Temporal Variation of Groundwater Level and Arsenic Concentration at Jianghan Plain, Central China

GEXPLO-05493; No of Pages 14 Journal of Geochemical Exploration xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Journal of Geochemical Exploration

journal homepage: www.elsevier.com/locate/jgeoexp

Temporal variation of groundwater level and arsenic concentration at Jianghan Plain, central China

Yanhua Duan a,c, Yiqun Gan a,c, Yanxin Wang a,c,⁎, Yamin Deng b,c, Xinxin Guo a,c, Chuangju Dong a,c a School of Environmental Studies, China University of Geosciences, 430074 Wuhan, China b Geological Survey, China University of Geosciences, 430074 Wuhan, China c State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences, 430074 Wuhan, China article info abstract

Article history: Elevated arsenic in groundwater affects some 60 thousand people in the Jianghan Plain, yet mitigation efforts are Received 28 April 2014 hindered by persistent uncertainty about the proximal source of As and mechanisms for its mobilization. Samples Accepted 5 December 2014 were collected every month from May 2012 to April 2014 in a set of 39 monitoring wells and two rivers in the Available online xxxx field monitoring site, Jianghan Plain, to evaluate the temporal variation of groundwater level and chemical composition for As and other constituents. Results showed that groundwater tables fluctuated during the sampling Keywords: period, with higher water levels during the rainy season and lower water levels during the dry season. Wells Temporal variation Groundwater level at 25 m and 50 m deep had lower water level than those of the wells at 10 m deep. Water levels of the wells Arsenic located in the south of the field monitoring site were always 1 m higher than the wells in the north. Only five Jianghan Plain wells showed no significant changes in As concentrations during the monitoring period; 23 wells exhibited seasonal variations in groundwater As concentration that were consistent from year to year; the rest of the 11 wells not only showed seasonal variations but also exhibited increasing trends over the monitoring period. Seasonal As concentration changes in groundwater were positively correlated to groundwater level changes, with lower concentration corresponding to lower water level during dry season and vice versa during rainy season. The rise in As concentration during the rainy season could be attributed to enhanced reductive dissolution of iron oxyhydroxides and/or reductive desorption of As(V) as the conditions turn to be more reducing, while during the dry season more As could be scavenged onto fresh iron oxyhydroxides. Meanwhile, temporal variations showed compatible trends between As and Fe, Fe(II), S2−, indicating that As variation would be related to Fe and S cycling in the aquifer systems. Additionally, the availability of labile organic carbon as a driver of microbial reduction may play an important role in controlling the spatial and temporal variation of groundwater As concentration in the Jianghan Plain. © 2014 Elsevier B.V. All rights reserved.

1. Introduction under reducing conditions (Smedley and Kinniburgh, 2002), and recent studies have shown that groundwater As concentration is highly vari- The presence of naturally elevated levels of arsenic in groundwater able at both local and regional scale in reducing aquifers from river has been reported worldwide, including in Bangladesh, India, Cambodia, deltas and inland basins (van Geen et al., 2003; Fendorf et al., 2010; Vietnam, Argentina, the United States, Chile, and China, with the Guoetal.,2012). The causes of the heterogenous As distribution include most serious waterborne endemic arsenic poisonings occurring in geologic setting, organic matter sources, water–rock interactions, Bangladesh, India, and China (Nickson et al., 1998; Nordstrom, 2002; groundwater ﬂow, and anthropogenic inﬂuences (Harvey et al., 2002; Smedley and Kinniburgh, 2002; Romero et al., 2003; Zheng et al., McArthur et al., 2004; Stute et al., 2007; Kocar et al., 2008; Neumann 2005; Harvey, 2008; Polizzotto et al., 2008; Deng et al., 2009; Fendorf et al., 2010; Guo et al., 2011; Freikowski et al., 2013). Such spatial vari- et al., 2010; Saha et al., 2010; Xie et al., 2011; Li et al., 2012; Guo et al., ability naturally leads to the concern that groundwater As concentra- 2013a). In terms of redox conditions favorable for As enrichment in tion may temporarily change as well. the subsurface, high As groundwater has been more frequently found Temporal change of As concentrations has been reported in Bangladesh (Cheng et al., 2005; Dhar et al., 2008), West Bengal of India (Savarimuthu et al., 2006; Farooq et al., 2011), Vietnam (Berg ⁎ Corresponding author at: No. 388, Lumo Road, Wuhan 430074, China. Tel.: +86 27 et al., 2001), Hetao basin and Yichuan Plain of China (Guo et al., 2012, 67883998; fax: +86 27 87481030. 2013b; Han et al., 2012), Ganges Floodplain of Nepal (Brikowski et al., E-mail addresses: [email protected] (Y. Duan), [email protected] (Y. Gan), [email protected] (Y. Wang), [email protected] (Y. Deng), 2014), Duero River Basin of Spain (Mayorga et al., 2013), Nevada and [email protected] (X. Guo), [email protected] (C. Dong). Snohomish County of USA (Frost et al., 1993; Steinmaus et al., 2005;

Thundiyil et al., 2007), Ouro Preto of Brazil (Goncalves et al., 2007), and Mn oxides/hydroxides under reducing conditions, while microbial deg- Zimapán Valley of Mexico (Rodríguez et al., 2004). Temporal variations radation of organic matter may also facilitate the release of As into of As concentrations in different sites show different trends. Several groundwater (Gan et al., 2014). Elevated As in groundwater affects the studies have documented that As concentrations in some wells might health of some 60 thousand people in the Jianghan Plain (Wang and vary dramatically by season (Frost et al., 1993; Savarimuthu et al., Zhao, 2007; Li et al., 2010), and it is crucial to characterize temporal var- 2006; Goncalves et al., 2007; Farooq et al., 2011; Guo et al., 2012; iations in groundwater As concentration to understand the mechanisms Brikowski et al., 2014), while in other sites the temporal variation was of As cycling and to help to develop effective strategies for sustainable inconsistent from year to year (Dhar et al., 2008), although some exploitation of groundwater resource. studies suggest absence of any significant change in As concentration The objectives of this study are to (1) fill a major gap in our under- (McArthur et al., 2004; Cheng et al., 2005; Steinmaus et al., 2005; standing of the temporal variability in groundwater level and As con- Thundiyil et al., 2007). Anthropogenic or natural factors responsible centration in the Jianghan Plain; (2) provide new insights into the for the observed groundwater As concentration changes include di- mechanisms of As mobilization under reducing conditions in the study lution by recharge of water with low As concentrations, seasonal area; and (3) help us highlight the important implications of temporal changes in redox conditions, pumping rates, water table depths, or shifts variations in As concentration on the exposure and the precision of in direction of hydraulic gradient (Tareq et al., 2003; Rodríguez et al., the health risk assessments. 2004; Cheng et al., 2005; Goncalves et al., 2007; Benner et al., 2008; Dhar et al., 2008; Kocar et al., 2008; Polizzotto et al., 2008; Guo et al., 2. Regional hydrogeology 2012; Brikowski et al., 2014). Several studies have been conducted for revealing the high As Jianghan Plain is an alluvial plain formed by the Yangtze and Han groundwater in the Jianghan Plain since it was first noted in 2005. In rivers located in the Middle Reaches of the Yangtze River that includes May 2006, the Center for Endemic Disease Control of Xiantao and the central and southern regions of Hubei Province (Fig. 1). It has a Hubei Province investigated 19 towns in Xiantao, where 60% of the sub-tropical monsoonal climate and annual temperature ranges be- 2538 km2 fertile alluvial–lacustrine sediments deposited by rivers and tween 15 °C and 17 °C. The average annual precipitation in the region lakes is farmland cultivated with rice, rape, cotton, and vegetables, and is 1269 mm, increasing from 800 mm in the northwest to 1500 mm or 24% is pond for aquariculture. The results showed that 863 wells in 12 more in the southeast, 30–50% of which occurs in summer. The average towns (179 villages) had As levels exceeding China's National Drinking annual evaporation is 1200 mm, which is almost equal to the total pre- Water Standard of 10 μg/L. The hydrogeochemical characteristics of the cipitation. The non-frost season is 200–290 days. shallow groundwater in the Jianghan Plain have been investigated by us Jianghan Plain is a semi-closed Quaternary basin with a higher eleva- (Zhou et al., 2013; Gan et al., 2014). Groundwater with high As was tion in the north and a lower elevation in the south. The middle region found in wells at depths of 10 to 45 m along rivers. The main potential in the plain is a low alluvial plain with elevations of 20–27 m in the mechanism for the release of As is the reductive dissolution of Fe and southeast and 30–40 m in the northwest, while the outlying areas

Fig. 1. Location of the Shahu ﬁeld monitoring site and the monitoring wells.

