uncorrected proof

AUTHOR'S PROOF!

JrnlID 11356_ArtID 1462_Proof# 1 - 06/02/2018

Environmental Science and Pollution Research https://doi.org/10.1007/s11356-018-1462-z

1 3 2

RESEARCH ARTICLE

4 5 6 7

Dissolved organic matter distribution and its association with colloidal aluminum and iron in the Selenga River Basin from Ulaanbaatar to Lake Baikal Morimaru Kida 1 & Orgilbold Myangan 2 & Bolormaa Oyuntsetseg 3 & Viacheslav Khakhinov 4 & Masayuki Kawahigashi 2 & Nobuhide Fujitake 1

11 12

Received: 13 November 2017 / Accepted: 1 February 2018 # Springer-Verlag GmbH Germany, part of Springer Nature 2018

13 14 15 16 17 18 19 20 21 22 23 24 25

Abstract The Selenga River Basin (Mongolia and Russia) has suffered from heavy metal contamination by placer gold mining and urban activities in recent decades. The objectives of this study were to provide the first distribution data of dissolved organic matter (DOM) and humic substances (HS) in this data-scarce region, and to investigate their association with dissolved and colloidal metals. Two sampling campaigns were conducted in August of 2013 and 2014. A constant proportion of HS (%HS; coefficient of variation of 2%) was observed from the headwater of Tuul River to the end of the delta before Lake Baikal, spanning > 1000 km in distance. The relationships were determined as [HS] = 0.643 × [DOM] (R2 = 0.996, P < 0.001), and this value (%HS = 64.3) is recommended as an input parameter for metal speciation modeling based on samples collected from the rivers. The DOM and metal (Al and Fe) concentrations in samples doubled through the Zaamar Goldfield mining area, but the influence was mitigated by mixing with the larger Orkhon River, which has better water quality. Metals were mainly present as colloids and had a strong positive correlation with DOM (Al r = 0.81, P < 0.01; Fe r = 0.61, P < 0.01), suggesting that DOM sustains colloidal Al and Fe in solution and they are co-transported in the Selenga River Basin. Land use changes affect water quality and metal speciation and therefore have major implications for the fate of metals.

26 27

Keywords Contamination . DOM . Heavy metal . Humic substances . Metal speciation . Model VI . Stockholm Humic Model . Visual MINTEQ

PR O O

N C O R R EC TE D

U

28 Q1

F

8 9 10

Morimaru Kida Research Fellow of Japan Society for the Promotion of Science Responsible editor: Philippe Garrigues * Nobuhide Fujitake [email protected] 1

Graduate School of Agricultural Science, Kobe University, 1 Rokkodai, Nada, Kobe, Hyogo 657-8501, Japan

2

Department of Geography, Graduate School of Urban Environmental Sciences, Tokyo Metropolitan University, Minamiosawa 1-1, Hachioji, Tokyo 192-0397, Japan

3

Department of Chemistry, National University of Mongolia, Ulaanbaatar, Mongolia

4

Department of Chemistry, Buryat State University, Buryatia, Russia

Introduction

29

The bulk of organic matter in water is present as dissolved organic matter (DOM). DOM is a heterogeneous mixture of various organic materials, and is involved in a range of physical, chemical, and biological processes in aquatic systems. These processes are largely driven by the hydrophobic acid fraction of DOM, also known as humic substances (HS), with acidic functional groups such as carboxyl and phenolic hydroxyl groups (Bell et al. 2015). Among the many roles of HS in aquatic systems, enhancement to the solubility of metals and hydrophobic contaminants is particularly biogeochemically and ecologically important. HS affect metal speciation through complexation and change the mobility, bioavailability, and toxicity of metals (Bergamaschi et al. 2012; Taylor et al. 2015; Blazevic et al. 2016). HS also

30 31 32 33 34 35 36 37 38 39 40 41 42 43

AUTHOR'S PROOF!


Environ Sci Pollut Res

F

The objective of this study was to provide the first quantitative data of DOM and HS distribution along the Selenga River Basin, and to investigate their effects on metal distributions. The DOM, HS, and metal (Fe and Al) contents can be affected by changes in the water quality at the confluences between the main river and the tributaries. The Selenga River and the major tributaries with varying anthropogenic activities were sampled to examine the influences. Metals were size-fractionated into total, dissolved, and colloidal fractions to investigate the metal speciation under different water conditions. Sedimentary and suspended metal distributions and loads in the Mongolian territory were investigated in our previous research (Myangan et al. 2017).

97 98 99 100 101 102 103 104 105 106 107 108 109

110

Study site

111

The Selenga River Basin has an area of 447,000 km2 and is characterized largely by a highly continental climate with very cold winters and a limited natural water availability. It contains the world’s largest inland continental delta with an area of 600 km2 (Chalov et al. 2017). The main tributaries of the Selenga River in Mongolia are the rivers Tuul, Orkhon, and Kharaa with more than half of Mongolia’s population and the main industrial districts concentrated on their banks (Karthe et al. 2017). As shown in Fig. 1, the Tuul River flows west through a valley of northern Mongolia where its capital, Ulaanbaatar, is located at about 1350 m above mean sea level. It turns north at the border of Bulgan aimag to meet the Orkhon River. The Orkhon River joins the Selenga River near the border of Mongolia and Russia, which flows into Lake Baikal after it runs > 400 km in Russian territory. The Selenga River is the largest river draining into Lake Baikal, with annual discharge of about 30 km3 of water and 3.5 million tons of sediment into the lake (Thorslund et al. 2014). The rivers are typically covered by ice during winter. The main land use of the Selenga basin is grazing, with other uses including mining, forestry, and row crop agriculture (Stubblefield et al. 2005). The Zaamar Goldfield is the largest gold mining site within the Tuul River Basin and stretches for about 60 km along the Tuul River (Chalov et al. 2015). The mining activities of many mining companies within the Zaamar Goldfield, as well as illegal mining, are reported to have a serious impact on the water quality within the river basin (Chalov et al. 2012). Another anthropogenic activity influencing the Selenga River Basin is intensive agriculture. Over 60% of Mongolian agricultural products are produced in the Selenga River Basin (UNEP-NISD 2008). Land degradation caused by mining does not cover a large land area as agricultural land and grassland account for the bulk of the area (Sinha and Balganjav 2011; Priess et al. 2015).

