0) Prince of Wales Is. 5 43.3 3.15 1.55 0% (0) 0% (0) Prince Patrick Is. 35 24.9 4.86 1.70 3% (1) 0% (0) Somerset Is. 13 48.6 3.71 1.62 8% (1) 8% (1) Southampton Is. 35 251 1.16 3.06 0% (0) 0% (0) Victoria Is. 89 53.9 1.93 2.09 12% (11) 24% (21) Yukon 40 435 2.56 1.17 5% (2) 5% (2) Grand total 12% (136) 12% (133)

140

–2 –1 Table B4. Number of sites, mean CL(A) (calculated with the Ecosystem ANClimit value; meq·m ·yr ), mean sulphur deposition (meq·m–2·yr–1) from both model, and exceedances (percentage of total sites and number of sites under both models) for each region of the Canadian Arctic. No. of CL(A) Deposition Model Exceedances Gong et al. Moran et al. Gong et al. Moran et al. Region sites Ecosystem (2018) (2016) (2018) (2016) Axel Heiberg Is. 47 44.1 1.72 1.59 9% (4) 6% (3) Baffin Is. 125 98.9 2.27 3.38 9% (11) 10% (12) Banks Is. 45 18.5 1.37 1.56 9% (4) 9% (4) Bathurst Is. 67 62.0 2.85 1.63 10% (7) 9% (6) Bylot Is. 36 20.7 2.32 1.27 8% (3) 3% (1) Coats Is. 10 230 1.39 3.15 0% (0) 0% (0) Cornwallis Is. 44 86.4 2.05 1.49 5% (2) 2% (1) Devon Is. 40 32.4 2.49 1.25 3% (1) 0% (0) Ellef Ringnes Is. 25 32.7 4.40 2.31 60% (15) 48% (12) Ellesmere Is. 143 264 4.20 0.63 3% (4) 0% (0) King William Is. 4 30.4 3.39 1.90 25% (1) 0% (0) Melville Is. 48 5.58 2.93 1.66 44% (21) 38% (18) Northwest Territories 96 48.8 2.28 3.46 24% (23) 24% (23) Nunavut 183 30.6 2.99 3.01 16% (29) 17% (32) Prince Charles Is. 5 103 3.97 1.63 0% (0) 0% (0) Prince of Wales Is. 5 43.4 3.15 1.55 0% (0) 0% (0) Prince Patrick Is. 35 25.3 4.86 1.70 3% (1) 0% (0) Somerset Is. 13 48.3 3.71 1.62 8% (1) 8% (1) Southampton Is. 35 253 1.16 3.06 0% (0) 0% (0) Victoria Is. 89 53.9 1.93 2.09 13% (12) 24% (21) Yukon 40 438 2.56 1.17 5% (2) 5% (2) Grand total 12% (142) 12% (137)

141

–1 Figure B1. Critical loads of Acidity for 1138 sites using an ANClimit value for Brown Trout (8 µeq L ), and relative sensitivity to acidification. 142

– Figure B2. Critical loads of Acidity for 1138 sites using an ANClimit value for Ecosystem (not organic acid adjusted; 20 µeq L 1), and relative sensitivity to acidification. 143

–1 Figure B3. Critical loads (with ANClimit value of Brown Trout, 8 µeq L ) exceedances under 2010 with shipping sulphur deposition scenario. 144

–1 Figure B4. Critical loads (with ANClimit value of Ecosystem, 20 µeq L ) exceedances under 2010 with shipping sulphur deposition scenario. 145

–1 Figure B5. Critical loads (with ANClimit value of Brown Trout, 8 µeq L ) exceedances under 2010 without shipping sulphur deposition scenario. 146

–1 Figure B6. Critical loads (with ANClimit value of Ecosystem, 20 µeq L ) exceedances under 2010 without shipping sulphur deposition scenario. 147

Figure B7. Sulphur deposition (meq·m–2·yr–1) from both model; with shipping (top) and without shipping (bottom) among sites in the Canadian Arctic.

148