Table 1 Summary of statistical data for major constituents of groundwater and surface water in the ﬁeld site, Jianghan Plain.

− − 2− Points Depth pH Eh (mV) EC K Na Ca Mg Cl HCO3 (mg/L) SO4 (mg/L) (m) (μS/cm) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L)

Avg Stdev Avg Stdev Avg Stdev Avg Stdev Avg Stdev Avg Stdev Avg Stdev Avg Stdev Avg Stdev Avg Stdev

SY01 10 7.0 ± 0.22 −123 ± 53 1356 ± 145 1 ± 0 28 ± 2 218 ± 17 52 ± 4 15 ± 13 1084 ± 102 14 ± 18 25 7.4 ± 0.17 −126 ± 52 798 ± 65 3 ± 0 18 ± 1 124 ± 9 28 ± 3 9 ± 5 628 ± 48 7 ± 4 50 7.2 ± 0.18 −108 ± 73 704 ± 54 1 ± 0 16 ± 6 100 ± 9 24 ± 3 4 ± 2 535 ± 56 5 ± 8 SY02 10 7.0 ± 0.13 −98 ± 48 1003 ± 133 2 ± 1 12 ± 4 157 ± 23 43 ± 13 15 ± 13 754 ± 155 20 ± 18 25 7.2 ± 0.13 −137 ± 47 782 ± 51 2 ± 1 24 ± 8 108 ± 29 32 ± 7 4 ± 2 616 ± 51 3 ± 2 50 7.3 ± 0.14 −151 ± 49 815 ± 60 2 ± 1 37 ± 15 104 ± 9 27 ± 8 6 ± 1 632 ± 49 3 ± 2 SY03 10 6.9 ± 0.18 −112 ± 35 1106 ± 247 2 ± 1 15 ± 3 174 ± 29 36 ± 8 23 ± 8 866 ± 195 5 ± 9 25 7.3 ± 0.15 −126 ± 63 866 ± 124 2 ± 0 20 ± 4 130 ± 24 27 ± 3 9 ± 14 655 ± 82 10 ± 16 50 7.3 ± 0.10 −140 ± 34 786 ± 80 2 ± 1 19 ± 3 112 ± 15 31 ± 8 5 ± 2 621 ± 47 3 ± 2 SY04 10 7.0 ± 0.19 −114 ± 41 1174 ± 115 1 ± 0 20 ± 4 194 ± 23 43 ± 5 7 ± 10 964 ± 137 8 ± 12 25 7.1 ± 0.15 −102 ± 58 992 ± 68 3 ± 0 26 ± 3 139 ± 16 31 ± 3 5 ± 3 759 ± 85 3 ± 2 50 7.1 ± 0.12 −113 ± 35 747 ± 48 2 ± 0 17 ± 2 108 ± 15 25 ± 2 4 ± 5 597 ± 70 3 ± 2 SY05 10 6.9 ± 0.15 −120 ± 42 919 ± 124 2 ± 1 14 ± 4 129 ± 25 27 ± 6 15 ± 10 708 ± 91 4 ± 3 25 7.2 ± 0.14 −128 ± 40 783 ± 73 2 ± 0 17 ± 1 114 ± 18 26 ± 2 4 ± 3 627 ± 59 3 ± 2 50 7.2 ± 0.17 −109 ± 45 792 ± 92 2 ± 1 17 ± 4 110 ± 9 28 ± 7 6 ± 14 625 ± 68 6 ± 15 SY06 10 6.7 ± 0.16 −101 ± 35 1066 ± 161 2 ± 1 16 ± 4 175 ± 18 27 ± 2 19 ± 6 784 ± 139 25 ± 28 25 7.1 ± 0.14 −106 ± 50 919 ± 77 2 ± 1 18 ± 2 132 ± 12 30 ± 2 5 ± 2 711 ± 70 3 ± 2 50 7.2 ± 0.11 −96 ± 44 794 ± 57 2 ± 0 17 ± 5 112 ± 12 27 ± 3 4 ± 2 615 ± 55 3 ± 2 SY07 10 7.0 ± 0.16 −105 ± 47 1022 ± 147 3 ± 1 22 ± 11 143 ± 26 29 ± 7 15 ± 6 802 ± 89 4 ± 4 25 7.2 ± 0.13 −134 ± 36 924 ± 112 2 ± 0 36 ± 22 135 ± 21 31 ± 5 6 ± 3 741 ± 88 3 ± 2 50 7.0 ± 0.21 −139 ± 47 902 ± 82 2 ± 1 34 ± 11 133 ± 18 27 ± 3 22 ± 9 679 ± 74 8 ± 7 SY08 10 7.1 ± 0.25 −118 ± 42 1043 ± 144 1 ± 0 24 ± 1 170 ± 17 39 ± 6 16 ± 14 833 ± 119 5 ± 4 25 7.2 ± 0.15 −129 ± 35 777 ± 95 2 ± 1 18 ± 3 107 ± 10 24 ± 3 5 ± 3 581 ± 52 4 ± 4 50 7.2 ± 0.17 −130 ± 46 822 ± 131 2 ± 1 19 ± 4 112 ± 9 31 ± 14 5 ± 1 638 ± 108 4 ± 2 SY09 10 7.0 ± 0.17 −120 ± 36 1060 ± 188 2 ± 0 27 ± 6 156 ± 13 47 ± 9 10 ± 20 877 ± 159 9 ± 22 25 7.2 ± 0.12 −118 ± 49 803 ± 203 2 ± 1 18 ± 5 101 ± 6 24 ± 1 5 ± 2 532 ± 37 9 ± 10 50 7.2 ± 0.18 −119 ± 68 795 ± 71 2 ± 0 21 ± 5 113 ± 10 27 ± 2 4 ± 3 620 ± 44 3 ± 2 SY10 10 7.0 ± 0.23 −74 ± 48 1071 ± 117 3 ± 0 21 ± 4 141 ± 13 46 ± 5 11 ± 3 842 ± 98 4 ± 3 25 7.2 ± 0.18 −142 ± 34 846 ± 110 1 ± 0 24 ± 6 126 ± 11 31 ± 4 4 ± 2 681 ± 94 3 ± 2 50 7.2 ± 0.14 −137 ± 45 813 ± 39 2 ± 1 27 ± 9 119 ± 16 28 ± 6 5 ± 1 630 ± 46 3 ± 2 SY11 10 7.0 ± 0.16 −85 ± 53 1017 ± 156 2 ± 2 21 ± 5 143 ± 19 42 ± 12 13 ± 3 803 ± 77 4 ± 2 25 7.3 ± 0.13 −141 ± 45 866 ± 119 1 ± 0 26 ± 7 130 ± 11 29 ± 3 9 ± 5 683 ± 86 4 ± 2 50 7.2 ± 0.16 −142 ± 49 875 ± 179 3 ± 1 40 ± 15 104 ± 14 36 ± 23 8 ± 2 690 ± 162 4 ± 2 SY12 10 6.9 ± 0.17 −80 ± 60 1042 ± 203 2 ± 2 22 ± 3 148 ± 12 56 ± 19 9 ± 5 858 ± 176 9 ± 4 25 7.1 ± 0.14 −118 ± 56 877 ± 118 2 ± 0 29 ± 4 125 ± 13 28 ± 12 5 ± 2 726 ± 114 3 ± 2 50 7.3 ± 0.17 −142 ± 56 945 ± 142 4 ± 2 77 ± 40 91 ± 12 22 ± 6 19 ± 15 669 ± 67 12 ± 8 SY13 10 7.0 ± 0.18 −99 ± 46 1209 ± 188 1 ± 0 26 ± 5 158 ± 11 70 ± 23 6 ± 4 1004 ± 231 5 ± 3 25 7.1 ± 0.14 −111 ± 45 954 ± 181 2 ± 1 35 ± 13 143 ± 11 29 ± 6 6 ± 1 734 ± 215 7 ± 9 50 7.3 ± 0.15 −147 ± 45 869 ± 133 3 ± 1 82 ± 44 78 ± 13 21 ± 3 16 ± 11 625 ± 61 8 ± 5 Surface DJR 7.8 ± 0.66 91 ± 65 443 ± 101 4 ± 1 11 ± 3 57 ± 21 13 ± 4 26 ± 15 214 ± 73 31 ± 8 water TSR 7.1 ± 0.99 82 ± 79 1057 ± 998 7 ± 4 36 ± 33 260 ± 466 31 ± 12 334 ± 500 267 ± 54 27 ± 13 Pond 7.9 ± 0.37 57 ± 52 527 ± 105 10 ± 3 15 ± 4 61 ± 19 14 ± 3 77 ± 90 231 ± 21 29 ± 25

