JES-00117; No of Pages 10 J O U RN A L OF E N V I RO N ME N TA L S CI EN CE S X X (2 0 1 4 ) XX X–XXX

Available online at www.sciencedirect.com

ScienceDirect www.journals.elsevier.com/journal-of-environmental-sciences

4Q1

Qiong Wu1 , Xinghui Xia1,⁎, Xinli Mou1,2 , Baotong Zhu1 , Pujun Zhao1 , Haiyang Dong1

5 6 7

O

R O

2

F

3

Effects of seasonal climatic variability on several toxic contaminants in urban lakes: Implications for the impacts of climate change

1

1. State Key Laboratory of Water Environment Simulation, Key Laboratory of Water and Sediment Sciences of Ministry of Education, School of Environment, Beijing Normal University, Beijing 100875, China. E-mail: [email protected] 2. School of Chemical and Environmental Engineering, Chongqing Three Gorges University, Wanzhou 404100, China

AR TIC LE I NFO

ABSTR ACT

D

11 10

Article history:

Climate change is supposed to have influences on water quality and ecosystem. However, only

Received 8 January 2014

few studies have assessed the effect of climate change on environmental toxic contaminants

14 19

Received in revised form 30 April 2014

in urban lakes. In this research, response of several toxic contaminants in twelve urban lakes

15 20

Accepted 30 April 2014

in Beijing, China, to the seasonal variations in climatic factors was studied. Fluorides, volatile

E

12 17 18 13

22 40 23

T

16 21

phenols, arsenic, selenium, and other water quality parameters were analyzed monthly from 2009 to 2012. Multivariate statistical methods including principle component analysis, cluster analysis, and multiple regression analysis were performed to study the relationship between

41 24 42 25

Climatic variations

contaminants and climatic factors including temperature, precipitation, wind speed, and

Urban lakes

sunshine duration. Fluoride and arsenic concentrations in most urban lakes exhibited a

43 26 44 27

Volatile phenols

significant positive correlation with temperature/precipitation, which is mainly caused by

Arsenic

rainfall induced diffuse pollution. A negative correlation was observed between volatile phenols

45 28 46 29

Fluorides

47 30 48 31

Water quality

35 36 37 38 39 51 50 49 52

E

R R

and biodegradation rates caused by higher temperature. Selenium did not show a significant response to climatic factor variations, which was attributed to low selenium contents in the lakes and soils. Moreover, the response degrees of contaminants to climatic variations differ

N C O

34

and temperature/precipitation, and this could be explained by their enhanced volatilization

Selenium

U

33

C

Keywords:

32

Q2

P

8

among lakes with different contamination levels. On average, temperature/precipitation contributed to 8%, 15%, and 12% of the variations in volatile phenols, arsenic, and fluorides, respectively. Beijing is undergoing increased temperature and heavy rainfall frequency during the past five decades. This study suggests that water quality related to fluoride and arsenic concentrations of most urban lakes in Beijing is becoming worse under this climate change trend. © 2014 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Published by Elsevier B.V.

⁎ Corresponding author. E-mail: [email protected] (Qiong Wu), [email protected] (Xinghui Xia), [email protected] (Xinli Mou), [email protected] (Baotong Zhu), [email protected] (Pujun Zhao), [email protected] (Haiyang Dong).

http://dx.doi.org/10.1016/j.jes.2014.04.001 1001-0742/© 2014 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Published by Elsevier B.V.

Please cite this article as: Wu, Q., et al., Effects of seasonal climatic variability on several toxic contaminants in urban lakes: Implications for the impacts of climate change, J. Environ. Sci. (2014), http://dx.doi.org/10.1016/j.jes.2014.04.001

Q3

2

J O U RN A L OF E N V I RO N ME N TA L S CIE N CE S X X (2 0 1 4 ) XXX –XXX

In this research, twelve lakes in urban areas of Beijing were 112

54 53

Introduction

55

The impacts of climate change on aquatic environment have

studied so far, to the variation of climatic factors. Apart from the 115

56

attracted extensive attention among governments and scientists

toxic contaminants, general physical parameters such as pH, 116

57

worldwide over the recent years. Climate change is associated with

dissolved oxygen (DO), conductivity, secchi depth, and water 117

58

changes in long-term weather conditions including precipitation,

temperature were monthly measured from 2009 to 2012. The 118

59

temperature, wind speed, and sunshine duration as well as

concentration and distribution of contaminants were assessed. 119

60 Q4

short-term extreme weather events (Larsen et al., 2011; Mitsch and

The effects of seasonal climatic factors (air temperature, precip- 120

61 Q5

Hernandez, 2013; Xia et al., 2012). Impacts of climate change on water

itation, wind speed, and sunshine duration) on toxic contami- 121

62

quantity have been widely studied (Park et al., 2011; Ferrer et al., 2012;

nants were studied, and the potential key climatic drivers for toxic 122

63

Mu et al., 2013). These effects coupled with climate change would

pollutant concentration variations and the effect mechanism 123

64

further influence the water quality and aquatic ecosystem directly or

were discussed. Accordingly, the effect of climate change on lake 124

65

indirectly.