112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145

PR O O

Materials and methods

N C O R R EC TE D

mobilize hydrophobic contaminants in aquatic systems through complexation (Chiou et al. 1986). Therefore, the interaction with HS must be considered to appropriately understand metal and contaminant transport and subsequent effects on the aquatic ecology. However, although HS generally predominate in DOM, both HS concentration and quality can considerably vary among different systems (e.g., Spencer et al. 2012; Tsuda et al. 2016) and even within a given system (Kida et al. 2015), which further complicates understanding of metal and contaminant dynamics. The Selenga River Basin, located in Russia and Mongolia, has suffered from heavy metal contamination mainly due to placer gold mining and industrial activities (Pfeiffer et al. 2015; Hofmann et al. 2015). The Selenga River contributes more than half of the total inflow to Lake Baikal, the world’s deepest and the largest freshwater resource, and has been designated a UNESCO World Heritage Site. Over the last 30 years, there has been rapid population growth and urbanization in Mongolia. Many of the tributaries of Selenga River, including Tuul River and Orkhon River, have been increasingly used for placer gold mining (see Farrington (2000) for a detailed description). The placer gold mining industry in Mongolia has increased by 17 times in the last 10 years (Sorokovikova et al. 2013). According to Batimaa et al. (2011), a total of 784 enterprises are engaged in mining, of which 204 smallscale gold mining companies were operating on 6,065,298 ha of land. These large changes have caused major impacts on the water cycle and water quality of the Selenga River Basin. The currently biggest scientific and political problem concerning environmental risk assessment in the Selenga River Basin is a scarcity in data. Karthe et al. (2015) described Mongolia as not only a water-scarce but also a data-scarce country with regard to environmental information. As comprehensive monitoring of rivers in Mongolia is still in its initial stage, published data have been limited to basic information such as surface and ground water quality (Byambaa and Todo 2011; Hofmann et al. 2011; Altansukh et al. 2012) and sediment load to Lake Baikal (Chalov et al. 2015; Myangan et al. 2017). Recently, an increasing amount of literature has focused on metal contamination (Batjargal et al. 2010; Brumbaugh et al. 2013; Pfeiffer et al. 2015; Lychagin et al. 2017; Batbayar et al. 2017; Thorslund et al. 2017). However, there is little study of DOM (e.g., Batsaikhan et al. 2017) except for a few detailed studies in Lake Baikal (Yoshioka et al. 2002; Sugiyama et al. 2014). Considering the above-mentioned importance of HS in metal mobilization and the generally high-pH water conditions in the Selenga River Basin, which lowers the solubility of metals and enhances precipitation or adsorption to suspended solids (SS) (Jamiyanov et al. 2011; Thorslund et al. 2014), it is important to investigate HS distribution in the Selenga River Basin.

U

44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96

AUTHOR'S PROOF!


Environ Sci Pollut Res Fig. 1 Location of the sampling sites in the Selenga River Basin. Dashed line indicates the border between Mongolia and Russia, while solid lines indicate rivers. Water flow directions are from rivers to Lake Baikal. Names of rivers and lakes are in italics. Numbers for sampling points are described in Table 1

N

26–28 Baikal

22–25

20

21 Ulan-Ude 19

35,Gusinoye

18 32,Khuvsgul

RUSSIA 30

14–16

Eg

MONGOLIA

10

31, Delgermurun

PR O O

Selenga

6

Orkhon

Zaamar Goldfield

N C O R R EC TE D

river sample lake sample

100 km

Sampling

147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170

Two sampling campaigns were conducted in August of 2013 and 2014, both during the rainy season with the second highest peak in river hydrographs following the spring ice and snowpack melt (Chalov et al. 2015). The rivers Tuul, Orkhon, Kharaa, Sharyn, Yeroo, Eg, Delgermurun, Buir, and Selenga and the lakes Gusinoye, Khuvsgul, Sharyn Tsagaan, Erkhel, and Baikal were sampled with the main sampling focused along the Tuul–Orkhorn–Selenga river transect. Sampling sites were selected mainly at tributary junctions and set both upstream and downstream from the junctions. Specific sampling site locations are provided in Fig. 1. After the confirmation of uniformity in water properties by a multiparameter meter (HI 9828, Hanna Instruments, RI, USA), surface water samples for DOM and HS quantification were directly taken with 280-mL volume polyethylene terephthalate bottles after rinsing three times with the sampled water. Water samples for trace metal quantification were also collected directly into plastic bottles or using a water sampler (Van Dorn water sampler, RIGO, Tokyo, Japan), as described in Myangan et al. (2017). The water samples were collected without spatial replicates because of varying widths and depths of the rivers. It is particularly difficult to collect replicates in small tributaries with narrow width and shallow depth.

U

146

Yeroo

F

33,Erkhel

17 29 11–13 Sharyn

9 34,Sharyn 8

7

5

Tsagaan

Kharaa

4

3 Tuul

Ulaanbaatar

2

1

The samples for DOM and HS quantification were filtered through glass fiber filters (GF-F, Advantec, Tokyo, Japan) on site and stored in a cooler. The samples for trace metal quantification were filtered with the GF-F and divided evenly into two subsamples. One set was acidified with concentrated HNO3 for determination of the total metal concentration. The other set was not acidified and kept in plastic tubes. All samples were sent back to Japan for further analysis. The pH and electrical conductivity (EC) were recorded in situ using the multiparameter meter.

171 172 173 174 175 176 177 178 179 180

Analysis

181

The filtrates were quantitatively analyzed for DOM and HS as reported (Tsuda et al. 2012). Samples of L. Erkhel (Point 33) and L. Sharyn Tsagaan (Point 34) were diluted before analysis to achieve dissolved organic carbon (DOC) concentration ≤ 40 mg C L−1 to keep them in the range of the applied method. Briefly, 0.4 mL of purified DAX-8 resin (DOC bleed < 0.05 mg C L−1) and each sample were put in a glass vial in triplicate with a resin:liquid ratio of 1:50 (v/v) and acidified with 1 mL 1 M H2SO4 to adjust the pH to < 2, consuming < 100 mL of each sample. After shaking for 24 h in the dark at 5 °C, the DOC concentration of the supernatant (the non-HS [NHS] fraction) was analyzed after filtration through the GF-F filter. The HS concentration was then calculated as the

182 183 184 185 186 187 188 189 190 191 192 193 194

AUTHOR'S PROOF!