− by Emilie (2009) showed that arsenic concentration ranged from 4 to spectrophotometer (HACH 2800, USA). Concentrations of HCO3 were 15.5 mg/kg, with high rates of arsenic found at depths of 3 m and measured within 24 h using acid–base titration methods. The total con- 21 m. The results of the sediment core samples collected in the central centration of dissolved ions (Ca, Mg, Na, K, Fe, and Mn) was determined region of Jianghan Plain (112°59′ E, 30°02′ N) indicated that the main using an inductively coupled plasma atomic emission spectrometer heavy minerals of the upper sediments (0–100 m) were enriched with (ICP-AES) (IRIS Intrepid II XSP, USA). The accuracy of most major ele- Fe and Mn, such as epidote, hornblende, pyroxene, garnet, hematite, ments analyzed by ICP–AES was found to be generally within 4%. Anions − − 2− and limonite, ilmenite, magnetite and iron-stained rock debris (Kang such as Cl ,NO3 , and SO4 were determined using an ion chromato- et al., 2009). A large number of organic materials in the sediments graph (Dionex 2500, USA). The analytical precision reported by the have also been reported (Zeng et al., 2004). laboratories was better than 5%. DOC was determined using total organic carbon analyzers (multi N/C 3100, Jena and Vario EL cube, 3. Materials and methods Elementar, Germany). Total dissolved As was determined using a hydride generation-atomic fluorescence spectrometer (HG-AFS, 930, 3.1. Well construction Titan, China). The detection limit of the instrument for As was 0.05 μg/L and the relative standard deviation was b1.0%. All samples were diluted A total of 39 monitoring wells were installed at 13 points (Fig. 1), several times to adjust for the operating range and then analyzed. each with a nest of borehole at three different depths: 10 m, 25 m, Because of the unavailability of the analytical instrument, only the and 50 m. All the 10 m deep wells and two 25 m deep wells were drilled samples collected in July 2013 were collected for arsenic species analysis. by hollow-stem augers, while all the 50 m deep wells and the rest of Samples were acidified with concentrated HCl and preserved 25 m deep wells were drilled by direct mud rotary. Hollow-stem by 0.25 mol/L EDTA solution; the HDPE bottles were also wrapped with auger drilling is accomplished using a series of interconnected auger tinfoil to protect the samples from light. Arsenic species of the flights with a cutting head at the lowermost end. As the augers are rotat- groundwater samples were separated by High Performance Liquid ed and pressed downward, the cuttings are rotated up the continuous flights. In mud rotary drilling, the drilling fluid is pumped down the drill rods and through a bit that is attached at the lower end of the drill rods. The fluid circulates back to the surface by moving up the an- Table 2 nular space between the drill rods and the wall of the borehole. At the Summary of statistical data for redox-sensitive constituents of groundwater and surface water in the field site, Jianhan Plain. surface, the fluid discharges through a pipe and enters into a segregated − sedimentation pit. The settling pit overflows into a suction pit where a Points Depth As (μg/L) Fe (mg/L) Fe2+ Mn S2 (μg/L) pump recirculates the fluid back through the drill rods. The inside diam- (m) (mg/L) (mg/L) eter of PVC pipes for the 10 m and 25 m deep wells is 50 mm, and that Avg Stdev Avg Stdev Avg Stdev Avg Stdev Avg Stdev for the 50 m deep wells is 75 mm. The well screen is 1 m long at the end SY01 10 35 ± 18 13.1 ± 5.2 9.1 ± 4.1 4.8 ± 1.0 9 ± 8 of the wells, and the filter pack surrounding the well screen is pure fine 25 75 ± 24 1.8 ± 1.1 1.7 ± 1.5 0.8 ± 0.3 18 ± 13 sand. A 1.5 m thick filter pack seal filled with bentonite pellets was 50 21 ± 5 3.0 ± 1.0 2.8 ± 1.0 0.4 ± 0.1 34 ± 28 placed in the annulus above the filter pack. The rest part of the annular SY02 10 5 ± 3 2.9 ± 1.9 2.6 ± 1.7 1.0 ± 0.9 47 ± 53 fi 25 81 ± 21 4.2 ± 1.0 3.4 ± 1.4 0.4 ± 0.7 44 ± 37 space is back lled with local deposits. The wells were then developed 50 38 ± 12 2.5 ± 1.2 2.3 ± 1.1 0.5 ± 1.0 66 ± 50 by air-lift pumping techniques for several hours to remove any water SY03 10 14 ± 12 14.6 ± 7.1 9.5 ± 5.1 0.9 ± 0.3 14 ± 11 or drilling fluids introduced into the well during drilling, stabilize the fil- 25 730 ± 325 7.7 ± 8.5 7.5 ± 7.0 4.9 ± 1.9 23 ± 21 ter pack and formation materials opposite the well screen, and to max- 50 50 ± 9 3.7 ± 2.4 3.6 ± 2.5 0.5 ± 0.1 8 ± 7 SY04 10 7 ± 3 7.3 ± 3.2 6.3 ± 3.3 0.6 ± 0.2 34 ± 39 imize the well efficiency and water inflow into the well. 25 50 ± 32 7.0 ± 4.6 6.1 ± 3.9 0.8 ± 0.4 12 ± 14 50 43 ± 18 3.6 ± 0.9 3.4 ± 1.1 0.3 ± 0.1 14 ± 13 3.2. Sampling and analytical methods SY05 10 70 ± 39 15.9 ± 8.3 9.5 ± 4.9 0.5 ± 0.2 21 ± 13 25 84 ± 32 4.9 ± 2.0 4.2 ± 2.0 0.2 ± 0.1 15 ± 25 Sampling of groundwater was carried out at the 39 monitoring wells 50 55 ± 14 3.0 ± 1.2 2.9 ± 1.6 0.5 ± 0.7 33 ± 30 SY06 10 11 ± 10 29.3 ± 12.6 18.6 ± 10.4 2.3 ± 1.1 34 ± 39 and two river points (R02 and R05) to characterize temporal changes in 25 81 ± 42 8.2 ± 4.9 5.7 ± 4.0 1.0 ± 1.2 14 ± 14 physical hydrology, major elements, and redox sensitive parameters 50 35 ± 16 3.2 ± 2.0 3.1 ± 2.2 0.6 ± 0.8 9 ± 8 such as iron and arsenic. Monthly samples were collected from May SY07 10 77 ± 57 12.0 ± 5.5 8.5 ± 4.1 2.5 ± 1.0 18 ± 38 2012 to December 2013. Groundwater samples of the well with 25 m 25 216 ± 182 5.0 ± 2.6 4.5 ± 2.6 1.6 ± 0.5 18 ± 19 50 39 ± 16 5.2 ± 2.2 4.9 ± 2.3 2.4 ± 1.3 209 ± 104 deep at SY09 were not collected after May 2013 because the monitoring SY08 10 5 ± 4 9.0 ± 4.7 7.3 ± 3.8 2.4 ± 2.6 37 ± 44 fi well was destroyed by rice transplanter. Waters from a sh pond near 25 52 ± 26 5.9 ± 2.3 5.5 ± 2.3 1.1 ± 1.0 36 ± 53 SY03 were collected monthly from December 2012. The wells were typ- 50 46 ± 14 3.2 ± 1.2 2.9 ± 1.3 0.5 ± 0.6 61 ± 56 ically purged by pumping for 5–10 min until the temperature, electrical SY09 10 17 ± 6 8.7 ± 2.7 7.4 ± 3.0 0.9 ± 1.6 16 ± 14 conductivity (EC), Eh and pH had stabilized. Water samples were then 25 35 ± 22 3.9 ± 2.7 2.8 ± 2.7 0.9 ± 0.6 14 ± 16 fi 50 37 ± 18 5.0 ± 3.2 4.5 ± 3.1 1.9 ± 1.9 41 ± 38 collected in 50 mL HDPE and brown glass bottles after being ltered SY10 10 4 ± 2 3.9 ± 2.1 3.0 ± 2.2 2.2 ± 1.6 24 ± 19 on site (using 0.45 μm membrane filters) and divided into four bottles. 25 8 ± 3 6.1 ± 2.0 5.3 ± 2.3 0.5 ± 0.1 8 ± 9 The first bottle (HDPE) was acidified with concentrated HCl and 50 26 ± 6 2.3 ± 1.2 1.9 ± 1.0 0.9 ± 0.8 111 ± 86 wrapped with tinfoil for analysis of total dissolved As, the second bottle SY11 10 16 ± 10 4.3 ± 2.7 3.5 ± 2.7 1.5 ± 0.8 6 ± 5 25 36 ± 16 5.8 ± 2.5 5.0 ± 2.5 1.5 ± 0.9 24 ± 22 (HDPE) was acidified to pH ≤ 2 using ultra-pure HNO for analysis of 3 50 30 ± 7 2.8 ± 1.6 2.5 ± 1.6 1.1 ± 0.5 135 ± 122 fi dissolved ions, the third bottle (brown glass) was acidi ed with H3PO4 SY12 10 15 ± 7 2.1 ± 1.8 1.3 ± 1.0 1.1 ± 0.7 22 ± 48 for analysis of dissolved organic carbon (DOC), and the last bottle 25 92 ± 36 7.0 ± 3.3 5.8 ± 3.4 1.2 ± 0.7 27 ± 25 (HDPE) was not acidified for the analysis of anions. Another unfiltered 50 35 ± 9 2.7 ± 2.8 2.5 ± 2.7 1.0 ± 0.8 144 ± 112 − SY13 10 21 ± 11 5.5 ± 2.6 5.0 ± 2.7 1.3 ± 0.7 8 ± 7 sample (500 mL) was collected for HCO3 analysis. All samples were 25 33 ± 16 5.3 ± 2.8 4.0 ± 2.8 1.2 ± 0.9 22 ± 28 stored in a refrigerator (temperature is about 4 °C) immediately until 50 28 ± 6 2.2 ± 0.9 2.1 ± 0.8 0.2 ± 0.2 198 ± 102 analyses were conducted. Surface DJR 4 ± 2 ––0.4 ± 0.5 – Temperature, pH, EC, and Eh were measured on site using a portable Water TSR 6 ± 3 0.5 ± 1.0 – 0.5 ± 0.5 – pH, EC, and Eh meter (HQ40D Field Case, cat. No: 58258-00, HACH, Col- Pond 5 ± 5 ––0.2 ± 0.2 – orado, USA). Fe(II) and sulfide were measured on site using a portable –: not determined.

