Journal of Hydrology 508 (2014) 12–27

Contents lists available at ScienceDirect

Journal of Hydrology

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

Chemical and isotopic constraints on evolution of groundwatersalinization in the coastal plain aquifer of Laizhou Bay, China

0022-1694/$ - see front matter � 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.jhydrol.2013.10.040

⇑ Corresponding author. Present address: Jia11#, Datun Road, Chaoyang District,100101 Beijing, China. Tel./fax: +86 10 64889849.

E-mail address: dmeihan@gmail.com (D.M. Han).

D.M. Han a,⇑, X.F. Song a, Matthew J. Currell b, J.L. Yang c, G.Q. Xiao c

a Key Laboratory of Water Cycle & Related Land Surface Processes, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing100101, Chinab School of Civil, Environmental and Chemical Engineering, RMIT University, Melbourne 3001, Australiac Tianjin Institute of Geology and Mineral Resources, Tianjin 300170, China

a r t i c l e i n f o s u m m a r y

Article history:Received 19 June 2013Received in revised form 28 September 2013Accepted 26 October 2013Available online 1 November 2013This manuscript was handled by CorradoCorradini, Editor-in-Chief, with theassistance of Michel Bakalowicz, AssociateEditor

Keywords:Laizhou BayCoastal aquifersGroundwater hydrochemistryStable isotopesSaltwater intrusion

A hydrochemical-isotopic investigation of the Laizhou Bay Quaternary aquifer in north China providesnew insights into the hydrodynamic and geochemical relationships between freshwater, seawater andbrine at different depths in coastal sediments. Saltwater intrusion mainly occurs due to two cones ofdepression caused by concentrated exploitation of fresh groundwater in the south, and brine water forsalt production in the north. Groundwater is characterized by hydrochemical zonation of water types(ranging from Ca–HCO3 to Na–Cl) from south to north, controlled by migration and mixing of saline waterbodies with the regional groundwater. The strong adherence of the majority of ion/Cl ratios to mixinglines between freshwater and saline water end-members (brine or seawater) indicates the importanceof mixing under natural and/or anthropogenic influences. Examination of the groundwater stable isotoped18O and d2H values (between �9.5‰ and �3.0‰ and �75‰ and �40‰, respectively) and chloride con-tents (�2 to 1000 meq/L) of the groundwater indicate that the saline end-member is brine rather thanseawater, and most groundwater samples plot on mixing trajectories between fresh groundwater(d18O of between �6.0‰ and �9.0‰; Cl < 5 meq/L) and sampled brines (d18O of approximately �3.0‰

and Cl > 1000 meq/L). Locally elevated Na/Cl ratios likely result from ion exchange in areas of long-termfreshening. The brines, with radiocarbon activities of �30 to 60 pMC likely formed during the Holocene asa result of the sequence of transgression-regression and evaporation; while deep, fresh groundwater withdepleted stable isotopic values (d18O = �9.7‰ and d2H = �71‰) and low radiocarbon activity (<20 pMC)was probably recharged during a cooler period in the late Pleistocene, as is common throughout northernChina. An increase in the salinity and tritium concentration in some shallow groundwater sampled in the1990s and re-sampled here indicates that intensive brine extraction has locally resulted in rapid mixingof young, fresh groundwater and saline brine. The d18O and d2H values of brines (��3.0‰ and �35‰) aremuch lower than that of modern seawater, which could be explained by 1) mixing of original (d18Oenriched) brine that was more saline than presently observed, with fresh groundwater recharged by pre-cipitation and/or 2) dilution of the palaeo-seawater with continental runoff prior to and/or during brineformation. The first mechanism is supported by relatively high Br/Cl molar ratios (1.7 � 10�3–2.5 � 10�3)in brine water compared with �1.5 � 10�3 in seawater, which could indicate that the brines originallyreached halite saturation and were subsequently diluted with fresher groundwater over the long-term.Decreasing 14C activities with increasing sampling depth and increasing proximity to the coastline indi-cate that the south coastal aquifer in Laizhou Bay is dominated by regional lateral flow, on millennialtimescales.

� 2013 Elsevier B.V. All rights reserved.

1. Introduction

Currently about 44 percent of the world’s population lives with-in 150 km of the coast (UN Atlas, 2010). Along much of China’s18,000 km of continental coastline, population densities average

between 580 and 1945 per square kilometer, and of China’s 1.3 bil-lion people, over 60% live in 14 coastal provinces (Shi, 2012). Therate of population growth in coastal areas is accelerating andincreasing water resources demand and economic developmentadds to pressure on the environment.

Seawater intrusion is a widespread geologic hazard in manycoastal regions around the world. It generally occurs when with-drawal of fresh groundwater from coastal aquifers results indeclining groundwater levels, facilitating lateral and/or vertical

D.M. Han et al. / Journal of Hydrology 508 (2014) 12–27 13

migration of saline water, causing deterioration of groundwaterquality (Barker et al., 1998; Barlow and Reichard, 2010). Saliniza-tion of coastal aquifers can occur due to simple, direct seawaterintrusion, but it can also involve a range of complex geochemicalprocesses which control water quality in different ways; e.g., in-ter-aquifer mixing, mobilization of brines, water–rock interactionand anthropogenic contamination (Vengosh et al., 2005). Risingsea levels, and increasing climate variability (e.g. longer droughts)are likely to place increasing pressure on coastal aquifers both innorthern China and globally in coming years (Werner andSimmons, 2009; Green et al., 2011). In this context, a rigorousunderstanding of the physical and chemical processes involved insalinization of coastal aquifers is vital for the future managementof these vulnerable water resources.

The Bohai Sea Coast is the region most seriously affected by sea-water intrusion in China, particularly Laizhou Bay in northernShandong province (Fig. 1). The problem has resulted due to inten-sive groundwater extraction during rapid development in the re-gion since the 1970s, and has been recognized by localauthorities since 1976 (Ji, 1991). The coastal aquifers of LaizhouBay have recently been the focus of attention due to increasingstress on fresh water resources and environmental degradation.Management of fresh groundwater resources in coastal aquifers re-quires an understanding of the processes controlling groundwatergeochemical evolution and flow system dynamics. By examininghydrochemical and isotopic data together with geology, topogra-phy and hydrogeological data, seawater intrusion can be compre-hensively assessed as a geologic hazard (e.g. Vengosh et al.,2005; Andersen et al., 2005; Jorgensen et al., 2008). The approachof using multiple sources of data can improve understanding ofhydrogeologic processes and aid the reliability of groundwaterflow models (e.g. Vengosh et al., 2002; Carrera et al., 2005). How-ever, comprehensive approaches to groundwater problems involv-ing use of chemistry and isotopic indicators are few in coastalaquifer research in China.

The objectives of the present paper are to provide an under-standing of the evolution of groundwater in the south coastal

Fig. 1. Location map of the study area, modified after TJR (2006). TDS concentrations weretable height contour. P–P0 line is the location of the cross-section in Fig. 2. Right-upper

plain of Laizhou Bay in relation to recharge, salinity sources,mixing behavior and palaeo-evolution of groundwater by con-sidering a wide selection of geochemical indicators. The dataexamined includes field observations of water table fluctuationsand analysis of hydrochemical and stable isotopic compositionsof groundwater. It is likely that much of the water beingexploited was recharged under different climatic conditions tothe present day (e.g. Chen et al., 2003; Kreuzer et al., 2009)and that the distribution of salinity has resulted from a combi-nation of past and recent processes, including a number ofphases of regression, transgression, brine formation and mixing.The results have significant implications for the management ofthe exploitation and recharge of coastal aquifers that are inten-sively exploited globally, and demonstrate the value of usingisotopic indicators in conjunction with other hydrogeologicalinformation.

2. Site description

2.1. Background of study area

The study area is the southern coastal area of Laizhou Bay,including the 43 km-long coastline and adjacent area of approxi-mately 1400 km2 (Fig. 1). Elevation of the coastal plain ranges from30 m a.s.l. in the south to 1–2 m a.s.l. in the north (Chen et al.,1997)with an average slope of 0.5‰ towards the sea. The Wei River, YuRiver and Jiaolai River flow through the study area into LaizhouBay. The area belongs to a sub-humid monsoon climate, with meanair temperature of 11.9 �C. The annual mean precipitation is660 mm, generally concentrated in June, July and August. Averageannual potential evapotranspiration is approximately 1400 mm.Past work in this region has been carried out examining the mech-anism, development, and preventative countermeasures for sea-water intrusion (e.g. Zhang and Peng, 1998; Xue et al., 2000), andthe formation of preferential channels of salt water intrusion(e.g. Zhang et al., 1996; Li et al., 2000).

measured in November 2005. Drawdown cone refers to area enclosed by 0 m watermap for showing the sampling wells in the study area.