water quality was analyzed, and the changing trend of lake water 125

selected to study the response of toxic contaminants, including 113

O

F

volatile phenols, fluorides, arsenic, and selenium which are rarely 114

Compelling evidences have been reported that increasing

quality in the urban areas of Beijing under the context of climate 126

67

temperature may alter the biotransformation of contaminants to

change was explored and future lake management measures 127

68

more bioactive metabolites and induce contaminants such as

were proposed.

69

persistent organic pollutants (POPs) releasing from sediments or

70

ice (Noyes et al., 2009; Whitehead et al., 2009; Petrovic et al., 2011).

71

Ma et al. (2011) analyzed records of the concentrations of POPs at the

72

Zeppelin and Alert Stations in Arctic air since the early 1990s and

73

compared it with the model simulation results of the effect of

74

climate change on their atmospheric abundances; they suggested

75

that a wide range of POPs stored in the ice or water had been

76

remobilized into the Arctic atmosphere over the past two decades

77

due to climate warming. Dissolved ions and heavy metal in Arctic

78

lakes also showed an increasing trend due to enhanced temperature

79 Q6

induced snow melt (Macdonald et al., 2005; Zhulidov et al., 2011; Liu et

80

al., 2012). For instance, Thies et al. (2007) observed a substantial

81

increase in solute concentration at two alpine lakes (Rasass See and

82

Schwarzsee ob Solden) in European Alps, in which electrical

83

conductivity increased by 18-fold and 3-fold during the past two

84

decades, respectively. Variations in precipitation linked with

85

climate change would affect water quality through influencing wet

86

deposition and diffused pollution of chemical contaminants. It has

87

been reported that a 20% increase/decrease in precipitation could

88

result in a 53% and 4% decrease/increase in perturbed air concen-

89

tration of γ-hexachlorocyclohexanes (HCHs) and α-HCH, respec-

90 Q7

tively (Ma and Cao, 2010). Jeppesen et al. (2009) predicted that the

91

phosphorus loading would increase by 3.3–16.5 times to Danish

92

streams due to the increasing precipitation under future climate

93

change scenario A2 during the period 2071–2100. In addition,

94

precipitation induced variations in river flow can influence the

95

concentration of toxic contaminants in the aquatic environment.

96 Q8

For example, Petrovic et al. (2011) analyzed 72 pharmaceutical

97

compounds in the Llobregat River at its mouth in Mediterranean,

98

and showed that the concentration of chemical pollutants exhibited a

99

variability of the same order as riverflows. Therefore, variations in

100

climatic factors coupling with each others may have a profounding

101

influence on transformation and migration of environmental con-

102

taminants in the aquatic system.

R O

66

P

1. Materials and methods 1.1. Site description

However, most studies about response of water quality to climate

104

change focus on remote arctic and alpine water bodies (Battarbee et

105

al., 2012; Todd et al., 2012). In urban area, the redistribution and

106

transformation of contaminants will also be affected by variations in

107

precipitation, temperature, and wind speed as well as sunshine

108

duration. Therefore, we hypothesize that environmental contami-

109

nants in the lakes of urban areas are likely be susceptible to variations

110

in climatic factors, and these effects may be different among lakes

111

with different contamination levels.

130 129 131 132

1.2. Sample collection and laboratory analysis

155

The water sampling was carried out monthly except the frozen period (mainly from January to March) from January 2009 to December 2012. The sampling sites were set in the center of the lakes. Secchi depth was measured using a Secchi disk; water temperature, pH, and conductivity of water were measured in situ with a multi-parameter meter (Mettler Toledo, SG23). DO concentration was measured with an oxygen meter (Mettler Toledo, SG9-FK2). Water samples for chemical analysis were collected under 0.5 m at the sampling sites and stored in glass bottles.