Fig. 2. Temporal variation of precipitation in the ﬁeld monitoring site.

Chromatography (HPLC) (PerkinElmer Series 200) with an anion- 10 μm), and then quantiﬁed by ICP-MS (PerkinElmer, NexIon 300). exchange column (Hamilton PRP-X100, 250 × 4.1 mm, 10 μm) Themobilephasewas15mmol/L(NH4)2HPO4 solution (pH 6.0); protected by a Hamilton guard column (PRP-X100, 20 × 4.1 mm, the ﬂow rate was 1.0 mL min−1.

Fig. 3. Temporal variations of river water level and groundwater level in the ﬁeld monitoring site.

Water levels of the wells and the four rivers were also monitored the dry season. Fluctuation amplitude of water levels at different points every half a month over the same time period using an electric water- was different. Wells (SY11, SY12, and SY13) near the Lüfeng River al- level beeper. ways had higher water level and greater fluctuation amplitude. The fluctuations in water levels of the 25 m and 50 m deep wells 4. Results followed similar patterns, except for two wells at SY06-25 m and SY07-25 m, the temporal variation patterns of which were more consis- A summary of statistical data of the hydrochemical parameters, tent with those of the corresponding 10 m deep wells. About 2 m drop major constituents and redox-sensitive constituents of groundwater of the groundwater level was observed at SY06-25 m in the last four and surface water samples is given in Tables 1 and 2. months of the monitoring period. Variations in groundwater levels in 25 m and 50 m deep wells commonly had about 2 months delay relative 4.1. Water levels to the precipitation cycle, and varied from 21.13–22.27 m asl and 21.44– 22.26 m asl during the rainy season to 20.33–21.05 m asl and 20.07– In the field monitoring site, groundwater and surface water tables 21.02 m asl during the dry season, respectively. And their water levels fluctuated during the sampling period. As expected, there was little were lower than those of the 10 m deep wells at each point. The rainfall in our study area between November and March; it precipitated water level of the wells SY01 and SY13 located in the south of the then in April–May to peak in June–July (Fig. 2). Patterns of temporal var- field monitoring site was always 1 m higher than the wells SY05, iation in levels of surface water (except Lüfeng River) and groundwater SY06, and SY10 in the north. with 10 m deep were consistent with those of precipitation (Figs. 2, 3). Additionally, nine of the 13 monitoring points were constructed Unconnected with other rivers, Lüfeng River showed no temporal vari- in cultivated fields where rice, cotton, and wheat have been the ation in water level, and the average level being 22.84 m asl which was main crops. Temporal variations in groundwater levels at these higher than that of the other rivers' water level during the whole mon- points may also be related to irrigation between later May and Mid- itoring period. Water levels of the wells at 10 m varied seasonally from dle September when water was diverted from the Tongshun River 21.50–23.71 m asl during the rainy season to 20.27–22.01 m asl during for irrigation.