14 D.M. Han et al. / Journal of Hydrology 508 (2014) 12–27

2.2. Hydrogeological conditions

Depositional facies of the aquifer sediments change from southto north, from alluvium in the south to diluvium, to marine sedi-ments on the coastal plain. The coastal plain consists of silty-mud-dy coastal sediments that are typical in China, while the seafloor ofLaizhou Bay is composed of an epi-continental sediment plain witha gently sloping gradient of 0.3–0.8‰ (Yang, 2005). The primaryaquifers in the study area are composed of Quaternary sediments,with a thickness of 30–50 m in the south, up to 360 m in the north(Peng et al., 1992). Close to the piedmont area, shallow and deepgroundwater is unconfined, while towards the coastline, deepgroundwater becomes confined due to increasing occurrence ofupper sandy clay. The strata in the upper proluvial fan in the southare mainly gravel and grit, grading to fine sand, silt, sandy clay, andsilty clay towards the coast. The aquifers are characterized by acomplex multi-layered framework (Fig. 2), often trending fromcoarse grain size in the lower part to fine-grained in the upper part.Coarse sand layers are highly permeable, with hydraulic conductiv-ities of 35–150 m/d (TJR, 2006). Bedrock, which outcrops in thesouthern mountainous regions of Laizhou Bay (TJR, 2006), ismainly composed of Archaeozoic and Proterozoic metamorphicrock and Cretaceous and Neogene basalt, andesite, trachyte andpyroclastic rock that consist of anorthite, albite, K-feldspar and bio-tite (Ning, 2004). The minerals in the aquifer include quartz, anor-thite, albite, plagioclase feldspar, picrite, biotite, aragonite,dolomite, calcite, kaolinite, gibbsite and Ca-montmorillonite, andminor amounts of evaporites (e.g., gypsum, anhydrite, halite, poly-halite, bischofite, epsomite) (Zhang et al., 1996; Han et al., 1996;Xue et al., 2000). Gypsum deposits are common in the coastal salt-water zone (Zhang and Peng, 1998). The clay layers are composedof illite, kaolinite, chlorite, and calcite-rich nodules (Zhao, 1996;Han et al., 1996).

Under natural conditions, fresh groundwater flows from southto north through the alluvial and marine sediments toward thecoastline. The unconfined aquifer is recharged by precipitation,horizontal flow of water from bedrock in the south, vertical infil-tration along the stream channel of Wei River and irrigation returnflows, and it is mainly discharged by groundwater exploitation andevaporation in shallow aquifers. Horizontal hydraulic gradientsvary from 2.64% in the upstream area to 1.64% along the coastline.Groundwater dynamics in the shallow unconfined and deepconfined aquifers are influenced by climatic and hydrological

Fig. 2. Change in groundwater levels of shallow aquifer and TDS concentrations along(2006). Water table in April 1992 obtained from Zhang et al. (1997). Legend: 1. Pre-Quaterclay; and 6. Boundary of bedrock.

conditions (e.g. rainfall and runoff amounts) and particularly in re-cent years, the volumes of groundwater exploitation. Tidal fluctu-ations in water level and inversion of water density with densesaltwater overlying less-dense freshwater can also drive down-ward leakage of saline water into coastal aquifers (Landmeyerand Belval, 1996), however these influences are considered to besmall in comparison to the hydraulic effects of groundwaterextraction, and have not been investigated in detail in this study.

2.3. Development of groundwater depression

Since the 1990s, over exploitation of fresh groundwater andbrine resources has resulted in variation of the natural groundwa-ter flow field, forming groundwater depression cones (Fig. 1). Thedepression cone near Changyi city initially formed in the shallowunconfined aquifer in 1984, and is currently characterized by anoval-shaped depression with northeast axis length of 24.3 kmand west-east direction 14.3 km (Fig. 1). The variation in the distri-bution of the shallow groundwater table and salinity (representedby the 2 g/L TDS contour) over recent years is shown in Fig. 2. Withthe increasing demand for fresh groundwater, the water table hasdescended and the 2 g/L TDS contour has moved inland towardsthe south. In 1985, the water table in the center of the depressioncone was at �2.4 m a.s.l. (water table depth of 9.8 m); in 1995 itwas at �9.3 m a.s.l. (water table depth of 17.5 m) (Zhang et al.,1997), and in 2007 it had declined to �14.5 m a.s.l. (water tabledepth of 22.7 m). The area encompassed by the cone of depression– defined as that enclosed by the 0 m a.s.l water table contour – in-creased from 170.0 km2 in 1984–248.6 km2 in 2000, to 353.1 km2

in 2007. The development of groundwater depression has allowedsaltwater (brine and/or seawater) to intrude into the fresh ground-water aquifers in a southward direction. The frontal interface ofsalt-water intrusion, north of Changyi, was pushed 2.9 km south-ward from 1980 to 2000 with a rate of 145 m/y, and a further3.2 km southward from 2000 to 2005 with a rate of 533.3 m/y(TJR, 2006). The interface between brackish and fresh groundwater(defined as 1 g/L TDS) in August 2007 was located at a distance of25.4 km from the coastline. Additionally, in the north of the studyarea, exploitation of brine water for the purpose of harvesting salt(by means of pumping and evaporation) have resulted in formationof a strip of decreased groundwater levels parallels the shoreline(Fig. 1), which began to appear in the 1990s. Currently, the width

the hydrogeological cross-section (P–P0 in Fig. 1) of study area. Modified after TJRnary Bedrock; 2. Gravel and sand; 3. Medium and fine sand; 4. Clayey Sand; 5. Sandy

Table 1Hydrochemical data from water samples in the south coastal aquifer of Laizhou Bay.

Watertype

Welllabel

Distance fromseashore (km)

Samplingtime

Welldepth (m)

EC(ms/m)

pH T (�C) Cl�(mg/L) NO�3(mg/L)

SO2�4

(mg/L)

HCO�3 (mg/L) Ca2+ (mg/L) Na+

(mg/L)K+

(mg/L)Mg2+

(mg/L)B(mg/L)

Sr(mg/L)

TDS (g/L)