Results and discussion

236

General DOM and HS distribution

237 238 239 240 241 242 243

The quantification results of DOM and HS carbon concentrations for the surface waters are presented in Table 1. More than two thirds of the sampled waters in the Selenga River Basin had DOC concentrations between 2.0 and 8.0 mg C L−1. In river samples, the DOC concentration ranged from 0.99 mg C L−1 at Point 30 (upper stream of the River Eg) to 22.7 mg C L−1 at Point 29 (River Buir, brown-colored water from a

244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277

%HS along the Tuul–Orkhorn–Selenga transect

278

Figure 2 shows the relationships between DOM and HS concentrations for water samples along the Tuul–Orkhorn– Selenga transect (Points 1–25). The relationships were determined as follows: y = 0.643×, R2 = 0.996 (P < 0.001). Note that we set the y intercept as zero because DOM without HS has not yet been reported (Watanabe et al. 2012). Surprisingly, there was little fluctuation in %HS (61.7–67.7%, CV 2.0%) from the upper stream of Ulaanbaatar City to the Selenga delta with a total distance of > 1000 km and mixing of several rivers of various sizes, despite the fact that samples were collected in two different years. This result is highly relevant to metal speciation models, such as the Humic Ion-binding Model VI (Tipping 1998) and Stockholm Humic Model (SHM) (Gustafsson 2001). In the modeling, users need to input the proportion of the “active” part of DOM to complex with metals. Most often, it is assumed that the active DOM is HS

279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294

F

235

bog) with the median value of 5.08 mg C L−1. The proportion of HS (%HS) ranged from 36.1% at Point 30 (upper stream of the River Eg) to 67.7% at Point 13 (River Yeroo) with a median of 63.7%. The majority of the samples had %HS higher than 60%. These values are higher than those in Lake Biwa and other 20 freshwater lakes in Japan (mean ± standard deviation; 44.1 ± 5.7 and 48.5 ± 7.1%, respectively) (Kida et al. 2015; Tsuda et al. 2016) and an average for 30 US rivers (47.3 ± 7.9%) (Spencer et al. 2012). The DOC concentrations in two lakes, Erkhel and Sharyn Tsagaan, were as high as 48.3 and 142.7 mg C L−1, respectively, which were considerably higher than DOC concentrations in the rivers. This was attributed to evaporative concentration of DOC (Curtis and Adams 1995; Osburn et al. 2011) because these two lakes had no outflowing river and EC values were very high (3268 and 2853 mS m−1, respectively; Table 1). It is likely that the harsh semiarid continental climate of the basin increases the evaporation rate (Hülsmann et al. 2015). The Buir River (Point 29) was brown colored and had high DOC concentration (22.7 mg C L−1), reflecting peat-derived DOC input from the bog. River water samples along the Tuul–Orkhorn– Selenga transect (Points 1–25) had DOC concentrations between 3.66 and 9.73 mg C L−1, while Lake Baikal had low DOC concentrations at every sampling point (≤ 1.4 mg C L−1). The %HS in freshwater lakes (Lakes Baikal, Khuvsgul, and Gusinoye) was lower than in rivers (45.4 ± 7.3 versus 62.4 ± 5.6%, respectively). Hayakawa et al. (2003) suggested that flocculation and photobleaching could represent major mechanisms to selectively remove HS in Lake Khuvsgul. Generally, selective flocculation, photodegradation, and subsequent microbial processing of riverine-derived, HS-rich DOM in lakes are well understood (e.g., Farjalla et al. 2009; Helms et al. 2013; Weyhenmeyer et al. 2014), and the same mechanisms seem to hold for the freshwater lakes in this study.

PR O O

difference between the bulk DOC and NHS concentrations. Organic carbon contamination from the system was also determined in quintuplicate with ultrapure water (18.2 MΩ; Advantec, Tokyo, Japan) as a blank and used for correction. The DOC was measured as non-purgeable organic carbon (NPOC) by Pt-catalyzed high-temperature combustion in a TOC-VCPH analyzer (Shimadzu, Kyoto, Japan). The sample was first sparged with a carrier gas for 90 s in the built-in syringe of the analyzer to remove any inorganic carbon prior to combustion. Calibration was performed by running five standards of a potassium hydrogen phthalate (Wako, Tokyo, Japan) solution over an appropriate range and one laboratory blank (i.e., ultrapure water). The elemental concentrations were calculated from the regression line (coefficient of determination, R2 > 0.9999). Reported DOC concentrations are average values of triplicate measurements. If the coefficient of variation (CV) exceeded 2.0% (about 40% of the all measurements), up to two additional analyses were performed and outliers were eliminated. Metal concentrations (Al and Fe) were measured in two fractions: non-acidified water samples from the field were filtered through a 0.025-μm pore size polysulfonate membrane filter (Millipore, Tokyo, Japan) to determine concentrations of dissolved metals (determined here as < 0.025 μm). The GF-F filtered samples were used for the determination of the total Al and Fe concentrations. Metal concentrations of the GF-F filtered and membrane filtered samples were analyzed with inductively coupled plasma atomic emission spectrometry (ICP-AES; ICPE-9000, Shimadzu, Kyoto, Japan). Subsequently, the colloidal metal concentrations were calculated as the difference in metal concentrations between the GF-F filtered and membrane filtered samples. The detection limit of the analyzer was approximately 10 μg L−1. Details of the ICP-AES analysis and sample handling can be found elsewhere (Myangan et al. 2017). In this study, if the dissolved metal concentrations were below the detection limit, colloidal metal concentrations were set equal to the corresponding total metal concentrations. All chemicals used were special grade and all glassware used was acid-washed and muffled at 450 °C for more than 3 h before use.

U

N C O R R EC TE D

195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234

AUTHOR'S PROOF!