Fig. 4. Temporal variations in EC, Eh, and pH values of surface water and groundwater in the ﬁeld monitoring site. In each panel, data for groundwater from 10 m, 25 m, and 50 m deep wells are plotted using the average value of the 13 points in every month.

Fig. 5. Temporal variations in major dissolved ions of the surface water and groundwater in the ﬁeld monitoring site. In each panel, data for groundwater from 10 m, 25 m, and 50 m deep wells are plotted using the average value of the 13 points in every month.

4.2. Major constituents groundwater over time was considerable and closely correlated with water levels: lower Eh values were related to higher water levels in In general, groundwater in the ﬁeld monitoring site was nearly neu- July–September, and vice versa in March–May (Figs. 3, 4). The EC tral, with pH values ranging from 6.4 to 7.8 over the monitoring period value (mean 1084 μS/cm) of groundwater at 10 m was much higher of 20 months. The pH value of groundwater from 25 m and 50 m deep than that at 25 m and 50 m (mean 862 and 820μS/cm, respectively). wells was little higher than that from 10 m deep wells (Fig. 4). The tem- However, the temporal variation patterns of groundwater EC values poral variability of the pH observed in the 39 monitoring wells was were similar among three different depths of wells. Compared with quite small, with a RSD b 4% over the monitoring period. Negative Eh groundwater, the temporal variations in pH, Eh, and EC values of surface values of the groundwater indicated strongly reducing conditions of water were more obvious and erratic. the aquifers. The average Eh value of groundwater at depths of 25 The concentrations of Ca and Mg in groundwater were higher at (−125 mV) and 50 m (−128 mV) was slightly lower than that of 10 m (mean 162 mg/L and 43 mg/L) than those at 50 m (mean groundwater at 10 m (−103 mV) (Fig. 4). Seasonal variation of Eh in 107 mg/L and 27 mg/L), while the Na concentrations showed a contrary

Fig. 6. Temporal variations in major dissolved anions of the surface water and groundwater in the ﬁeld monitoring site. In each panel, data for groundwater from 10 m, 25 m, and 50 m deep wells are plotted using the average value of the 13 points in every month.

Please cite this article as: Duan, Y., et al., Temporal variation of groundwater level and arsenic concentration at Jianghan Plain, central China, J. Geochem. Explor. (2014), http://dx.doi.org/10.1016/j.gexplo.2014.12.001 Y. Duan et al. / Journal of Geochemical Exploration xxx (2014) xxx–xxx 9 tendency. For most of groundwaters, concentrations of Ca, Mg, and Na at 25 m, but showed the lowest As at 50 m. Arsenic concentration in kept relatively stable, with %RSDs at below 15%, 15%, and 20%, respec- wells at SY05 decreased with depth. The vertical distribution of As at tively. However, decreasing trends of Na concentration with time SY10 was completely contrary to that at SY05: As concentration rose were observed in groundwater at 50 m and those of Mg concentration with the increasing depth, although wells at this point always had a at 10 m were observed over the monitoring period (Fig. 5). very low As content. Bicarbonate was the major anionic species present in groundwater There were three different patterns of temporal variation of ground- − irrespective of seasons. The HCO3 concentrations in groundwater water As concentration. Only five wells, SY02-10 m, SY04-10 m, SY08- were much higher at 10 m (860 mg/L) than at 25 m (673 mg/L) and 10 m, SY10-10 m and SY10-25 m, showed no significant changes (less − 50 m (629 mg/L). The temporal variability of HCO3 concentrations in than 10 μg/L) in As concentrations during the monitoring period. 23 the wells at different depths followed similar pattern (Fig. 6). In com- wells exhibited seasonal variations in groundwater As concentration − 2− parison, concentrations of Cl and SO4 in groundwater were much that were consistent from year to year (Fig. 8). The largest amplitude lower. Groundwater at 10 m had slightly higher Cl− (mean 15 mg/L) of the fluctuations was observed at SY03-25 m, ranging between 101 2− and SO4 (mean 9 mg/L) concentrations than at 50 m (mean 8 mg/L and 1072 μg/L. The rest of the 11 wells not only showed seasonal varia- − 2− and 5 mg/L for Cl and SO4 ) and 25 m (mean 6 mg/L and 4 mg/L for tions but also exhibited increasing trends over the monitoring period − 2− − Cl and SO4 ). Temporary spikes were observed both in Cl and (Fig. 9), which was most significant in well SY07-25 m. And seasonal 2− SO4 concentration in February 2013 when the rainfall was lowest concentration changes in groundwater from these 34 wells were posi- over a year. tively correlated to groundwater level changes, with lower concentra- By contrast, temporal variations of major dissolved ions and anions tion corresponding to lower water level during dry season and higher of the surface water were more significant and erratic, especially for concentration corresponding to higher water level during rainy season the samples collected from Tongshun River. Concentrations of major (Figs.3,8,9). dissolved ions (Ca. Mg, Na, and K) and chloride in Tongshun River water increased sharply in some months, probably due to discharge of pollutants into the river (Figs. 5, 6). According to our investigation, we 4.4. Redox-sensitive constituents found out a processing plant and pig farms near the river; it is very com- mon for the plant and pig farm to use calcium hypochlorite and/or lime Concentrations of redox-sensitive constituents Fe, Fe(II), Mn, and for the wastewater treatment and pigpen sterilization, so concentra- S2− in the field monitoring site varied both spatially and temporally. tions of major dissolved ions and chlorine in this river water increased Dissolved Fe concentrations spanned one order of magnitude, ranging sharply in some months probably due to the discharge of treated waste- from 1.8 ± 1.1 mg/L at SY01-25 m to 29.3 ± 12.6 mg/L at SY06-10 m water from the plant and pig farm. (Table 2). Both Fe and Fe(II) concentrations decreased with depth (Fig. 10), with Fe(II) as the dominant Fe species in groundwater samples 4.3. Arsenic which accounted for 72%, 84%, and 91% of dissolved iron at 10 m, 25 m and 50 m respectively. In parallel with changes in concentration of As, The arsenic concentrations in groundwater from the 39 wells ranged concentration of Fe and Fe(II) rose during the rainy season and dropped from 4 ± 2 μg/L at SY10-10 m to 730 ± 325 μg/L at SY03-25 m over the during the dry season (Figs. 8, 9, 10). And like As concentration, an in- monitoring period (Table 2). Results of As speciation measurement of creasing trend was also observed in Fe and Fe(II) concentration of groundwater samples collected in July 2013 showed that the predomi- groundwater samples from most of the monitoring wells at 25 m and nant species was As(III) except five wells (SY01-25 m, SY02-50 m, 50 m. SY04-50 m, SY12-25 m, SY13-25 m) which had a mixture of both As compared with Fe, Mn concentration was much lower in the field As(III) and As(V) (Fig. 7). Arsenate in these five wells accounted for monitoring site, ranging from 0.2 ± 0.1 mg/L at SY05-25 m to 4.9 ± 24–49% of the total As (average 33%). The characteristics of As specia- 1.9 mg/L at SY03-25 m (Table 2). Vertically, Mn concentration de- tion clearly reflect the strongly reducing conditions in the aquifers. Par- creased with depth (Fig. 10), with insignificant seasonal variation but ticulate As was very low or undetected in groundwater except two wells an increasing tendency in 50 m deep wells. Sulfide concentration at SY06-50 m and SY13-25 m where particulate As accounted for 28% ranged over two orders of magnitude, from 6 ± 5 μg/L at SY11-10 m and 22% of the total arsenic. to 209 ± 104 μg/L at SY07-50 m (Table 2). The concentration of S2− in The vertical distributions of As in groundwater were different at 50 m deep wells was much higher than that in 10 m and 25 m deep different places. For nine out of the 13 monitoring points, higher As con- wells. Season variation in S2− concentration was consistent with that centration was observed in wells at 25 m, followed by wells at 50 m and in As and Fe concentrations, with lower concentration in dry season then at 10 m. Wells at SY01 and SY07 also had higher As concentration and higher concentration in rainy season (Figs. 8, 9, 10).