Fresh SG03 45.4 August 2009 14 102 8.0 15.3 62.5 89.2 72.3 162.6 161.5 41.0 0.6 19.5 0.03 0.88 0.6Fresh SG04 41.2 August 2009 14 139 7.7 20.1 148.5 115.0 116.2 158.4 206.6 73.0 1.9 30.0 0.04 0.94 0.8Fresh SG06 36.8 August 2009 12 154 7.9 14.6 120.0 213.6 158.9 167.9 258.6 52.3 0.2 29.6 0.03 1.05 1.0Fresh SG07 33.6 August 2009 25 119 8.4 15.7 88.9 245.6 61.6 120.7 179.1 43.4 2.0 28.4 0.03 0.99 0.8Fresh SG08 31.7 August 2009 28 122 8.5 22.7 105.6 176.3 103.6 132.7 188.3 33.1 0.6 36.2 0.03 1.45 0.8Fresh SG11 23.7 August 2009 23 166 8.1 16.5 134.6 7.0 117.3 304.3 16.4 387.0 14.0 8.7 1.19 0.15 1.0Fresh SG29 22.1 August 2009 36 155 7.6 15.1 144.4 118.7 120.3 218.2 250.9 58.8 1.2 35.1 0.04 1.21 0.9Fresh SG53 32.2 August 2009 25 142 8.3 15.7 95.2 346.5 65.4 136.9 201.2 52.1 1.2 36.1 0.03 1.27 0.9Fresh SG42 40.8 August 2009 30 99 8.2 18.9 119.5 31.1 95.0 113.8 99.5 54.3 3.5 16.4 0.03 0.43 0.5Fresh SG57 31.6 August 2009 22 115 8.1 16.2 140.4 90.2 82.7 119.6 159.1 69.3 1.1 20.2 0.04 0.64 0.7Fresh SG58 32.2 August 2009 40 92 8.3 21.2 90.2 54.9 77.4 111.2 131.2 50.3 0.7 18.5 0.03 0.63 0.5Fresh SG59 29.4 August 2009 33 93 8.7 24.2 116.8 37.3 26.6 132.7 86.5 57.8 0.4 38.6 0.05 1.52 0.5Fresh SG03 45.4 June 2008 17 102 7.4 15.7 73.1 81.3 77.8 159.5 108.3 43.0 1.1 20.2 0.03 0.89 0.6Fresh SG04 41.2 June 2008 14 170 7.2 17.8 176.5 218.4 152.4 192.2 171.6 89.3 2.0 33.7 0.04 1.01 1.0Fresh SG06 36.8 June 2008 12 111 7.3 15.3 91.0 130.1 110.2 126.8 140.8 41.9 0.5 22.4 0.03 0.72 0.7Fresh SG07 33.6 June 2008 25 105 7.7 16.6 83.4 222.1 54.6 134.5 141.5 43.0 1.9 27.0 0.05 0.96 0.7Fresh SG30 30.4 June 2008 30 154 7.6 14.9 118.9 228.2 115.7 199.8 144.6 65.5 2.0 44.9 0.06 1.29 0.9Fresh SG40 14.8 June 2008 14 214 8.2 15.4 301.1 101.2 107.6 30.5 413.9 19.3 24.0 2.13 0.38 1.0Fresh SG34 23.7 June 2008 40 128 8.6 15.6 141.6 95.0 220.6 2.3 298.0 7.7 3.1 0.71 0.04 0.8Fresh SG33 24.1 June 2008 38 112 7.6 15.8 100.4 20.8 88.6 287.7 57.9 101.7 14.4 43.5 0.29 0.77 0.7Fresh SG29 22.1 June 2008 36 122 7.6 15.5 117.7 83.4 89.1 256.7 124.6 51.8 1.4 26.9 0.04 0.94 0.8Fresh SG43 25.8 June 2008 45 202 8.1 16.3 90.7 59.6 489.7 6.0 270.8 11.4 6.4 0.77 0.09 0.9Fresh SG42 40.8 June 2008 30 85 7.2 16.8 113.0 12.6 96.9 90.3 106.2 61.0 5.1 16.5 0.04 0.43 0.5Fresh DG14 23.3 August 2009 210 95 8.2 112.3 23.9 165.2 80.3 102.1 1.4 21.9 0.06 0.79 0.5Brackish SG10 25.4 August 2009 36 334 8.2 15.9 288.4 645.0 381.6 31.7 680.0 28.4 35.7 1.35 0.45 2.1Brackish SG12 22.0 August 2009 50 236 8.2 16.2 256.5 2.9 276.9 329.2 10.0 511.8 16.2 9.9 0.84 0.12 1.4Brackish SG19 19.0 August 2009 25 558 8.0 16.2 1288.2 204.3 300.3 43.9 1103.0 29.2 60.4 1.50 0.81 3.0Brackish SG21 18.0 August 2009 6 388 7.9 15.8 646.6 164.5 448.7 359.4 336.9 410.6 22.8 96.9 0.26 2.02 2.5Brackish SG04’ 23.7 August 2009 43 216 7.7 27.2 322.8 26.6 151.5 254.4 27.9 421.1 11.8 22.1 0.67 0.37 1.2Brackish SG40 14.8 August 2009 14 227 8.4 21.2 338.0 185.8 275.4 122.2 273.3 11.6 76.7 0.48 1.08 1.3Brackish SG39 18.2 August 2009 40 289 8.0 15.2 619.3 22.7 171.5 228.2 260.2 206.3 7.5 74.7 0.13 1.65 1.6Brackish SG33 24.1 August 2009 38 462 8.4 15.7 1145.4 286.4 267.0 245.3 213.3 715.5 1.5 77.2 0.23 2.03 3.0Brackish SG44 25.4 August 2009 26 235 8.0 16 220.1 130.7 143.3 359.4 49.4 390.1 20.8 47.9 0.65 0.58 1.4Brackish SG47 21.1 August 2009 35 430 7.8 15.4 747.5 417.9 331.0 39.7 954.9 32.2 10.0 1.15 0.45 2.5Brackish SG50 19.0 August 2009 50 544 8.0 17.2 1220.2 296.2 275.4 17.0 1129.0 21.2 27.8 1.69 0.26 3.0Brackish SG51 16.5 August 2009 40 448 8.0 15.1 841.2 16.1 395.2 288.5 42.2 891.8 33.3 45.0 1.05 0.42 2.6Brackish SG52 16.7 August 2009 35 327 7.5 23.1 498.3 465.6 314.8 220.1 420.3 18.2 85.0 0.43 1.40 2.0Brackish SG54 22.4 August 2009 37 200 7.4 17.5 236.2 74.5 146.9 301.6 200.9 186.1 10.0 52.3 0.26 1.14 1.2Brackish SG02 51.3 June 2008 14 202 7.0 15.2 392.7 221.5 123.7 209.4 225.8 121.5 0.7 51.3 0.06 1.95 1.3Brackish SG10 25.4 June 2008 40 318 8.0 16 276.1 649.8 406.2 11.6 681.5 29.1 36.9 1.47 0.35 2.1Brackish SG12 22.0 June 2008 35 190 8.2 14.6 259.2 12.7 190.7 248.5 24.3 373.4 21.6 24.1 0.86 0.29 1.2Brackish SG36 19.2 June 2008 29 265 8.3 16.8 516.2 91.6 294.4 8.5 549.2 16.4 15.0 1.36 0.17 1.5Brackish SG39 18.2 June 2008 40 242 7.6 17.8 637.4 11.6 111.2 235.2 186.0 156.3 7.3 75.1 0.17 1.52 1.4Brackish SG35 22.4 June 2008 37 279 8.0 15 371.3 148.2 199.8 372.0 28.8 583.8 25.7 30.0 1.04 0.40 1.8Brackish SG44 25.4 June 2008 26 246 8.0 15.2 280.9 129.3 149.5 775.2 31.7 402.3 31.1 69.4 0.74 0.64 1.9Brackish SG47 21.1 June 2008 35 494 8.2 16.5 878.3 438.1 663.2 20.4 1006.0 37.3 16.3 1.24 0.26 3.1Brackish SG48 15.6 June 2008 30 554 7.6 13.6 1320.6 23.6 242.9 551.2 282.6 810.6 5.2 120.3 0.60 1.85 3.4Brackish DG04 24.1 August 2009 65 807 7.7 17.3 2120.3 25.6 354.6 215.1 268.4 1501.0 27.9 189.4 0.65 2.58 4.7Brackish DG07 11.8 August 2009 270 472 8.0 20.6 1026.3 8.0 200.1 174.4 25.3 743.9 6.8 72.8 0.09 0.44 2.3Brackish DG09 18.7 August 2009 210 204 8.1 19.6 464.3 71.8 130.6 162.3 201.1 3.2 48.2 0.05 1.52 1.1Brackish DG10 5.7 August 2009 180 1487 7.6 19.5 5049.2 605.5 135.1 554.4 2420.0 15.4 369.4 0.32 5.09 9.1Brackish DG04 24.1 June 2008 38 304 7.7 18.3 641.2 26.8 170.2 275.2 36.6 590.5 18.9 50.3 0.86 0.66 1.8Brackish DG08 9.4 June 2008 180 188 7.8 22.9 334.3 94.7 344.8 12.6 388.9 1.9 13.6 0.07 0.15 1.2Saline SG55 21.4 August 2009 18 2860 7.0 16.4 9713.2 1132.1 194.1 1228.0 4359.0 58.5 988.0 0.65 13.16 17.7

(continued on next page)

D.M

.Han

etal./Journal

ofH

ydrology508

(2014)12–

2715

Tabl

e1

(con

tinu

ed)

Wat

erty

peW

ell

labe

lD

ista

nce

from

seas

hor

e(k

m)

Sam

plin

gti

me

Wel

lde

pth

(m)

EC (ms/

m)

pHT

(�C

)C

l�(m

g/L)

NO� 3

(mg/

L)SO

2� 4

(mg/

L)

HC

O� 3

(mg/

L)C

a2+

(mg/

L)N

a+

(mg/

L)K

+

(mg/

L)M

g2+

(mg/

L)B (m

g/L)

Sr (mg/

L)TD

S(g

/L)

Sali

ne

SG56

4.8

Au

gust

2009

2357

307.

116

.620

734.

525

32.2

255.

760

8.1

1370

0.0

469.

515

90.0

8.40

8.29

39.9

Sali

ne

SG38

16.9

Jun

e20

0845

2470

7.5

15.5

7727

.016

18.5

273.

621

7.4

5213

.013

7.7

511.

42.

943.

4515

.7Sa

lin

eD

G05

8.9

Au

gust

2009

125

5400

7.4

16.4

1947

0.5

2276

.314

6.9

905.

511

070.

019

1.3

1824

.00.

7311

.24

35.9

Sali

ne

DG

115.

1A

ugu

st20

0918

057

207.

318

.521

010.

924

37.7

149.

561

9.3

1222

0.0

267.

817

09.0

1.05

8.97

38.4

Sali

ne

DG

068.

9Ju

ne

2008

8645

407.

417

1610

1.3

1882

.615

2.2

620.

491

26.0

177.

114

29.0

0.79

7.78

29.5

Sali

ne

DG

058.

9Ju

ne

2008

125

2240

7.4

17.6

8709

.622

8.5

734.

816

2.2

1182

.036

01.0

37.9

920.

90.

3311

.50

15.6

Bri

ne

DG

08’

9.4

Au

gust

2009

5814

,860

6.5

16.6

6420

1.0

8526

.519

0.2

1046

.039

280.

010

14.0

5997

.03.

4023

.47

120.

3B

rin

eD

G12

9.1

Au

gust

2009

6514

,250

6.5

16.2

6728

4.6

9099

.120

4.6

1052

.036

860.

096

3.2

5548

.04.

0619

.77

121.

0B

rin

eD

G13

4.7

Au

gust

2009

7515

,120

6.4

16.3

6646

4.6

8152

.114

8.2

1532

.040

510.

011

37.0

7153

.02.

1923

.97

125.

1B

rin

eD

G07

11.8

Au

gust

2007

6013

,750

6.6

7108

8.0

8576

.646

3.0

1006

.731

495.

595

0.5

4767

.611

8.9

Bri

ne

SG41

8.9

July

2007

3014

,100

6.5

6075

6.3

6932

.686

5.2

1076

.333

140.

069

9.0

4600

.110

8.4

Bri

ne

SG27

8.5

Nov

embe

r20

0540

15,2

606.

887

674.

010

670.

955

8.0

975.

837

513.