4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 29 30 31 Lakes 26 27 28 32 33 34 35

pH

EC (mS m−1)

3.76 3.14 2.76

65.4 64.2 63.4

ND ND 7.85

ND ND 19.6

8.38 9.73 3.75 6.10 5.05 5.56 5.10 7.86 5.41 6.62 5.76 5.69 7.63 7.34 3.92 4.29 3.77 3.68

5.35 6.27 2.31 3.89 3.22 3.57 3.18 5.11 3.46 4.48 3.70 3.62 5.06 4.68 2.44 2.74 2.35 2.34

63.8 64.4 61.7 63.8 63.6 64.3 62.4 65.0 63.9 67.7 64.2 63.7 66.3 63.8 62.3 63.9 62.4 63.6

8.03 8.19 7.95 8.39 8.07 8.39 8.47 8.17 8.35 7.66 7.93 7.82 7.81 7.80 8.21 8.13 8.15 8.22

33.2 34.5 20.0 28.1 27.6 27.7 26.7 25.7 30.7 7.8 19.5 20.0 19.4 20.1 28.0 20.7 18.3 17.8

106.4923 106.3981 106.3286 106.2837 106.2009 100.0555 100.1512

3.67 3.67 3.70 3.66 22.7 0.99 1.74

2.29 2.31 2.33 2.32 13.7 0.36 0.94

62.5 63.1 63.0 63.3 60.3 36.1 54.4

8.24 8.01 8.11 8.04 7.91 8.31 8.20

17.9 19.6 19.3 20.6 96.0 29.7 33.0

106.1715 106.1640 106.2063 100.4023 99.9162 106.0290 106.4479

1.36 1.24 1.41 1.27 48.3 142.7 6.11

0.65 0.62 0.67 0.41 18.0 63.2 2.98

47.9 50.3 47.7 32.4 37.2 44.3 48.7

7.82 7.69 7.60 8.33 8.96 10.1 8.22

13.6 15.0 14.9 27.5 3268 2853 498

Latitude (N, decimal)

Longitude (E, decimal)

DOM (mg C L−1)

14 River Tuul 14 River Tuul 13 River Tuul

47.8780 47.8899 47.8628

107.6121 106.9099 105.1976

5.74 4.90 4.36

13 River Tuul 13 River Tuul 13 River Orkhon 13 River Orkhon 13 River Orkhon 13 River Kharaa 13 River Orkhon 13 River Sharyn 13 River Orkhon 13 River Yeroo 13 River Orkhon 13 River Orkhon 13 River Selenga 13 River Selenga 14 River Selenga 14 River Selenga 14 River Selenga 14 River Selenga

48.4897 48.8242 48.9666 48.9663 49.1283 49.4893 49.7007 49.8572 49.8821 49.9146 50.2417 50.2525 50.2525 50.3125 51.0933 51.7367 52.0326 52.0332

104.5472 104.8088 104.8769 104.8770 105.3453 105.8954 105.9028 106.1357 106.1474 106.1623 106.1699 106.1378 106.1378 106.1335 106.6459 107.4672 107.5018 106.8248

14 River Selenga 14 River Selenga 14 River Selenga 14 River Selenga 13 River Buir (colored water) 14 upper stream of River Eg 14 River Delgermurun

52.1907 52.2155 52.2198 52.2616 50.2315 50.1156 49.6007

14 Lake Baikal 14 Lake Baikal 14 Lake Baikal 14 Lake Khuvsgul 14 Lake Erkhel 13 Lake Sharyn Tsagaan 14 Lake Gusinoye

52.1820 52.2575 52.0872 50.7224 49.9532 49.6464 51.2936

HS (mg C L−1)

F

Rivers 1 2 3

%HS (%)

Description

PR O O

Sampling point

Description, location, and chemical characteristics of samples

N C O R R EC TE D

Table 1

U

t1:1

Numbers attached before sample descriptions indicate sampling year (i.e., 2013 and 2014) ND not determined

295 296 297 298 299 300 301

and 65% of DOM are present as HS while 35% are inert (Bryan et al. 2002), although NHS such as amino acids and saccharides can contribute to metal binding in some systems. This assumption is very close to our findings (64.3%). However, we notice that a few studies (e.g., Thorslund et al. 2014) applied SHM to the water samples collected in the Selenga River Basin, assuming a HS/DOC ratio of 1.65,

which corresponds to %HS of 82.5% when HS is assumed to be 50% carbon. Although metal speciation modeling is important and such study is encouraged especially in areas such as the Selenga River Basin because of well-known metal contamination, we highly recommend application of the value of 64.3% (possibly 65%) for rivers and 45.4% for freshwater lakes in the Selenga River Basin to improve model predictions

302 303 304 305 306 307 308

AUTHOR'S PROOF!



HS (mgC L−1)

6

y = 0.643x R² = 0.996 P < 0.001

5

4

3 L. Gusinoye 2 3

5

7

9

11

DOM (mgC L−1)

Fig. 2 Relationships between DOM and HS concentrations for the river samples along the Tuul–Orkhon–Selenga River transect (Nos. 1–25). Open circles are samples collected in 2013 while plus marks in 2014. Solid line indicates a regression line for the river samples by least squares method. Lake Gusinoye (filled circle) is shown for reference and not included in the regression

in future research. This could also lead to improving predictions of organo-metal transport. The low variability in %HS is unusual in other systems; in general, %HS fluctuates even within a given system if sampled from different horizontal or vertical points (Kida et al. 2015). Several reasons could be proposed for the uniformity of the %HS, such as similar geological conditions along the rivers or terrestrial DOM input with similar composition, since the sampling campaigns were conducted in the same month when the land was apparently covered by identical plant species. Sinyukovich et al. (2010) also attributed low variability of the chemical compositions of Selenga waters to similar geological and climatic conditions throughout the basin, including Mongolia. In addition, increases to river flows under high precipitation during summertime (Karthe et al. 2014) may have lessened the opportunities both for photodegradation of DOM by enhanced turbidity and for microbial processing of DOM by decreasing water residence time within the Tuul–Baikal transect, both of which can often cause DOM composition to shift in opposite directions (Hansen et al. 2016). Another explanation could be a large contribution by groundwater in the Selenga River Basin because of the extensive areas of alluvial unconfined aquifers, which may narrow the variation in %HS through high degrees of water-rock interaction (Thorslund et al. 2017).