Fig. 7. Arsenic speciation in groundwater samples collected in July 2013 from the ﬁeld monitoring site.

Fig. 8. Temporal variation in As concentration of groundwater: As concentration in these wells showed seasonal variations during the monitoring period.

Fig. 9. Temporal variation in As concentration of groundwater: As concentration in these wells showed seasonal variations and increasing trends over the monitoring period.

5. Discussion with surface water bodies such as ponds, irrigation channel, and wet- lands (Fig. 1). The seasonal variation of groundwater As concentration In recent decades, surface water quality in our study area has been observed in 30 wells corresponded well with that of groundwater level degraded as a result of the rapid development of industry and agricul- (Figs. 3, 8, 9). Similar results have been reported in other waterborne ture and the rapidly increasing population. Accordingly, groundwater arsenic poisoning affected regions of the world. In northern China, arse- has become a critical source for drinking water supply. However, the nic concentration in groundwaters also rose with water levels. In the spatial and temporal variations of groundwater arsenic concentration Hetao Basin, Inner Mongolia, arsenic concentration was generally high make it difficult for groundwater quality and health risk assessment. between June and August when groundwater table was high, and de- Seasonal variations in arsenic concentration cause some wells to be clas- creased as the water table declined in September–October, and reached sified as safe during one season but unsafe during the rest of the year. the highest in November when the water table was the highest due to Increasing trends of arsenic concentration in some wells cause the the winter irrigation (Guo et al., 2012). In the Yinchuan Plain, shallow wells to be classified as safe at the beginning of the installation but groundwater also showed increasing As concentrations during May or more and more unsafe under long time use. It urgently needs to reveal October with rising water level associated with spring irrigation and the factors controlling the spatial and temporal variation of groundwa- summer irrigation (Han et al., 2012). High As concentrations during ter arsenic concentration and the mechanisms of arsenic mobilization monsoon due to the high water levels in West Bengal and Manipur, under reducing conditions to protect groundwater quality. India, also been observed (Savarimuthu et al., 2006; Oinam et al., 2011). In Araihazar area in Bangladesh, As concentration was high at 5.1. Effects of water level on groundwater As concentration one well during the rainy season (Dhar et al., 2008) and at another well in May-June (Cheng et al., 2005). For decades, the field monitoring site has been used as the farmland The response of As concentration to groundwater level may be relat- for growth of cotton, wheat, and rice. Precipitation/evaporation to- ed to a more reducing condition in the aquifers during high water level gether with irrigation using surface water induced seasonal variation periods. Williams and Oostrom (2000) reported that dissolved oxygen of groundwater level, with highs in July–October and lows in March– concentrations of water in the saturated zone decreased with elevation May (Fig. 3). Difference in fluctuation amplitude of water levels at differ- of water table in column experiments. The higher As concentration ent points could be related to their differences in hydraulic connection would be associated with lower Eh in groundwater during the rainy

Fig. 10. Temporal variation in Fe, Fe(II), S2−, and Mn concentrations of groundwater in the ﬁeld monitoring site. In each panel, data for groundwater from 10 m, 25 m, and 50 m deep wells are plotted using the average value of the 13 points in every month.

Fig. 11. Correlations between groundwater arsenic concentration and DOC in typical wells.