514

53.2

5518

.414

4.6

Sali

ne

SWA

ugu

st20

0977

507.

428

.231

761.

155

63.4

14.4

860.

919

410.

066

9.5

4011

.02.

464

13.1

62.3

16 D.M. Han et al. / Journal of Hydrology 508 (2014) 12–27

of the depression varies from 2.9 km in the east to 7.1 km in thewest.

Groundwater salinity in the study area varies both in the verti-cal and lateral directions. Subsurface brines with an average TDS of110 g/L occur in the northern part of the study area. Brine water inQuaternary aquifers is mainly distributed within 10 km of theshoreline, (Zhang and Peng, 1998) extending to a depth of up to60 m (Jiang, 1992). The brines have previously been proposed tohave originated from palaeo-seawater intrusion, following threeoccurrences of marine incursion and regression along the coast ofLaizhou Bay since the Upper Pleistocene (Xue et al., 2000). Thereare three major layers containing brine in the sediments, in thedepth ranges 0–14.8 m, 33.2–42.3 m, and 58.6–74.1 m, with thick-nesses of 7.3 m, 12.7 m, and 9.9 m, respectively. There are a num-ber of clay inter-bed layers between different layers of marinedeposits, which locally form confining layers, and probably inhib-ited vertical mixing between these individual layers of brine undernatural conditions. In the downstream region of the Wei River, athick paleochannel containing coarse grained sediments occurs,deposited in the Late Pleistocene (Fig. 1). Paleochannels may pro-vide further potential conduits for brine and/or seawater intrusion(Li et al., 2000).

The cones of depression have become the dominant influenceon groundwater flow directions, resulting in hydraulic gradientsdirected from the northern saltwater areas towards the freshersouth (with local influence of the cones also important). Althoughthere may be shallow groundwater discharge into sea, it is impos-sible that groundwater within our investigated depths currentlydischarges to the sea due to these gradients.

2.4. Environmental issues

Environmental problems caused by salt water intrusion, such asdegeneration of groundwater quality, soil salinization and ecolog-ical deterioration, have become a major obstacle in the future eco-nomic development of this coastal area. Due to the lack ofsignificant recharge to the deep confined aquifer and intrusion ofsaltwater due to pumping, there is a significant risk that this waterresource may be irreversibly compromised. That rapid changes tothe distribution of groundwater levels and quality have occurredover a timescale of years indicates that the flow field is highlyresponsive to changes in the water balance. Climate change is alsopredicted to reduce overall rainfall in the study area (Liu et al.,2004). In the 1980s when rainfall was the lowest over the last fivedecades, saltwater intrusion occurred rapidly in the study area,while the velocity of intrusion decreased as rainfall briefly in-creased after 1990 (Feng et al., 2006). This is probably attributableto the responsiveness of coastal water users to water shortage –e.g. they turn to groundwater in the absence of rainfall. Future pro-tection and management of the groundwater resources in thecoastal region of Laizhou Bay and other areas must take into ac-count the likely impacts on the flow field driven by water demand,which is in turn responsible for the extent of saltwater intrusion,whether seawater or brine (e.g. Ferguson and Gleeson, 2012).

3. Materials and methods

3.1. Sampling

Water sampling (a total of 65 samples) was conducted at differ-ent aquifer depths over 2 campaigns from June 2008 to August 2009in the Changyi-Liutuan area of Laizhou Bay (Fig. 1). Groundwatersamples can be divided into shallow groundwater (<60 m depth,hereafter denoted as SG) and deep groundwater (>60 m depth, here-after denoted as DG), with shallow groundwater corresponding to

Table 2Isotope composition for water samples in the south coastal area of Laizhou Bay.

Label Water type Sampling time d18O(‰) d2H(‰) Label Water type Sampling time d18O(‰) d2H(‰)

SG03 Fresh August 2009 �55 �7.5 SG47 Brackish August 2009 �49 �5.4SG04 Fresh August 2009 �50 �6.6 SG50 Brackish August 2009 �57 �7.5SG06 Fresh August 2009 �46 �6.0 SG51 Brackish August 2009 �56 �7.5SG07 Fresh August 2009 �51 �6.6 SG52 Brackish August 2009 �54 �7.4SG08 Fresh August 2009 �48 �6.0 SG54 Brackish August 2009 �53 �7.3SG11 Fresh August 2009 �58 �8.0 SG02 Brackish June 2008 �58 �7.0SG29 Fresh August 2009 �56 �7.0 SG10 Brackish June 2008 �53 �6.5SG53 Fresh August 2009 �57 �7.4 SG12 Brackish June 2008 �57 �8.0SG42 Fresh August 2009 �40 �4.4 SG36 Brackish June 2008 �52 �8.3SG57 Fresh August 2009 �36 �4.1 SG39 Brackish June 2008 �54 �6.7SG58 Fresh August 2009 �41 �5.0 SG35 Brackish June 2008 �59 �7.2SG59 Fresh August 2009 �54 �8.1 SG44 Brackish June 2008 �58 �7.5SG03 Fresh June 2008 �53 �6.9 SG47 Brackish June 2008 �51 �5.8SG04 Fresh June 2008 �49 �6.0 SG48 Brackish June 2008 �51 �7.0SG06 Fresh June 2008 �45 �5.2 DG04 Brackish June 2008 �58 �7.2SG07 Fresh June 2008 �53 �6.9 DG04 Brackish August 2009 �57 �7.8SG30 Fresh June 2008 �56 �7.3 DG07 Brackish August 2009 �70 �9.3SG40 Fresh June 2008 �45 �4.9 DG09 Brackish August 2009 �71 �9.6SG34 Fresh June 2008 �57 �7.4 DG10 Brackish August 2009 �62 �8.2SG33 Fresh June 2008 �61 �8.0 DG08 Brackish June 2008 �74 �9.6SG29 Fresh June 2008 �55 �6.8 SG55 Saline August 2009 �50 �7.0SG43 Fresh June 2008 �58 �7.6 SG56 Saline August 2009 �35 �3.6SG42 Fresh June 2008 �35 �3.6 SG38 Saline June 2008 �40 �3.7DG14 Fresh August 2009 �71 �9.7 DG05 Saline August 2009 �60 �7.5SG10 Brackish August 2009 �54 �6.8 DG11 Saline August 2009 �53 �6.4SG12 Brackish August 2009 �52 �6.7 DG06 Saline June 2008 �61 �7.2SG19 Brackish August 2009 �55 �7.4 DG05 Saline June 2008 �64 �7.7SG21 Brackish August 2009 �47 �6.9 DG08 Brine August 2009 �41 �3.9SG40 Brackish August 2009 �48 �6.0 DG12 Brine August 2009 �36 �3.5SG39 Brackish August 2009 �52 �6.3 DG13 Brine August 2009 �39 �4.0SG33 Brackish August 2009 �56 �7.2 SG41 Brine July 2006 �43 �6.2SG44 Brackish August 2009 �56 �7.0 SG27 Brine November 2005 �39 �4.7

D.M. Han et al. / Journal of Hydrology 508 (2014) 12–27 17

pore water in the shallow unconfined aquifer and deep groundwa-ter to water from the confined aquifer layer(s). Most groundwatersamples were collected along a monitoring profile oriented N-Sfrom the coast of Laizhou Bay (Fig. 2). Surface water samples(labeled in SU) were also collected. Groundwater samples have beenclassified into four types on the basis of salinity; fresh, brackish, sal-ine and brine water, with Total Dissolved Solids (TDS) concentra-tions of <1 g/L, 1–10 g/L, 10–100 g/L, and >100 g/L, respectively.

3.2. Analytical methods

Physical and chemical parameters, including depth to water ta-ble, temperature, pH and electrical conductivity (EC) were directlymeasured in the field using portable meters, allowing for temper-ature compensation and calibration with appropriate standards.Samples for major anion analysis were collected in polyethylenebottles, tightly capped and stored at 4 �C until analysis. All watersamples were filtered promptly after collection for analysis of hyd-rochemical composition using 0.45 lm membrane filters. Samplesfor cation analysis (Na, K, Mg and Ca) were preserved in acid-washed polyethylene bottles, and acidified to pH�2 with 6 NHNO3. The concentrations of cations, together with B and Sr, weremeasured using inductively coupled plasma analysis (ICP-OES) onfiltered samples. Concentrations of Cl, SO4, NO3 and F were deter-mined by a High Performance Ion Chromatograph (SHIMADZU)at the Key Laboratory of Water Cycle & Related Land Surface Pro-cesses, Institute of Geographic Sciences and Natural Resources Re-search, China Academy of Sciences. Alkalinity was determined onfiltered samples in the field, by titration with H2SO4 (0.22N) onthe day of sample collection. Charge balance errors in all analysesare less than 8%. The hydrochemical and physical data are shown inTable 1.