334 335

DOM distribution along the Tuul–Orkhorn–Selenga transect to Lake Baikal

336 337

Since DOM and HS concentrations fluctuated in parallel, we focused on the changes in DOM concentrations along the

338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376

Metal distribution along the Tuul–Orkhorn–Selenga transect

377 378

Table 2 shows the metal (Al and Fe) concentrations in the three different size fractions. There was little detection of Al and Fe in the dissolved fraction and they mostly existed as colloids. The low solubility of Al and Fe was also estimated by Visual MINTEQ speciation modeling in the study of Thorslund et al. (2014) for river samples collected in the Tuul River. Prevailing high-pH conditions (usually > 8) in the Selenga River Basin generally limited the dissolution of metals. However, there was an exception observed: in the River Yeroo and its downstream (Points 13 and 14), 58 and

379 380 381 382 383 384 385 386 387 388

U

N C O R R EC TE D

309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333

Tuul–Orkhorn–Selenga River transect including Lake Baikal to clarify driving factors that affected DOM concentrations (Fig. 3). The major changes in DOM concentrations occurred for the following three reasons: (1) input from the Zaamar Goldfield, (2) mixing with large rivers with different DOM concentrations, and (3) dilution at the inflow to Lake Baikal. The clearest evidence of impact on DOM concentration along the transect was found in the Tuul River where the Zaamar Goldfield is located. The DOM concentration increased by a factor of two from 4.36 to 9.73 mg C L−1 before and after the river passed through the mining area. Considering there was no detected increase in DOM concentration at Ulaanbaatar, where more than 1.3 million people (nearly half of Mongolia’s population) is concentrated, the effect of the Zaamar Goldfield on the river quality and material loading may be considered to be substantial. Consistent material inputs at the Zaamar Goldfield were reported in the previous literature such as suspended sediment (Chalov et al. 2015), heavy metals (Thorslund et al. 2014), and basic dissolved anions including Cl−, SO42−, and HCO3− (Chalov et al. 2012). The influence, however, was mitigated by mixing with the larger Orkhon River with lower DOM concentration at Point 7. After mixing with several smaller tributaries including the rivers Kharaa, Sharyn, and Yeroo, the Orkhon River is mixed with the larger Selenga River at Point 17. The higher DOM concentration of the Selenga River (7.63 mg C L−1) increased that of the Orkhon River from 5.69 to 7.34 mg C L−1. The DOM concentration decreased between Points 17 and 18, which may indicate a yearly difference, or removal of DOM under physicochemical processes such as sorption to SS, flocculation, or precipitation by changing water qualities, then remained stable until the end of the delta. The decrease of DOM within the delta was therefore not observed, which is contrary to the previously reported decrease in suspended and sediment materials (Chalov et al. 2017). In Lake Baikal, the DOM concentrations were significantly low, and this was attributed to the flocculation, photodegradation, and subsequent microbial processing within the lake (Farjalla et al. 2009; Helms et al. 2013; Weyhenmeyer et al. 2014).

F

N = 25

PR O O

7

AUTHOR'S PROOF!


Rv Selenga

Rv Yeroo

Rv Orkhon

PR O O

F

2

0

L Baikal

4

DELTA

Ulan-Ude

6

Rv Sharyn

8

Ulaanbaatar

DOM (mgC L−1)

10

Rv Kharaa

Zammar

DOM

RUSSIA

12

Fig. 3 Downstream profiles of DOC concentration through the Tuul–Orkhon–Selenga River transect to Lake Baikal (distances are not considered). Dotted lines indicate entrances of tributaries. Samples collected in the Selenga River Delta are shaded. Numbers for sampling points are described in Table 1

MONGOILA


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

18 19

20 21 22 23 24 25 26 27 28

Sampling point

SS concentration in the Yeroo River due to the lower exploited area could also prevent precipitation of the metals by sorption onto SS (Myangan et al. 2017).

N C O R R EC TE D

t2:1 t2:2

82% of Al and 91 and 98% of Fe, respectively, were found in the dissolved fractions. This was likely a result of the low pH and EC conditions in the Yeroo River (Table 1). The very low Table 2 Concentrations of total, dissolved, and colloidal metals in river waters from the Selenga River and its tributaries

U

389 390 391

Total Al (mg L−1)

Fe (mg L−1)

Dissolved Al (mg L−1)

Fe (mg L−1)

Colloidal Al (mg L−1)

Fe (mg L−1)

1 2 3 4 5 6 7 8 9 10 11

0.025 0.022 0.147 0.225 0.359 0.120 0.189 0.122 0.110 0.168 0.238

0.056 0.047 0.080 0.128 0.265 0.069 0.124 0.092 0.082 0.114 0.119

0.016 0.019 – – 0.025 – 0.013 – – – –

– – – 0.027 0.051 – 0.031 – – 0.019 –

0.010 0.003 0.147 0.225 0.334 0.120 0.176 0.122 0.110 0.168 0.238

0.056 0.047 0.080 0.101 0.214 0.069 0.093 0.092 0.082 0.096 0.119

12 13 14 15 16 17 18 19 20 21 22 23 24

0.109 0.196 0.136 0.165 0.142 0.166 0.053 0.017 0.046 0.023 0.038 0.065 0.078

0.064 0.134 0.098 0.224 0.112 0.114 0.052 0.040 0.071 0.055 0.069 0.089 0.105

– 0.113 0.112 0.013 – – – – 0.028 0.023 – – –

– 0.122 0.095 0.033 0.030 0.020 – – – – – – –

0.109 0.083 0.024 0.152 0.142 0.166 0.053 0.017 0.018 0.0 0.038 0.065 0.078

0.064 0.012 0.003 0.191 0.082 0.094 0.052 0.040 0.071 0.055 0.069 0.089 0.105

Sampling point

Data of sample nos. 3–17 were obtained from Myangan et al. (2017) − below the detection limit

392 393 394

AUTHOR'S PROOF!