Freikowski et al., 2013). The availability of labile organic carbon as a and more detailed study, like the hydrology of the field site, characteris- driver of microbial reduction may therefore play an important role in tics of DOC, groundwater arsenic species temporal variation, arsenic controlling the spatial and temporal variation of groundwater As con- speciation and DOM properties in sediments, are needed since the tem- centration. High concentration of dissolved organic carbon (DOC) in poral variation of groundwater As concentration has important implica- groundwater has been reported in the Jianghan Plain (Gan et al., 2014). tions for its health risk assessment. In the study area, large amounts of organic matter are left over in the paddy and wheat fields each year. In paddy cultivation, harvested crop is cut from the middle of stem and the remaining half of the stem and Acknowledgments roots are plowed back for next cultivation. During the next cultivation, paddy fields are flooded with rain water and irrigation water, the de- The research work was financially supported by the National Natural composition of such plant remains increases the availability DOC. On Science Foundation of China (No. 40830748, No. 41120124003, and No. the other hand, infiltrating irrigation water will transport labile carbon 41102153), the China Geological Survey (No. 1212011121142), and the to directly increase the DOC concentration in groundwater. Temporal Ministry of Science and Technology (2012AA062602). Constructive variations of groundwater arsenic concentration in shallower well comments provided by Dr. Yan Zheng, Dr. Ratan Dhar, and another (10 m) at SY05 which was installed in paddy fields corresponded well anonymous reviewer helped to clarify and improve this manuscript. with that of DOC (Fig. 11). The high concentration of labile organic carbon as a driver of microbial reduction may well explain the higher As concentration in this shallower well. References Apart from various other factors, distance from pools rich in organic fi Benner, S.G., Polizzotto, M.L., Kocar, B.D., Ganguly, S., Phan, K., Ouch, K., Sampson, M., matter (paddy elds, ponds and waste dumping sites) may also play an Fendorf, S., 2008. Groundwater flow in an arsenic-contaminated aquifer, Mekong important role in the seasonal variation of As concentrations in ground- Delta, Cambodia. Appl. Geochem. 23, 3072–3087. water. A sharp increase in As concentration at SY07-10 m and 25 m in Berg, M., Tran, H.C., Nguyen, T.C., Pham, H.V., Schertenleib, R., Giger, W., 2001. Arsenic – contamination of groundwater and drinking water in Vietnam: a human threat. August December 2013 was mainly attributed to the sharp increase in Environ. Sci. Technol. 35, 2621–2626. DOC concentration (Fig. 11). The dumping of animals' waste, which Brikowski, T.H., Neku, A., Shrestha, S.D., Smith, L.S., 2014. Hydrologic control of temporal was only about five meters away from SY07, provided high concentra- variability in groundwater arsenic on the Ganges floodplain of Nepal. J. Hydrol. 518, – tion of available DOC. The available DOC can greatly accelerate the rate 342 353. Cheng, Z., van Geen, A., Seddique, A.A., Ahmed, K.M., 2005. Limited temporal variability of of microbially mediated reduction after infiltration into aquifers with arsenic concentration in 20 wells monitored for 3 years in Araihazar, Bangladesh. En- precipitation and/or irrigation water. However, the characteristics of viron. Sci. Technol. 29, 4759–4766. the DOC merit further study to reveal the mechanism on arsenic Deng, Y.M., Wang, Y.X., Ma, T., Gan, Y.Q., 2009. Speciation and enrichment of arsenic in strongly reducing shallow aquifers at western Hetao Plain, northern China. Environ. releasing. Geol. 56, 1467–1477. Dhar, R.K., Zheng, Y., Stute, M., van Geen, A., Cheng, Z., Shanewaz, M., Shamsudduha, M., 6. Conclusions Hoque, M.A., Rahman, M.W., Ahmed, K.M., 2008. Temporal variability of groundwater chemistry in shallow and deep aquifers of Araihazar, Bangladesh. J. Contam. Hydrol. 99, 97–111. In the field monitoring site, groundwater water tables fluctuated Emilie, E., 2009. Behavior of arsenic in the Jianghan Plain, China. Study field: Sha Hu vil- over the sampling period, with higher water levels during the rainy sea- lages, Hubei, P.R. ChinaA Dissertation Submitted to China University of Geosciences for the Master Degree. son and lower water levels during the dry season. Arsenic concentration Farooq, S.H., Chandrasekharam, D., Norra, S., Berner, Z., Eiche, E., Thambidurai, P., Stüben, in 34 of the 39 monitoring wells showed temporal variations which D., 2011. Temporal variations in arsenic concentration in the groundwater of were positively correlated to groundwater levels changes, with lower Murshidabad District, West Bengal, India. Environ Earth Sci. 62, 223–232. Fendorf, S., Michael, H.A., van Geen, A., 2010. Spatial and temporal variations of ground- concentration corresponding to lower water level during dry season water arsenic in south and southeast Asia. Science 328, 1123–1127. and higher concentration corresponding to higher water level during Freikowski, D., Neidhardt, H., Winter, J., Berner, Z., Gallert, C., 2013. Effect of carbon rainy season. Water level, redox-sensitive constituents, and labile sources and of sulfate on microbial arsenic mobilization in sediments of West Bengal, – organic carbon are there potential factors controlling the temporal India. Ecotoxicol. Environ. Saf. 91, 139 146. Frost, F., Frank, D., Pierson, K., Woodruff, L., Raasina, B., Davis, R., Davies, J., 1993. Aseason- variation of groundwater arsenic concentration. The rise in As concen- al study of arsenic in groundwater, Snohomish County, Washington, USA. Environ. trations during the rainy season could be attributed to enhanced reduc- Geochem. Health 15, 209–214. tive dissolution of iron oxyhydroxides and/or reductive desorption of Gan, Y.Q., Wang, Y.X., Duan, Y.H., Deng, Y.M., Guo, X.X., Ding, X.F., 2014. Hydrogeochemistry and arsenic contamination of groundwater in the Jianghan Plain, central China. As(V) as the conditions turn to be more reducing, while during the J. Geochem. Explor. 138, 81–93. dry season more As is scavenged onto fresh iron oxyhydroxides. Mean- Goncalves, J.A.C., de Lena, J.C., Paiva Jr., J.F., H.A.N., Pereira, J.C., 2007. Arsenic in the while, temporal variations showed compatible trends between As and groundwater of Ouro Preto (Brazil): its temporal behavior as influenced by the hydric – 2− regime and hydrogeology. Environ. Geol. 53, 785 793. Fe, Fe(II), S , indicating that As variation would be related to Fe and Goswami, A.M., Nath, B., Jana, J., Sahu, S.J., Sarkar, M.J., Jacks, G., Bhattacharya, P., S cycling in the aquifer systems. However, this mechanism remains Mukherjee, A., Polya, D.A., Jean, J.S., Chatterjee, D., 2008. Hydrogeochemical behavior ambiguous until further analysis of the sediments chemistry and miner- of arsenic-enriched groundwater in the deltaic environment: comparison between two study sites in West Bengal, India. J. Contam. Hydrol. 99, 22–30. alogical compositions are made. Additionally, the availability of organic Guo, H.M., Zhang, B., Li, Y., Berner, Z., Tang, X.H., Norra, S., Stuben, D., 2011. carbon as a driver of microbial reduction played an important role in Hydrogeological and biogeochemical constrains of arsenic mobilization in shallow controlling the spatial and temporal variations of groundwater arsenic aquifers from the Hetao basin, Inner Mongolia. Environ. Pollut. 159, 876–883. Guo, H.M., Zhang, Y., Jia, Y.F., Zhao, K., Li, Y., Tang, X.H., 2012. Dynamic behaviors of water in the Jianghan Plain. Nevertheless, the characteristics of the DOC levels and arsenic concentration in shallow groundwater from the Hetao Basin, Inner merit further study to reveal the mechanism on arsenic releasing. Mongolia. Journal of Geochemistry Exploration 135, 130–140. It is unclear why groundwater As concentrations in some wells Guo, H.M., Liu, C., Lu, H., Wanty, R.B., Wang, J., Zhou, Y.Z., 2013a. Pathways of coupled showed increasing trends over the monitoring period. Specific condi- arsenic and iron cycling in high atsenic groundwater of the Hetao basin, Inner Mongolia, China: an iron isotope approach. Geochim. Cosmochim. Acta 112, 130–145. tions such as proximity to waste dumping and heterogeneity of aquifers Guo, H.M., Zhang, Y., Jia, Y.F., Zhao, K., Kangjoo, K.I.M., 2013b. Spatial and temporal evolu- are potential reasons for such increasing trends. And it is not known tions of groundwater arsenic approximately along the flow path in the Hetao basin, – whether pumping from these wells disperses more oxygen into the Inner Mongolia. Chin. Sci. Bull. 58, 3070 3079. Han, S.B., Zhang, F.C., An, Y.H., Wang, Y.S., Wu, X., Wang, C., 2012. Spatial and temporal aquifers since the wells are unsealed during pumping. Another question patterns of groundwater arsenic in shallow and deep groundwater of Yinchuan is why the water level, redox sensitive elements, DOC, and arsenic in Plain, China. J. Geochem. Explor. 135, 71–78. groundwater show fluctuation over time, major ions and anions do Handler, R.M., Beard, B.L., Hohnson, C., Scherer, M.M., 2009. Atom exchange between aqueous Fe (II) and goethite: an Fe isotope tracer study. Environ. Sci. Technol. 43, not. The temporal variability makes characterization of groundwater 1102–1107. As hazard much more difficult. A longer term continuous monitoring Harvey, C.F., 2008. Poisoned waters traced to source. Nature 454, 415–416.