The analysis of stable isotopes of water (d18O and d2H) werecarried out at the Key Laboratory of Water Cycle & Related Land

Surface Processes, Institute of Geographic Sciences and Natural Re-sources Research, China Academy of Sciences, using a Finigan MAT253 mass spectrometer. The d18O and d2H values were measuredrelative to internal standards that were calibrated using IAEA stan-dards, using equilibration with CO2–He for 18O and H–He for 2H.The d18O and d2H values are given hereafter in d-units calculatedwith respect to VSMOW (Vienna Standard Mean Ocean Water) ex-pressed in ‰. The analytical precision of long-term standard mea-surements for d2H is ±2‰ and for d18O is ±0.5‰. Results are shownin Table 2. Some discrepancies in the measurement of the d2H andd18O concentrations and activities from saline water, due to the ef-fect of hydration of ions, have been pointed out (Horita and Gat,1988; Cartwright et al., 2009). While this effect may have some im-pact on our results – particularly for d2H, where high salinity inbrines may cause discrepancies of up to 10‰; this will not be sub-stantial for the majority of the sample set. In order to reduce theeffect of this discrepancy in our interpretations, we have generallyused relationships between d18O (where the maximum expecteddifference is <0.5‰) and Cl to discuss different groundwater mix-ing processes. The analytical methods used for determining radio-carbon and tritium activities in water samples have been describedin Han et al.(2011).

4. Results

4.1. Groundwater level and salinity monitoring

There are 5 long-term observation wells (locations ‘‘a–e’’ inFig. 1) monitoring water table dynamics in the shallow Quaternaryaquifers in the study area. Plots of observed water levels in theunconfined aquifer are shown in Fig. 3. The monitoring resultscan be summarized as follows: (i) Qingxiang well (‘‘a’’ in Fig. 3)shows the water table increasing by 4.3 m from March 2006 to

Fig. 3. Plots of observed water levels in five wells from the unconfined aquifer (seeFig. 1 for locations) Ground surface elevation is 2.5 m a.s.l. in ‘‘a’’-Qingxiang,4.6 m a.s.l. in ‘‘b’’-Nanfanjiazhuang, 6.2 m a.s.l. in ‘‘c’’-Liujiapu, 7.1 m a.s.l. in ‘‘d’’-Lijiapu, and 14.3 m a.s.l. in ‘‘e’’-Weihe River bridge.

18 D.M. Han et al. / Journal of Hydrology 508 (2014) 12–27

August 2007, and decreasing by 4 m to November 2009. The waterlevel is probably controlled by recent adjustment measuresrestricting brine exploitation in the area; a number of wells havebeen abandoned or had extraction limits emplaced since 2006.TDS values of the brines in this region range from 108.4to144.6 g/L. (ii) The water table at Nanfanjiazhuang (‘‘b’’ in Fig. 3)showed a decrease by 3.6 m during the monitoring period fromMarch 2006 to July 2008, meanwhile the TDS increased from2.5 g/L in November 2005 to 7.2 g/L in August 2007, indicative ofsaltwater intrusion. (iii) The water table at Lijiapu monitoring well(‘‘c’’ in Fig. 3) decreased 6.5 m from March 2006 to October 2010,with a velocity of �1.6 m/a. Groundwater similarly increased inTDS from 0.9 g/L in November 2005, to 1.7 g/L in August 2007.(iv) The monitoring well ‘‘d’’ in Liujiapu is located at the marginof the cone of depression. The water level declined by 6.4 m fromMarch 2006 to October 2010, although seasonal recharge duringJune to August allowed for some recovery of water levels each year.Monitoring well ‘‘e’’ near the Wei River Bridge, is located at a depthof 16 m in the piedmont plain, where the aquifer is composed ofcoarse sand and gravel (TJR, 2006). The water table in this well des-cended by 1.9 m from March 2006 to October 2010; although inthe dry season (outside of the monsoon period), an increase inwater table in this well may have be caused by decreased ground-water exploitation or use of transferred water for irrigation. Thegroundwater had a TDS of 0.8 g/L in November 2005 and 0.4 g/Lin August 2007, which is consistent with this mechanism.

4.2. Hydrochemistry

4.2.1. Salinity distribution of regional groundwaterGroundwater salinity and hydrochemistry exhibits regular

zonation from south to north with different types of groundwaterdistributed in strips parallel to the coast. The TDS concentrationsrange from 0.5 g/L in the south to 144.6 g/L in the north, withbrackish water between the fresh zone and brine zones (Fig. 4).In areas of intensive exploitation, the zoned pattern of salinitiesis disrupted (Fig. 4), showing that the pumping is having a majorimpact on flow fields and salinity distributions. The distributionof the salinities with depth depends on hydrogeological conditions,such as the presence of local paleo-channels and confining strata,as well as the intensity of groundwater exploitation. Cross sectionsof TDS and Cl distributions with depth (Figs. 2 and 5) provide adepiction of the degree of hydrochemical variability through thestrata. In general, the salinity pattern resembles a classic coastalsalt-fresh water wedge interface, with high salinities at depthand near the coast, with some local variations near the two pump-ing centers (Figs. 2 and 5).

4.2.2. Major ion trendsThe anions in most water samples are dominated by Cl (up to

93% of total anion concentrations, where% is proportion to total an-ions in miliequivalents), this is especially true (>85%) in saline andbrine water (Cl concentrations up to 88 g/L). The cations of thefresh groundwater are dominated by Ca (67%) while the brackish,saline and brine groundwater are characterized by higher Na con-centrations. Hydrochemical types of groundwater change fromsouth to north in the series (Fig. 4): Ca–HCO3�Cl�SO4, Ca–Cl�SO4

and Ca–Cl�HCO3(SO4) in the fresh water zone, to Na–Cl�HCO3 inthe brackish water zone, to Na–Cl in saline and brine water zones.Na–HCO3 type groundwater (such as SG11 and SG43) is also pres-ent sporadically near the center of groundwater depression cone,and fresh groundwater (such as SG40 and SG48) of Na–Cl type islocally present in the brackish water zone.

The evolution of the major ions in groundwater up to high salin-ities can be seen from Fig. 6. Towards the coastline, Na, SO4, and Mgconcentrations increase approximately linearly with chloride con-centrations (Fig. 6a, b and d). Apart from the brines, Mg contentsare generally well below those of seawater (104.7 meq/L)(Fig. 6d). The saline and brine water are characterized by low ratiosof Na/Cl, SO4/Cl, Ca/SO4, and Mg/Cl compared to the fresh andbrackish water (Fig. 7); indicating a contribution of mineral weath-ering to the overall solute load at low salinities, which is over-whelmed at high salinity. In the saline waters, Na/Cl ratios arenearly constant and are mostly between the seawater value of�0.86 and observed brine water value of �0.71 (Fig. 7). The freshand brackish groundwaters show a wide range of SO4/Cl ratios(0.09–1.74) relative to seawater (0.05) and brine (0.08–0.10), andplot above the mixing line between fresh water and seawater/brine, showing some sulfate enrichment relative to chloride(Fig. 7b).

The Br/Cl ratio may be used to indicate the origin of groundwa-ter salinity (Edmunds et al., 2006). In this study, there is a generallinear relationship between the bromide concentration and chlo-ride concentration (Fig. 8), with Br slightly enriched relative tothe expected path for meteoric water or seawater evaporation.The Br/Cl ratio maintains a constant value until halite saturationis reached during the evaporation of meteoric water or seawater(Bottomley et al., 1994). In this case, none of the samples were athalite saturation (see Section 4.2.3). Br/Cl ratios of most groundwa-ters plot above the evaporation line (Fig. 8), and are highest for sev-eral shallow brackish groundwater samples.

Fig. 4. Distribution of different hydrochemical types of shallow groundwater in the study area.

Fig. 5. Groundwater flow patterns along the Changyi-Liutuan cross-section (P–P0 location in Fig. 1). 1. Gravel and sand; 2. Medium and fine sand; 3. Clayey Sand; 4. Sandyclay; 5. Pre-Quaternary Bedrock; 6. Boundary of bedrock; 7. Shallow groundwater level in November 2005; 8. Potentiometric surface for the confined aquifer in November2005; 9. Contour line of Cl� concentration (in mg/L) in November 2005; and 10. Concentrated groundwater exploitation zone. Arrows indicate groundwater flow direction.

D.M. Han et al. / Journal of Hydrology 508 (2014) 12–27 19

4.2.3. Saturation indicesSaturation indices were calculated by PHREEQC 2 (Parkhurst

and Appelo, 1999) to better understand the hydrogeochemical pro-cesses that take place in the aquifer. The saturation indices of cal-cite (SIcalcite) and gypsum (SIgypsum) vary between �0.5 and +0.5and indicate saturation (or equilibrium) or near saturation with re-spect to these minerals in the saline and brine water (Fig. 9a and b).The evolution of halite saturation indices shows an increasing

trend when plotted against the chloride and sulfate contents; allsamples are below saturation (SIhalite < 0, Fig. 9c and d).

4.3. d18O and d2H compositions

Isotopic values of groundwater samples from the coastal aquiferat Laizhou Bay are related to the recharge sources of the water, mix-ing and modification by evaporation. Groundwater samples in the

Fig. 6. Hydrochemical relationships between selected ions in groundwater and mean seawater composition, as given by Stumm and Morgan (1981). Explanation: Seawaterevaporation trajectory from Fontes and Matray (1993): G, H, E, S, C and B stand for point of precipitation of gypsum, halite, epsomite, sylvite, carnallite and bischofite,respectively. The mean values for rainfall are referenced from Liu et al. (1993), Wang et al. (1993), and Huang et al.(1993). Mixing line between seawater and freshgroundwater (SG03) is shown in dashed lines, and mixing line between fresh groundwater (SG03) and brine water (DG13) is shown in solid red lines. Numbers along themixing lines show the percentage (%) of seawater/brine with 10% increments under the simple mixing behavior.