Given the strong correlations between DOM and HS (Fig. 2), we present the relationships between DOM and the dissolved and colloidal metals in Fig. 4. There was no significant correlation between the dissolved metals and DOM concentrations (P > 0.05) even when excluding the outliers (Points 13 and 14). In contrast, the colloidal metals showed strong correlations with DOM concentrations. This strongly suggests that DOM sustained colloidal Al and Fe in the solution and they were co-transported in the Selenga River Basin. This is also important in controlling the behavior and transportation of other heavy metals such as As, Cd, and Pb, because their solubility is controlled by sorption onto Al and Fe-(oxy)hydroxide colloids such as gibbsite and ferrihydrite (Eary 1999). Myangan et al. (2017) showed that the source of Al and Fe was surface soil in watersheds, which was confirmed by a constant Al and Fe composition in the SS. Therefore, land use and land cover changes in watersheds can also strongly affect the downstream loads of heavy metals. Furthermore, heavy metal partitioning between the colloidal and dissolved fractions depends on the water quality, and such speciation affects the metal bioavailability and toxicity (Taylor

No. 13

0.09

No. 14

0.06

0.03

0.00 3.0

5.0

7.0

9.0

11.0

DOM (mgC L−1)

F

425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445

Al Fe

0.12

0.4

Colloidal

PR O O

Relationships between DOM and metals

Dissolved

Dissolved metal (mg L−1)

424

0.15

Al Fe

Colloidal metal (mg L−1)

Similar to DOM, the major changes in metal concentrations occurred by (1) input from the Zaamar Goldfield, (2) mixing with large rivers with different metal concentrations, and (3) dissolution of sedimentary metals in the delta. There was no increase in metal concentrations at Ulaanbaatar. The Zaamar Goldfield had a major impact not only on DOM but also on metals. The total Al and Fe concentrations more than doubled through the mining area. The influence was rather masked by mixing with the larger Orkhon River with lower Al and Fe concentrations, but the metal concentration of the Orkhon River increased by a factor of 1.5 after mixing. The metal concentrations in the rivers Sharyn and Yeroo were higher than the main river. Interestingly, the metal concentrations decreased between Points 17 and 18, as in DOM. The reasons for this decrease are currently unclear, but a possible explanation is that slight increases in pH and EC, together with a large decrease in DOM, enhanced metal precipitation, or sorption onto SS during the > 100-km journey between the two points. Finally, the metal concentrations increased gradually within the delta, by up to two times, although the DOM did not increase. We attributed this to dissolution of metals from delta sediment by DOM (“peptization”), and especially by HS with acidic functional groups such as carboxyl and phenolic hydroxyl groups. Since metals bound to DOM are more resistant to precipitation and available for phytoplankton, this organometal complex is expected to be important in sustaining the primary production in Lake Baikal. Future research is required to quantify the contribution of this organo-metal complex to the production in the great lake.

N C O R R EC TE D

395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423

0.3

(r = 0.81, P < 0.01)

0.2

0.1 (r = 0.60, P < 0.01) 0.0 3.0

5.0

7.0

9.0

11.0

DOM (mgC L−1)

Fig. 4 Relationships between DOM and metal concentrations of different size fraction

446 447 448 449

Conclusion

450

Despite previous research over the last decade, the Selenga River Basin still remains a data-scarce region. This study provides important %HS distribution data, which can be used for better metal speciation modeling. Since speciation of metals critically affect their mobility, bioavailability, and toxicity, better prediction of nature of organo-metal complexes has important environmental applications. Although variabilities in DOM and metal concentration were clearly dependent on the mixing of rivers with different size and water conditions, Al and Fe were mainly present as organo-metal colloids. These organo-metal colloids are important in controlling the behavior and transportation of other heavy metals. In the Selenga Delta, Al and Fe were solubilized from the sediment

451 452 453 454 455 456 457 458 459 460 461 462 463

U

et al. 2015; Blazevic et al. 2016). The observed changes in pH and EC values at the tributary junctions and along the downstream flow (Table 1) therefore have major implications for the fate of metals.

AUTHOR'S PROOF!



473 474 475 476 477 478

Acknowledgements This study was financially supported by National University of Mongolia (grant number P2016-1222) and JSPS KAKENHI grant number 25304001. We would like to thank K. Maki from Kobe University for the assistance in data collection.

479 480

Conflict of interest The authors declare that they have no conflict of interest.

481 482

References

483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519

Altansukh O, Whitehead P, Bromley J (2012) Spatial patterns and temporal trends in the water quality of the Tuul River in Mongolia. Energy Environ Res 2(1):62–78. https://doi.org/10.5539/eer. v2n1p62 Batbayar G, Pfeiffer M, von Tümpling W, Kappas M, Karthe D (2017) Chemical water quality gradients in the Mongolian sub-catchments of the Selenga River basin. Environ Monit Assess 189(8):420–427. https://doi.org/10.1007/s10661-017-6123-z Batimaa P, Myagmarjav B, Batnasan N et al (2011) Urban water vulnerability to climate change in Mongolia. Minist Nature Environ Tour Rep 78P Batjargal T, Otgonjargal E, Baek K, Yang J-S (2010) Assessment of metals contamination of soils in Ulaanbaatar, Mongolia. J Hazard Mater 184(1-3):872–876. https://doi.org/10.1016/j.jhazmat.2010. 08.106 Batsaikhan B, Kwon J-S, Kim K-H, Lee YJ, Lee JH, Badarch M, Yun ST (2017) Hydrochemical evaluation of the influences of mining activities on river water chemistry in central northern Mongolia. Environ Sci Pollut Res 24(2):2019–2034. https://doi.org/10.1007/s11356016-7895-3 Bell NGA, Michalchuk AAL, Blackburn JWT, Graham MC, Uhrín D (2015) Isotope-filtered 4D NMR spectroscopy for structure determination of humic substances. Angew Chemie Int Ed 54(29):8382– 8385. https://doi.org/10.1002/anie.201503321 Bergamaschi BA, Krabbenhoft DP, Aiken GR, Patino E, Rumbold DG, Orem WH (2012) Tidally driven export of dissolved organic carbon, total mercury, and methylmercury from a mangrove-dominated estuary. Environ Sci Technol 46(3):1371–1378. https://doi.org/10. 1021/es2029137 Blazevic A, Orlowska E, Kandioller W, Jirsa F, Keppler BK, TafiliKryeziu M, Linert W, Krachler RF, Krachler R, Rompel A (2016) Photoreduction of terrigenous Fe-humic substances leads to bioavailable iron in oceans. Angew Chemie Int Ed 55(22):6417– 6422. https://doi.org/10.1002/anie.201600852 Brumbaugh WG, Tillitt DE, May TW, Javzan C, Komov VT (2013) Environmental survey in the Tuul and Orkhon River basins of north-central Mongolia, 2010: metals and other elements in