Harvey, C.F., Swartz, C.H., Badruzzaman, A.B.M., Keon-Blute, N., Yu, W., Ali, M.A., Jay, J., Saha, D., Sreehari, S.M.S., Dwivedi, S.N., Bhartariya, K.G., 2010. Evaluation of hydrogeo- Beckie, R., Niedan, V., Brabander, D., 2002. Arsenic mobility and groundwater extrac- chemical processes in arsenic-contaminated alluvial aquifers in parts of Mid-Ganga tion in Bangladesh. Science 298 (5598), 1602–1606. Basin, Bihar, Eastern India. Environmental Earth Sciences 61, 799–811. Islam, F.S., Gault, A.G., Boothman, C., Polya, D.A., Charnock, J.M., Chatterjee, D., Lloyd, J.R., Savarimuthu, X., Hira-Smith, M.M., Yuan, Y., von Ehrenstein, O.S., Das, S., Ghosh, N., 2004. Role of metal-reducing bacteria in arsenic release from Bengal delta sediments. Mazumder, D.N.G., Smith, A.H., 2006. Seasonal variation of arsenic concentrations in Nature 430, 68–71. tubewells in West Bengal, India. J. Health Popul. Nutr. 24, 277–281. Kang, C.G., Li, C.A., Wang, J.T., Zhao, L., 2009. Heavy minerals characteristics of sediments Smedley, P.L., Kinniburgh, D.G., 2002. A review of the source, behavior and distribution of in Jonathan plain and its indication to the forming of the Three Gorges. Earth Sci. arsenic in natural waters. Appl. Geochem. 17, 517–568. J. China Univ. Geosci. 34 (3), 419–426 (in Chinese with English abstract). Steinmaus, C.M., Yuan, Y., Smith, A.H., 2005. The temporal stability of arsenic concentra- Kocar, B.D., Polizzotto, M.L., Benner, S.G., Ying, S.C., Ung, M., Samreth, S., Suy, B., Phan, K., tions in well water in western Nevada. Environ. Res. 99, 164–168. Sampson, M., Fendorf, S., 2008. Integrated biogeochemical and hydrologic process Stute, M., Zheng, Y., Schlosser, P., Horneman, A., Dhar, R.K., Datta, S., Hoque, M.A., driving arsenic release from shallow sediments to groundwater of the Mekong Seddique, A.A., Shamsudduha, M., Ahmed, K.M., 2007. Hydrological control of As con- delta. Appl. Geochem. 23, 3059–3071. centrations in Bangladesh groundwater. Water Resour. Res. 43 (9), W09417. http:// Kocar, B.D., Borch, T., Fendorf, S., 2010. Arsenic repartitioning during biogenic sulfidization dx.doi.org/10.1029/2005WR004499. and transformation of ferrihydrite. Geochimica et Cosmochinica Acta 74 (3), Tareq, S.M., Safiullah, S., Anawar, H.M., Rahman, M.M., Ishizuka, T., 2003. Arsenic pollution 980–994. in groundwater: a self-organizing complex geochemical process in the deltaic sedi- Li, J.N., Yin, W.Y., Xu, H.T., Bie, M., Yan, X.D., 2010. Outcome analysis of screening on high mentary environment, Bangladesh. Sci. Total Environ. 313, 213–226. arsenic water in Honghu City, Hubei province in 2006 and 2007. Chin. J. Endemiol. 29, Thundiyil, J.G., Yuan, Y., Smith, A.H., Steinmaaus, C., 2007. Seasonal variation of arsenic 330–332 (in Chinese with English abstract). concentration in wells in Nevada. Environ. Res. 104, 367–373. Li, J.X., Wang, Y.X., Xie, X.J., Su, C.L., 2012. Hierarchical cluster analysis of arsenic and fluo- van Geen, A., Zheng, Y., Versteeg, R., Stute, M., Horneman, A., Dhar, R., Steckler, A., Gelman, ride enrichments in groundwater from the Datong basin, Northern China. Journal of A., Small, C., Ahsan, H., Graziano, J.H., Hussain, I., Ahmed, K.M., 2003. Spatial variability Geochemistry Exploration 118, 77–89. of arsenic in 6000 tube wells in a 25 km2 area of Bangladesh. Water Resour. Res. 39 Mayorga, P., Moyano, A., Anawar, H.M., García-Sánchez, A., 2013. Temporal variation of (5), 1140. http://dx.doi.org/10.1029/2002WR001617. arsenic and nitrate content in groundwater of the Duero River Basin (Spain). Phys. Wang, A.H., Zhao, S.J., 2007. Survey of endemic arsenic poisoning areas in Xiantao City, Chem. Earth 58–60, 22–27. Hubei Province. China Trop. Med. 7, 1486–1487 (in Chinese). McArthur, J.M., Banerjee, D.M., Hudson-Edwards, K.A., Mishra, R., Purohit, R., Ravenscroft, Williams, M.D., Oostrom, M., 2000. Oxygenation of anoxic water in a fluctuation water P., Cronin, A., Howarth, R.J., Chatterjee, A., Talukder, A., Lowry, D., Hounghton, S., table system: an experimental and numerical study. J. Hydrol. 230, 70–85. Chadha, D.K., 2004. Natural organic matter in sedimentary basins and its relation to Xie, X.J., Wang, Y.X., Ellis, A., Su, C.L., Li, J.X., Li, M.D., 2011. The sources of geogenic arsenic arsenic in anoxic ground water: the example of West Bengal and its worldwide in aquifers at Datong basin, northern China: constraints from isotopic and geochem- implications. Appl. Geochem. 19, 1255–1293. ical data. J. Geochem. Explor. 110, 155–166. Mladenov, N., Zheng, Y., Miller, M.P., Nemergut, D.R., Legg, T., Simone, B., Hageman, C., Ye, H.P., Xiong, P.S., Yang, Z.Y., Li, M.J., Du, D.Y., Zhou, S.H., Xiong, Q., Zhao, S.J., 2013. Rahman, M., Ahmed, K.M., Mcknight, D.M., 2010. Dissolved organic matter sources Preliminary study of the distribution profile of arsenic in sediment column of and consequences for iron and arsenic mobilization in Bangladesh aquifers. Environ. high-arsenic groundwater system at Jianghan Plain. Environ. Sci. Technol. (Chin. Sci. TechnoL. 44, 123–128. Ed.) 36, 24–27 (in Chinese with English abstract). Neumann, R.B., Ashfaque, K.N., Badruzzaman, A.B.B., Ali, M.A., Schoemaker, J.K., Harvey, Zeng, Z.H., Cai, W.D., Zhang, Z.L., 2004. The migration enrichment of Mn element in C.F., 2010. Anthropogenic influences on groundwater arsenic concentrations in groundwater and the controlling factors. Resour. Environ. Eng. 18 (4), 39–42 (in Bangladesh. Nat. Geosci. 3, 46–52. Chinese with English abstract). Nickson, R., McArthur, J., Burgess, W., Ahmed, K.M., Ravenscroft, P., Rahman, M., 1998. Zhao, D.J., Liu, Z.P., Xiong, Q.H., 2007. The vulnerability evaluation for groundwater pollu- Arsenic poisoning of Bangladesh groundwater. Nature 395, 338. tion in Jianghan Plain. Resour. En\viron. Eng. 21, 64–67 (in Chinese with English Nordstrom, D.K., 2002. Worldwide occurrences of arsenic in groundwater. Science 296, abstract). 2143–2145. Zheng, Y., van Geen, A., Stute, M., Dhar, R., Mo, Z., Cheng, Z., Horenman, A., Gavriell, I., Oinam,J.D.,Ramanathan,A.,Linda,A.,Singh,G.,2011.A study of arsenic and other Simpson, H.J., 2005. Geochemical and hydrogeological contrasts between shallow dissolved ion variations in the groundwater of Bishnupur district, Manipur, India. and deeper aquifers in two villages of Araihazar, Bangladesh: Implications for deeper Environmental Earth Science 62, 1183–1195. aquifers as drinking water sources. Geochimica et Cosmochinica Acta 69 (22), Polizzotto, M.L., Kocar, B.D., Benner, S.G., Sampsom, M., Fendorf, S., 2008. Near-surface 5203–5218. wetland sediments as a source of arsenic release to ground water in Asia. Nature Zhou, Y., 2009. Pesticide Transport and Enrichment in Water-Soil Environment: A Case 454, 505–508. Study of Jianghan Plain. A Dissertation Submitted to China University of Geosciences Rodríguez, R., Ramos, J.A., Armienta, A., 2004. Groundwater arsenic variations: the role of for the Doctor Degree of Philosophy. local geology and rainfall. Appl. Geochem. 19, 245–250. Zhou, Y., Wang, Y.X., Li, Y.L., Zwahlen, F., Boillat, J., 2013. Hydrogeochemical characteristics Romero, L., Alonso, H., Campano, P., Fanfani, L., Cidu, R., Dadea, C., Keegan, T., Thornton, I., of central Jianghan Plain, China. Environmental Earth Science 68, 765–778. Farago, M., 2003. Arsenic enrichment in waters and sediments of the Rio Loa (Second Region, Chile). Appl. Geochem. 18 (9), 1399–1416.