20 D.M. Han et al. / Journal of Hydrology 508 (2014) 12–27

area have d18O values between �9.7‰ and �3.5‰, and d2H valuesbetween �74‰ and �35‰ (Table 2). The stable isotope values areshown in Fig. 10, along with the GMWL (Craig, 1961) and a localmeteoric water line (LMWL: d2H = 7.4d18O + 1.1), obtained frommonthly rainwater data sampled at the Changyi weather stationin 2006 (n = 12, R2 = 0.91). The mean stable isotopic value in modernprecipitation is�6.9‰ for d18O, and�48‰ for d2H. Most water sam-ples, especially the shallow groundwater, fall along a mixing trendline between these rain-like values and shallow brine and/or in-shore seawater from Laizhou Bay (d2H = 5.4�d18O � 16.5, n = 63,r2 = 0.88; Fig. 10). The d2H and d18O values in deep fresh and brack-ish groundwater are in places depleted relative to rainfall and shal-low groundwater (Fig. 11a).

4.4. Carbon-14 and tritium

The measured groundwater carbon-14 activities vary between5.5 and 106.9 pMC. Most shallow fresh and brackish groundwatersamples are within the range 60–107 pMC, indicating there is rel-atively recent recharge from both before and during the atmo-spheric nuclear testing period of the 1950s and 1960s. Thecarbon-14 activity in the confined aquifer is less than 6 pMC, whichis consistent with the palaeo-water signatures and residence timeestimates previously reported in groundwater from the North Chi-na Plain (Chen et al., 2003; Kreuzer et al., 2009). There is a veryobvious decline in 14C activities with increasing sampling depth(Fig. 12a), and towards the coastline from the recharge area. Apartfrom some mixing effect (observed when looking at radiocarbon

and TDS contents – Fig. 13), the data indicate that variations of14C activities are mainly due to radioactive decay and reflect theresidence time in aquifer. Age dating estimates made on the basisof geochemical correction (Han et al., 2011) can be seen in Fig. 11c.

The 3H concentration of most groundwater samples (Fig. 11b),expressed as 3H units (TU), varies from below detection to 11.8,with a wide range (1.8–15.3 TU) observed in shallow fresh andbrackish groundwater samples. The last recorded 3H concentrationin precipitation in the North China Plain was 16 TU, monitored atthe Shijiazhuang Station of IAEA in 2002 (IAEA/WMO, 2006). Tri-tium concentrations above �5 TU are therefore probably indicativeof recharge in the post-nuclear testing era. The low salinitygroundwaters mostly have therefore had relatively short residencetimes in the flow system and most were recharged within the lastfive decades (or at least contain a substantial mixing componentfrom this time period). In deep groundwater, there is a fairlynarrow range of tritium values from below detection up to3.5 TU – the upper value is an indication that some modern waterhas locally reached the deeper aquifer (probably as a result of in-duced leakage from pumping).

5. Discussion

5.1. Origin and recharge of the groundwater

The d18O vs. d2H values of the fresh, shallow groundwater aremostly scattered close to or slightly to the right of the LWML

Fig. 7. Evolution of hydrochemical ratios with increasing salinity, and correspondence to seawater and brine compositions. Symbols and mixing lines are same as in Fig. 7.

Fig. 8. Chloride versus bromide (a); and versus corresponding elements to chloride molar ratios (b) of groundwater from the south coastal area of Lazihou Bay. The dashedlines represents seawater Br/Cl ratio (�1.5 � 10�3). Br concentrations of groundwater in this region referenced from Han et al.(1996) and Han et al.(2011). Symbols are sameas in Fig. 7.

D.M. Han et al. / Journal of Hydrology 508 (2014) 12–27 21

(Fig. 10a), indicating that modern precipitation is their predom-inant origin. Stream water from Weihe River has d18O valuesranging from �7.6‰ to �4.6‰ with mean value of �5.8‰),and tritium of 5.1–7.8 TU (mean 6.5 TU) that are also similarto some of the shallow groundwater compositions, which mayindicate that these are further recharge sources locally. Thetrend towards 18O and 2H depletion in the deep fresh groundwa-ter (Fig. 10b) can be attributed to a colder climate during pastrecharge (e.g. during the last glacial maximum; cf. Kreuzer etal., 2009). The carbon-14 vs. d18O relationship observed(Fig. 14) with modern waters (>60 pMC) having relatively en-riched stable isotopic values (d18O > �6.0‰) and older waterbeing more isotopically depleted, is characteristic of therecharge history of northern China during the last �30 kyr(Kreuzer et al., 2009; Currell et al., 2012). The radiocarbon data

also suggest that the brine water formed during the Holocene, asthe corrected 14C ages are approximately 2.3–7.0 ka BP (notwith-standing effects of mixing – discussed below in Section 5.2). Thesubstantially lower stable isotopic values (e.g. d18O < �3.0‰) inthe brine water compared to modern seawater, present an inter-esting question as to the sources of salinity and water in thebrine, and a couple of hypotheses are explored further in Section5.3.

5.2. Mixing processes

5.2.1. Major ion indicationsBased on a mass-balance model, mixing lines between two

end-member samples for a range of hydrochemical indicators werecalculated and plotted based on:

Fig. 9. Saturation indices of groundwater samples with increasing selected anion concentrations.

Fig. 10. Stable isotope compositions of groundwater and rainfall in the study area.Symbols are same as in Fig. 7.

22 D.M. Han et al. / Journal of Hydrology 508 (2014) 12–27

Cm ¼ Cf � X þ Cb=s � ð1� XÞ ð1Þ

where X is the mixing fraction of fresh groundwater, Cm is the ionconcentration of mixture sample; Cf for the ion concentration offresh groundwater; Cb/s for the ion concentration of brine or seawa-ter (separate lines on each plot in Figs. 6 and 7). The relatively closefit between the majority of samples and the mixing lines (Fig. 6a, b,and d) implies predominantly conservative mixing processes be-tween freshwater and a saline end-member. The scatter observedaround mixing lines for fresh and brackish groundwater suggeststhat additional processes, such as water–rock modification (adsorp-tion–desorption, dissolution-precipitation, and oxidation–reductionprocesses), also control the chemical composition of the groundwa-ter to an extent (e.g. Appelo and Postma, 2005).

5.2.2. Isotopic indicationsWith increasing salinity, both the shallow and deep groundwa-

ter evolve towards more enriched d18O and d2H values. However, ingeneral, the groundwater does not show a tendency to evolve to-wards standard isotopic seawater (e.g. VSMOW); rather, samplesplot on a mixing trend between fresh groundwater with a meteoricsignature, and brine, which has similar d18O and d2H values of be-tween �3.0‰ to �4.0‰ and �30‰ to 40‰ in both shallow anddeep aquifers (Figs. 10 and 15). This mixing trajectory betweenfresh groundwater and the coastal brine is not obvious from exam-ining the mixing lines in the ionic data alone (Figs. 6 and 7), as theionic ratios of typical seawater and the brine are very similar(whereas the stable isotopic composition is notably different).

Accounting for the influence of mixing with brine, many of thefresh and brackish water samples appear to be derived from a com-mon ‘‘source composition’’ along the LMWL, at around �60‰ and�8‰ for d2H and d18O, respectively. This source composition isdepleted relative to the average modern precipitation value,

Fig. 11. Different tracers’ compositions of water in the south coastal area of Laizhou Bay. Explanation: SG – shallow groundwater; DG – deep groundwater; F – fresh; BA –brackish; SA – saline; BR – brine; M – modern water. The data of tritium and corrected 14C age in groundwater are referenced from Han et al.(2011) and Xue et al.(2000). Thed18O data of rainfall are from the GNIP database (IAEA/WMO, 2006), IAEA network, the average monthly (1986–1990) monitored at Yantai station (37�320240 0N, 121�240000 0)some 250 km NE of Changyi city.

Fig. 12. Carbon-14 activities (pMC) for the groundwater in south coastal aquifer ofLaizhou Bay. (a) Variations in percentage modern carbon (pMC) with depth. (b)Variations in pMC with distance from the coastline. Arrows show general trend inaquifer. Symbols are same as in Fig. 7.

Fig. 13. Carbon-14 activities vs. TDS concentrations of groundwater. The lines showpredicted mixing trends between brines and fresh water in shallow and deepaquifer. Data referenced from Han et al. (2011). Symbols are same as in Fig. 7.

Fig. 14. d18O vs. carbon-14 activities for shallow and deep groundwaters from thesouth coastal aquifers of Laizhou Bay. Symbols are same as in Fig. 7.