U

N C O R R EC TE D

Compliance with ethical standards

streambed sediment and floodplain soil. Environ Monit Assess 185(11):8991–9008. https://doi.org/10.1007/s10661-013-3229-9 Bryan SE, Tipping E, Hamilton-Taylor J (2002) Comparison of measured and modelled copper binding by natural organic matter in freshwaters. Comp Biochem Physiol Part C Toxicol Pharmacol 133(1-2): 37–49. https://doi.org/10.1016/S1532-0456(02)00083-2 Byambaa B, Todo Y (2011) Technological impact of placer gold mine on water quality: case of Tuul River Valley in the Zaamar Goldfield, Mongolia. World Acad Sci Eng Technol 51:167–171 Chalov SR, Zavadsky AS, Belozerova EV, Bulacheva MP et al (2012) Suspended and dissolved matter fluxes in the upper Selenga river basin. Geogr Environ Sustain 5(03):78–94. https://doi.org/10.1017/ S0376892900029301 Chalov SR, Jarsjö J, Kasimov NS et al (2015) Spatio-temporal variation of sediment transport in the Selenga River Basin, Mongolia and Russia. Environ Earth Sci 73(2):663–680. https://doi.org/10.1007/ s12665-014-3106-z Chalov S, Thorslund J, Kasimov N, Aybullatov D, Ilyicheva E, Karthe D, Kositsky A, Lychagin M, Nittrouer J, Pavlov M, Pietron J, Shinkareva G, Tarasov M, Garmaev E, Akhtman Y, Jarsjö J (2017) The Selenga River delta: a geochemical barrier protecting Lake Baikal waters. Reg Environ Chang 17(7):2039–2053. https:// doi.org/10.1007/s10113-016-0996-1 Chiou CT, Malcolm RL, Brinton TI, Kile DE (1986) Water solubility enhancement of some organic pollutants and pesticides by dissolved humic and fulvic acids. Environ Sci Technol 20(5):502–508. https:// doi.org/10.1021/es00147a010 Curtis PJ, Adams HE (1995) Dissolved organic matter quantity and quality from freshwater and saltwater lakes in east-central Alberta. Biogeochemistry 30(1):59–76. https://doi.org/10.1007/ BF02181040 Eary LE (1999) Geochemical and equilibrium trends in mine pit lakes. Appl Geochem 14(8):963–987. https://doi.org/10.1016/S08832927(99)00049-9 Farjalla VF, Amado AM, Suhett AL, Meirelles-Pereira F (2009) DOC removal paradigms in highly humic aquatic ecosystems. Environ Sci Pollut Res 16(5):531–538. https://doi.org/10.1007/s11356-0090165-x Farrington J (2000) Environmental problems of placer gold mining in the Zaamar Goldfield, Mongolia. World Placer J 1:107–128 Gustafsson JP (2001) Modeling the acid–base properties and metal complexation of humic substances with the Stockholm Humic Model. J Colloid Interface Sci 244(1):102–112. https://doi.org/10.1006/jcis. 2001.7871 Hansen AM, Kraus TEC, Pellerin BA, Fleck JA, Downing BD, Bergamaschi BA (2016) Optical properties of dissolved organic matter (DOM): effects of biological and photolytic degradation. Limnol Oceanogr 61(3):1015–1032. https://doi.org/10.1002/lno. 10270 Hayakawa K, Sekino T, Yoshioka T, Maruo M, Kumagai M (2003) Dissolved organic carbon and fluorescence in Lake Hovsgol: factors reducing humic content of the lake water. Limnology 4(1):25–33. https://doi.org/10.1007/s10201-003-0092-3 Helms JR, Mao J, Schmidt-Rohr K, Abdulla H, Mopper K (2013) Photochemical flocculation of terrestrial dissolved organic matter and iron. Geochim Cosmochim Acta 121:398–413. https://doi.org/ 10.1016/j.gca.2013.07.025 Hofmann J, Hürdler J, Ibisch R, Schaeffer M, Borchardt D (2011) Analysis of recent nutrient emission pathways, resulting surface water quality and ecological impacts under extreme continental climate: the Kharaa River Basin (Mongolia). Int Rev Hydrobiol 96(5): 484–519. https://doi.org/10.1002/iroh.201111294 Hofmann J, Watson V, Scharaw B (2015) Groundwater quality under stress: contaminants in the Kharaa River basin (Mongolia). Environ Earth Sci 73(2):629–648. https://doi.org/10.1007/s12665014-3148-2

F

by complexation with HS, and the metal concentrations were doubled. This study highlights the importance of DOM in metal transportation in the Selenga River Basin. Finally, we need to stress that the Zaamar Goldfield maintained (in our 2014 field campaign) a major impact on the water quality of the Tuul River. Water quality monitoring in the Tuul River needs to be continued for assessing the transport and load of contaminants within the basin, and potentially for Lake Baikal.

PR O O

464 465 466 467 468 469 470 471 472

520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585

AUTHOR'S PROOF!