156

E

D

Beijing (39°28′N–41°05′N, 115°25′E–117°30′E), the capital of China, is located in the middle latitude, belonging to the eastern warm temperate monsoon zone with semi-humid continental climate and four distinct seasons. Annual precipitation ranges from 470 to 500 mm with uneven temporal and spatial rainfall distribution. In this research, twelve urban lakes with similar water areas and depth were studied. General characteristics and physical parameters as well as locations of the studied urban lakes are shown in Appendix Tables S1 and S2 and Fig. 1, respectively. The lakes are all located in the recreational parks with similar land use types and covering areas. Water source of Lake Tuancheng comes from Miyun Reservoir, an important drinking water area in Beijing (Fig. 1). Recharge sources of important landscape lakes such as Lakes Qianhai and Houhai mainly come from both Miyun and Guanting Reservoir. As for the general landscape lakes including Lakes Fuhai, Taoranting, Longtan, Qingnian, Lianhua, and Liuyin, recharge sources are mainly from Qinghe, Wujiacun, and No. 6 Water Reclamation Plant in spring season (March to May) and winter season (December), and most water quality parameters of the reclaimed water for landscape water reached Class IV of the national water quality standards, which is the minimum standard for industry and recreation.

T

C

E

R

R

O

C

N

U

103

128

Please cite this article as: Wu, Q., et al., Effects of seasonal climatic variability on several toxic contaminants in urban lakes: Implications for the impacts of climate change, J. Environ. Sci. (2014), http://dx.doi.org/10.1016/j.jes.2014.04.001

133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154

157 158 159 160 161 Q9 162 163 Q10 164 165

Kunyu River

Lakes

Lianhua Lake

O

South Moat

Longtan Lake

F

Taoranting Lake

N

North Moat

Liuyin Park Lake

Qianhai Lake Houhai Lake

Qingnian Lake

Xiaoyue River

R O

P

Rivers

Zizhuyuan Lake

Chang River

Yuyuantan Lake

Kunming Lake

D

E

Tuancheng Lake

Jingmi channel

Fig. 1 – Location of studied urban lakes in Beijing, China.

Beijing

Reservoirs Weather stations Rivers

T

C

E

R

R

N C O

U

J O U RN A L OF E N V I RO N ME N TA L S CI EN CE S X X (2 0 1 4 ) XX X–XXX

3

Please cite this article as: Wu, Q., et al., Effects of seasonal climatic variability on several toxic contaminants in urban lakes: Implications for the impacts of climate change, J. Environ. Sci. (2014), http://dx.doi.org/10.1016/j.jes.2014.04.001

4

J O U RN A L OF E N V I RO N ME N TA L S CIE N CE S X X (2 0 1 4 ) XXX –XXX

195

N

C

O

R

R

E

Before analysis, all data including water quality parameters 196 and climatic factors were standardized at a mean of 0 and 197 a standard deviation of 1, and tested for normality and 198 homogeneity of variance with the Kolmogorov–Smirnov (K–S) 199 test. The data were log-transformed if they were not normally 200 distributed. The spatial variability of water quality in different 201 lakes was determined from hierarchical agglomerative cluster 202 analysis (CA) by means of the Ward's method using squared 203 Euclidean distances as a measure of similarity to divide the 204 lakes into different types (Hussain et al., 2008). The Pearson 205 correlation coefficient was calculated and used to test the 206 significance of correlation between water quality parameters 207 and climatic factors; correlation was considered significant when 208 the significance level was smaller than 0.05. Principle component 209 analysis (PCA) technique was firstly conducted to extract the 210 independent variables of climatic factors for multiple regression 211 analysis; then it was used to analyze the effects of climatic factors 212 on water quality among lakes with different levels of contam213 inants. Kaiser–Meyer–Olkin (KMO) and Bartlett's sphericity tests 214 Q13 were performed (Shrestha and Kazama, 2007) to examine the 215 suitability of the data for PCA. Multiple regression analysis was 216 performed to indentify the climatic factors that made strong 217 contribution to the variation in water quality parameters. 218 Before analysis, Durbin–Watson (D–W) test was conducted to 219 examine the autocorrelation of the data set. For the data 220 without autocorrelation, ordinary least squares multiple regres221 sion analysis was performed; for the data with autocorrelation, 222 generalized least square models were applied to overcome serial 223 correlation.

U

2. Results and discussion

231 230

2.1. Variations in toxic contaminant concentrations

232

F

224

P

R O

O

It could be observed that a four-year mean value of volatile phenols ranged from 0.0007 to 0.0019 mg/L for the twelve urban lakes (Table 1). Lakes Qianhai, Houhai, and Longtan had the minimum value of 0.0007 mg/L, whereas Lake Yuyuantan had the maximum value of 0.0019 mg/L, followed by Lakes Kunming, Taoranting, and Lianhua with the same value of 0.0017 mg/L. The four-year mean value of volatile phenols indicated that volatile phenol concentrations in all the urban lakes belonged to Class III of national surface water standard for volatile phenols in China (