D.M. Han et al. / Journal of Hydrology 508 (2014) 12–27 23

suggesting either that preferential recharge takes place during thewettest season of the year (when isotopic value is depleted relativeto the average) and/or that deep groundwater contains palaeo-water with the depletion characteristic of the late Pleistocene agedgroundwaters of northern China. Additionally Fig. 13 shows radio-carbon activities vs. TDS concentrations for the groundwater in thisstudy area. The lines show two distinctive mixing trends betweenthe brines and fresh water in both the shallow and deep aquifers.

5.2.3. Mixing modelConcentrations of Cl together with d18O were further used to

examine the mixing trajectory and quantities for the different

waters (Fig. 15). The first obvious trend in this plot is that themajority of samples fall between fresh groundwater end members(with Cl concentrations of �2 meq/L and d18O values of ��6.0‰

and �9.0‰ for shallow and deep groundwater, respectively), andthe brines – with d18O values of ��3.0‰ and Cl concentrationsof �2000 meq/L. The relative contributions of freshwater and saltwater to the mixed samples was estimated using two end-mem-bers, based on the mass-balance mixing model (Eq. (1)), for bothCl and d18O. In general, the brine mixing proportions are estimated

Fig. 15. Relationship between chloride content and isotopic signature of surfaceand ground-water samples as a means to differentiate mixing processes in the area.Symbols are same as in Fig. 7.

Fig. 16. Distribution of cationic deltas of the groundwater samples with the changein distance from the coastline.

24 D.M. Han et al. / Journal of Hydrology 508 (2014) 12–27

to be between 5% and 30% of the mixed waters. The brine mixingprocess has important implications for assessing the extent of sea-water intrusion –a smaller mixing volume produces a much great-er salinisation effect in comparison to mixing with standardseawater.

The fresh-brine mixing pattern is adhered to strongly in thedeep samples, with increasing proportions of brine observed to-wards the coast, consistent with gradual mixing between the twoend-members along the regional S–N groundwater flow direction.The patterns in shallow groundwater are somewhat more com-plex; this likely reflects the added complexity in the water sourcesand processes in the shallow aquifer, e.g., inputs from modern pre-cipitation, surface water and leakage of irrigation water, plus evap-orative concentration from locally high water tables. Two shallowsaline groundwater samples also appear to fit a seawater-freshgroundwater mixing trajectory better than a brine-groundwatermix, suggesting classic seawater intrusion. The high values ofd18O observed in a number of shallow fresh groundwater samples,with relatively low Cl contents (Fig. 15) point to local evaporationprior to recharge where the water table is shallow, probably due toirrigation return flow.

5.3. Brine formation

The brine itself is characterized by both d18O and d2H values lo-cated below the LMWL; significantly lower than that of modern

seawater. This indicates that it either did not form due to simpleevaporative concentration of seawater and/or that it was subse-quently modified by mixing. The natural movement of groundwa-ter through the aquifer towards the coast, and the migration offresh, saline and brine water caused by intensive exploitation hasclearly resulted in mixing, which could have over-printed theinitial brine isotopic signatures, however, the extent of thisover-printing is now unclear. The current isotopic values may haveresulted from the mixture of originally more enriched brine andsubsequent mixing, or, the original brine may have been depletedrelative to standard seawater prior to the mixing (or a combinationof the two). One mechanism by which this could have occurred isthat the brine itself mixed with stream water (i.e. continental run-off) in Laizhou Bay during its formation (e.g. Dakin et al., 1983). It isinteresting to note that the stable isotopic values of the seawatercollected from Laizhou Bay in this study (Fig. 10 samples marked‘inshore seawater’) are more isotopically depleted than standardseawater – this may be indicative of a relatively large degree ofdilution of water in the bay with continental runoff; and may havebeen important in the brine formation process.

Another explanation is that there may have been an isotopedepletion effect occurring during brine formation by evaporation,this is known to occur during late stages of brine formation in hy-per-saline evaporating basins, due to reduction in the humiditycontrast between the boundary layer and atmosphere (Clark andFritz, 1997; Cartwright et al., 2009). Typically this effect occursonly at very high salinities, and so subsequent dilution (e.g. viasome mixing and displacement with regional groundwater flow)would have to have taken place to reach the observed brine com-positions. Elevated Br/Cl ratios (1.6 � 10�3–4.5 � 10�3, Han et al.,1996 and Han et al., 2011) in most of the saline and brine waterscompared to Br/Cl ratio (�1.5 � 10�3) in seawater (Davis et al.,1998) are consistent with an original brine that was more saline(e.g. at halite saturation), which subsequently was diluted by freshgroundwater or runoff (McCaffrey et al., 1987). The Br/Cl ratios inbrines (equal or greater than typical marine signatures) confirmthat evapotranspiration, rather than halite dissolution, was thedominant mechanism of brine formation. Based on the relationshipbetween the Br/Cl ratio in residual evaporated seawater and thedegree of evaporation (i.e., the ratio of Br dissolved in the residualbrine to its concentration in the initial seawater – cf. Vengosh andHendry, 2001), the maximum degree of evaporation of the brinesin the study area prior to any dilution with fresh water can be esti-mated at about 5.5 times.

5.4. Water–rock interaction

From the major ion and stable isotope data, it is clear that mix-ing between fresh groundwater and brine is the dominant controlon groundwater quality and hydrochemistry. However, the factthat ion ratios in many cases plot away from strict adherence tothe mixing lines (Fig. 7) indicate water–rock interaction (e.g. rockweathering, ion exchange) is also important, particularly in con-trolling the shallow and brackish groundwater chemistry. Satura-tion with respect to halite is not reached in any of the samples,precluding halite precipitation under current conditions – henceCl is assumed to behave relatively conservatively. An examinationof the ion/Cl ratios is therefore the primary means of detectingwater–rock interaction processes.

High Na/Cl ratios (>1.0) of most fresh and brackish groundwatersamples may be attributed to feldspar weathering (e.g.,2NaAlSi3O8 ðAlbiteÞ þ2CO2þ11H2O! 2Naþ þAl2Si2O5ðOHÞ4 ðKaoliniteÞþ4H4SiO4þ2HCO�3 ) and/or release of Na from exchange sites in theaquifer during freshening from palaeo seawater intrusion, whichhas not yet exhausted Na+ on the exchanger (Appelo and Postma,2005). Mg/Cl ratios in many of the brackish and fresh waters plot

D.M. Han et al. / Journal of Hydrology 508 (2014) 12–27 25

below the mixing line between the freshest water and seawater/brine, suggesting uptake on an exchanger. Weathering of K-feld-spar, Ca-bearing plagioclase and/or calcite could explain the rela-tively high K/Cl and Ca/Cl ratios of the freshest waters; while yetagain cation exchange may be important; e.g. an ‘intrusion signa-ture’ occurring due to recent migration of Na-rich water into fresh-water aquifers could explain the elevated Ca/Cl.

The wider variations of SO4/Cl ratios in fresh and brackishgroundwater than that in seawater and brine indicate both sulfateenrichment and depletion processes occurring. The decreasing SO4/Cl ratios with increasing salinity are likely caused by saline waterreaching saturation with respect to gypsum, rather than thede-sulfation of the aquifer sediments, which would be expectedto result in generation of acidity (e.g. Andersen et al., 2005). Somelocally elevated SO4/Cl ratios could relate to oxidation of sulfidicmaterial. High Ca/SO4 ratios in most fresh and brackish groundwa-ter samples would seem to rule out significant gypsum dissolution.In all cases, the ion/Cl ratios begin to approximate the predictedmixing lines more closely as salinity increases, due to other pro-cesses being overwhelmed by the total solute load.

To further determine the importance of water–rock interaction,the concentrations of cations were presented as a function of Clconcentration using ionic deltas (Pulido-Leboeuf, 2004). Fig. 16 de-picts the behavior of major cations with distance from the coast-line. The 4Na is usually positive, while the other values (4Ca,4Mg and 4K) are largely near zero, with no systematic trend.The relative excess of Na implies that during freshening of theaquifer, there is likely a direct cation-exchange of Ca and/or Mgin groundwater for Na in the clay layers present near the coastline.This process may be responsible for the Na–HCO3 type water ob-served near Liutuan (Fig. 4). It is notable that cation exchange ap-pears to be much more significant near the coast.

5.5. Anthropogenic influences on groundwater

The hydrochemical investigation by Xue et al.(2000) showedthat most groundwater in the fresh water zone in 1994 wasCa–HCO3 type. However, the current study shows that the ground-water in the fresh water zone has changed to Ca–HCO3�Cl type,indicating some influence of mixing with the more saline (andCl-rich) water. As noted earlier, increasing TDS values were alsoobserved in some of the monitoring wells between 2005 and2007 (Section 4.1). Shallow groundwater has also clearly been con-taminated by agricultural irrigation and is characterized by highNO3 concentrations (up to 346 mg/L in SG53) including withinthe fresh water zone (i.e. TDS < 1 g/L) south of Changyi city. Thiscompares with much lower concentrations observed before 2000(Xue et al., 2000). Relatively lower NO3 concentrations were foundin the shallow groundwater near the Wei River (31.1 mg/L inSG42), but overall concentrations are high in the shallow ground-water; this is a regional problem throughout the North China Plainarea (e.g. Chen, 2001), and indicates that irrigation return is now adominant recharge mechanism to the shallow aquifers (e.g. Currellet al., 2012). Locally high nitrate-enriched groundwater in the con-fined aquifer indicates the vertical leakage of modern high-nitratewater through the shallow formation into the underlying confinedaquifer. The high NO3 concentrations in groundwater have likelyresulted from the extensive use of nitrogen fertilizers in this area.