PR O O

F

Mongolia at the beginning of the 21st century. Russ Meteorol Hydrol 38(2):126–133. https://doi.org/10.3103/ S1068373913020106 Spencer RG, Butler KD, Aiken GR (2012) Dissolved organic carbon and chromophoric dissolved organic matter properties of rivers in the USA. J Geophys Res 117(G3):14P. https://doi.org/10.1029/ 2011JG001928 Stubblefield A, Chandra S, Eagan S, Tuvshinjargal D, Davaadorzh G, Gilroy D, Sampson J, Thorne J, Allen B, Hogan Z (2005) Impacts of gold mining and land use alterations on the water quality of central Mongolian rivers. Integr Environ Assess Manag 1(4):365– 373. https://doi.org/10.1002/ieam.5630010406 Sugiyama Y, Hatcher PG, Sleighter RL, Suzuki T, Wada C, Kumagai T, Mitamura O, Katano T, Nakano SI, Tanaka Y, Drucker VV, Fialkov VA, Sugiyama M (2014) Developing an understanding of dissolved organic matter dynamics in the giant Lake Baikal by ultrahigh resolution mass spectrometry. Limnology 15(2):127–139. https://doi. org/10.1007/s10201-014-0424-5 Taylor NS, Kirwan JA, Yan ND, Viant MR, Gunn JM, McGeer JC (2015) Metabolomics confirms that dissolved organic carbon mitigates copper toxicity. Environ Toxicol Chem 35(3):635–644. https://doi.org/ 10.1002/etc.3206 Thorslund J, Jarsjö J, Wällstedt T, Mörth CM, Lychagin MY, Chalov SR (2014) Geochemical controls on the partitioning and hydrological transport of metals in a non-acidic river system. Hydrol Earth Syst Sci Discuss 11(8):9715–9758. https://doi.org/10.5194/hessd-119715-2014 Thorslund J, Jarsjö J, Wällstedt T, Mörth CM, Lychagin MY, Chalov SR (2017) Speciation and hydrological transport of metals in non-acidic river systems of the Lake Baikal basin: field data and model predictions. Reg Environ Chang 17(7):2007–2021. https://doi.org/10. 1007/s10113-016-0982-7 Tipping E (1998) Humic Ion-Binding Model VI: an improved description of the interactions of protons and metal ions with humic substances. Aquat Geochem 4(1):3–48. https://doi.org/10.1023/A: 1009627214459 Tsuda K, Takata A, Shirai H et al (2012) A method for quantitative analysis of aquatic humic substances in clear water based on carbon concentration. Anal Sci 28(10):1017–1020. https://doi.org/10.2116/ analsci.28.1017 Tsuda K, Kida M, Aso S, Kato T, Fujitake N, Maruo M, Hayakawa K, Hirota M (2016) Determination of aquatic humic substances in Japanese lakes and wetlands by the carbon concentration-based resin isolation technique. Limnology 17(1):1–6. https://doi.org/10. 1007/s10201-015-0455-6 UNEP-NISD (2008) Integrated water management model on the Selenga River Basin: Status survey and investigation (Phase I). KEI 419P Watanabe A, Moroi K, Sato H, Tsutsuki K, Maie N, Melling L, Jaffé R (2012) Contributions of humic substances to the dissolved organic carbon pool in wetlands from different climates. Chemosphere 88(10):1265–1268. https://doi.org/10. 1016/j.chemosphere.2012.04.005 Weyhenmeyer GA, Prairie YT, Tranvik LJ (2014) Browning of boreal freshwaters coupled to carbon-iron interactions along the aquatic continuum. PLoS One 9(2):7P. https://doi.org/10.1371/journal. pone.0088104 Yoshioka T, Ueda S, Khodzher T, Bashenkhaeva N, Korovyakova I, Sorokovikova L, Gorbunova L (2002) Distribution of dissolved organic carbon in Lake Baikal and its watershed. Limnology 3(3): 159–168. https://doi.org/10.1007/s102010200019

N C O R R EC TE D

Hülsmann L, Geyer T, Schweitzer C, Priess J, Karthe D (2015) The effect of subarctic conditions on water resources: initial results and limitations of the SWAT model applied to the Kharaa River Basin in northern Mongolia. Environ Earth Sci 73(2):581–592. https://doi. org/10.1007/s12665-014-3173-1 Jamiyanov DTD, Ulzetueva ID, Gomboev BO, Khakhinov VV (2011) Transboundary problems of water resources quality in the Selenga River Basin. Mineral Mag 75:2050 Karthe D, Kasimov NS, Chalov SR et al (2014) Integrating multi-scale data for the assessment of water availability and quality in the Kharaa-Orkhon-Selenga River system. Geogr Environ Sustain 3: 65–86. https://doi.org/10.15356/2071-9388 Karthe D, Hofmann J, Ibisch R, Heldt S, Westphal K, Menzel L, Avlyush S, Malsy M (2015) Science-based IWRM implementation in a datascarce central Asian region: experiences from a research and development project in the Kharaa River Basin, Mongolia. Water 7(12): 3486–3514. https://doi.org/10.3390/w7073486 Karthe D, Chalov S, Moreido V, Pashkina M, Romanchenko A, Batbayar G, Kalugin A, Westphal K, Malsy M, Flörke M (2017) Assessment of runoff, water and sediment quality in the Selenga River basin aided by a web-based geoservice. Water Resour 44(3):399–416. https://doi.org/10.1134/S0097807817030113 Kida M, Maki K, Takata A, Kato T, Tsuda K, Hayakawa K, Sugiyama Y, Fujitake N (2015) Quantitative monitoring of aquatic humic substances in Lake Biwa, Japan, using the DAX-8 batch method based on carbon concentrations. Org Geochem 83–84:153–157. https:// doi.org/10.1016/j.orggeochem.2015.03.015 Lychagin M, Chalov S, Kasimov N, Shinkareva G, Jarsjö J, Thorslund J (2017) Surface water pathways and fluxes of metals under changing environmental conditions and human interventions in the Selenga River system. Environ Earth Sci 76(1):1–14. https://doi.org/10. 1007/s12665-016-6304-z Myangan O, Kawahigashi M, Oyuntsetseg B, Fujitake N (2017) Impact of land uses on heavy metal distribution in the Selenga River system in Mongolia. Environ Earth Sci 76(9):346–360. https://doi.org/10. 1007/s12665-017-6664-z Osburn CL, Wigdahl CR, Fritz SC, Saros JE (2011) Dissolved organic matter composition and photoreactivity in prairie lakes of the U.S. Great Plains. Limnol Oceanogr 56(6):2371–2390. https://doi.org/ 10.4319/lo.2011.56.6.2371 Pfeiffer M, Batbayar G, Hofmann J, Siegfried K, Karthe D, Hahn-Tomer S (2015) Investigating arsenic (As) occurrence and sources in ground, surface, waste and drinking water in northern Mongolia. Environ Earth Sci 73(2):649–662. https://doi.org/10.1007/s12665013-3029-0 Priess JA, Schweitzer C, Batkhishig O, Koschitzki T, Wurbs D (2015) Impacts of agricultural land-use dynamics on erosion risks and options for land and water management in northern Mongolia. Environ Earth Sci 73(2):697–708. https://doi.org/10.1007/s12665-0143380-9 Sinha S, Balganjav K (2011) From vulnerability to sustainability: environment and human development. UNDP, Mong Hum Dev Rep 2011:148P Sinyukovich VN, Sorokovikova LM, Tomberg IV, Tulokhonov AK (2010) Climate changes and the Selenga River chemical flow. Dokl Earth Sci 433(2):1127–1131. https://doi.org/10.1134/ S1028334X10080295 Sorokovikova LM, Popovskaya GI, Tomberg IV, Sinyukovich VN, Kravchenko OS, Marinaite II, Bashenkhaeva NV, Khodzher TV (2013) The Selenga River water quality on the border with

U

586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 706

646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705