The middle layer (33–42 m) of brines were observed to have atritium concentration less than 1 TU in April 1994 (Xue et al.,2000), while this had reached 5.3 TU in July 2006 (Han et al.,2011). This is compelling evidence that the intensification of brineextraction has led to increased mixing between modern, shallowrecharge and the brine. The increased hydraulic gradients betweenshallow and deep layers may enhance the downward movement ofthe shallow groundwater into the deep aquifer. The wide range of

variation in tritium in the shallow fresh and brackish groundwatermay reflect the complexity of groundwater flowpaths and variationof travel time in the regional system, including the up-coning deepgroundwater caused by strong exploitation. Additionally, the leak-age of modern high-nitrate water through the shallow aquifer intothe underlying deeper aquifer suggests the presence of intercon-nected pathways vertically form the relatively rapid flow.

6. Conclusions and environmental implications

This paper has examined the processes involved in saltwaterintrusion into the aquifer below the Wei River plain, an exampleof complex groundwater dynamics and salinization mechanismsin the coastal plains of northern China – where groundwaterextraction is intensive. Physical and chemical parameters wereexamined as multiple lines of evidence to refine understandingof the evolution of groundwater salinization in the coastal plainaquifer of Laizhou Bay. The main findings are:

1. The continued decrease in groundwater levels in recent yearsdue to groundwater extraction for water supply and salt pro-duction is having noticeable water quality impacts in the aqui-fer, inducing mixing and salinization. This is most severe inareas of intensive extraction, for example, near the Changyigroundwater depression.

2. Mixing between fresh groundwater derived from meteoricsources and saline brines stored in the aquifer near the coastis supported by ion/chloride ratios but is particularly evidentwhen looking at stable isotope compositions, and radiocarbondata. This has important implications for the assessment of sal-ine intrusion in coastal aquifers. Firstly, analysis of mixing orthe extent of intrusion may not accurately determine salinitysources unless isotope data are examined in conjunction withtraditional hydrochemical analysis. Secondly, brines, whichcan occur in coastal systems due to a history of sea-level rise,fall and evaporative enrichment, can pose an equal or greaterwater quality threat than seawater, and cause more rapid sali-nization due to their higher salinity. Detailed analysis of coastalgroundwater geochemistry is therefore extremely important toassessing the risk posed by groundwater extraction in coastalareas, as the traditional conceptual model of a simple fresh/sea-water interface may not be adequate.

3. Ion/chloride ratios (e.g., Br/Cl and Na/Cl) in the system suggestan original evaporative source of salinity in the brines, whichwas intensive during formation, but was probably followed bydilution with meteoric water, resulting in d18O and d2H valueslower than seawater. The variation in chemical composition ofgroundwater demonstrates some complex (i.e. non-conserva-tive) hydrogeochemical processes, such as mineral weatheringand cation exchange that are more noticeable at low salinities.

4. The different stable and radio-isotopic values of shallow anddeep groundwaters provide a distinction between modern andpalaeo-recharge. The isotopic signature of groundwaterrecharged during the late Pleistocene differs markedly fromthat of the Holocene, characterized by depleted stable isotopiccompositions and low carbon-14 activities in the deepestwaters, furthest from the recharge areas.

5. Anthropogenic input into shallow groundwater is reflected bythe distribution of NO�3 concentrations, which can be attrib-uted to the general over-utilization of fertilizers in agricul-tural activities. Elevated and increasing levels of tritium,chloride and salinity in areas of intensive pumping supportthe hypothesis that induced mixing and water quality degra-dation have taken place over relatively short time scales (e.g.decades).

26 D.M. Han et al. / Journal of Hydrology 508 (2014) 12–27

At present, local government is adopting measures to try to re-duce local groundwater usage, such as transfer of water from theYellow River, officially forbidding or restricting local groundwaterpumping, and attempting to change from concentrated pumpingto more distributed pumping. Although it is very difficult forgroundwater depression cones to recover over short time periods,the water level data indicate that levels in some parts of the system– e.g. the unconfined aquifer near recharge sources, can locally re-cover relatively quickly, and thus these management measuresmay be able to slow the saline intrusion and even decrease the in-truded area. However, agricultural water consumption has untilnow been the main driver of water demand and the well-fieldsfor agricultural irrigation, will likely continue pumping into the fu-ture, providing a gradient favorable to inland migration of salinewater. If groundwater pumping for irrigation is not restricted, thenthe region around Changyi city will likely continue to be affectedby worsening saltwater intrusion (due to brine migration). Addi-tionally, effective management of the pumping of brine groundwa-ter may be able to restrict the movement of saltwater to the south.To some extent, the water divide existing between the southerncone of depression in the fresh water zone, and the northerndepression cone in the brine zone may restrict movement of waterbetween the two zones, i.e. forming a ‘double hydraulic barrier’. Ifno effective measures implemented in this area for relieving thesaltwater intrusion, the cones of depression will be connectedand develop into one regional cone, which could seriously threatenlocal sustainable development.

Acknowledgements

This research is financially supported by the National ScienceFoundation of China (No. 40801018). This work also forms partof the project of China Geology Survey, titled ‘‘Assessment of Vul-nerability and Investigation of Environmental Geology in the KeySection of Circum-Bohai-Sea Region’’, and we acknowledge thesupport of the China Geological Survey. We are thankful to Dr.Chen Zongyu for his constructive comments on the manuscript.We wish to thank Dr. Xie Hailan and Mr. Pan Tong from TianjinInstitute of Geology and Mineral Resources, for their help and sup-port during water sampling in the field and data collection.

References

Andersen, M.S., Jakobsen, V.N.R., Postma, D., 2005. Geochemical processes andsolute transport at the seawater/freshwater interface of a sandy aquifer.Geochimica and Cosmochimica Acta 69, 3979–3994.

Appelo, C.A.J., Postma, D., 2005. Geochemistry, Groundwater and Pollution, seconded. A.A. Balkema Publishers, Leiden, pp. 242–247.

Barker, A.P., Newton, R.J., Bottrell, S.H., 1998. Processes affecting groundwaterchemistry in a zone of saline intrusion into an urban aquifer. AppliedGeochemistry 13, 735–749.

Barlow, P.M., Reichard, E.G., 2010. Saltwater intrusion in coastal regions of NorthAmerica. Hydrogeology Journal 18, 247–260.

Bottomley, D., Gregoige, D.C., Raven, K.G., 1994. Saline groundwaters and brines inthe Canadian Shield: Geochemical and isotopic evidence for a residual evaporitebrine component. Geochimica et Cosmochimica Acta 58 (5), 1483–1498.

Carrera, J., Alcolea, A., Medina, A., Hidalgo, J., Slooten, L.J., 2005. Inverse problem inhydrogeology. Hydrogeology Journal 13, 206–222.

Cartwright, I., Hall, S., Tweed, S., Leblanc, M., 2009. Geochemical and isotopicconstraints on the interaction between saline lakes and groundwater insoutheast Australia. Hydrogeology Journal 17, 1991–2004.

Chen, H.H., Zhang, Y.X., Wang, X.M., Ren, Z.Y., Li, L., 1997. Salt-water intrusion in thelower reaches of the Weilhe River, Shandong Province, China. HydrogeologyJournal 5 (3), 82–88.

Chen, Z.Y., 2001. Groundwater Resources Evolution Based on PaleoenvironmentalInformation from Groundwater System in North China Plain. Ph.D. thesis, JilinUniversity, pp. 33–52 (in Chinese).

Chen, Z.Y., Qi, J.X., Xu, J.M., Xu, J.M., Ye, H., Nan, Y.J., 2003. Palaeoclimaticinterpretation of the past 30 ka from isotopic studies of the deep confinedaquifer of the North China plain. Applied Geochemistry 18, 997–1009.

Clark, I.D., Fritz, P., 1997. Environmental Isotopes in Hydrogeology. LewisPublishers, New York.

Craig, H., 1961. Standard for reporting concentration of deuterium and oxygen-18 innatural water. Science 133, 1833–1834.

Currell, M.J., Han, D.M., Chen, Z.Y., Cartwright, I., 2012. Sustainability ofgroundwater extraction in northern China: dependence on palaeowaters andeffects on water quality, quantity and ecosystem health. Hydrological Processes26, 4050–4066.

Dakin, R.A., Farvolden, R.N., Cherry, J.A., Fritz, P., 1983. Origin of dissolved solids ingroundwater of Mayne island, British Columbia, Canada. Hydrogeology Journal63, 233–270.

Davis, S.N., Whittemore, D.O., Fabryka-Martin, J., 1998. Use of chloride/bromide instudies of potable water. Ground Water 36, 338–351.

Edmunds, W.M., Ma, J.Z., Aeschbach-Hertig, W., Kipfer, R., Darbyshire, D.P.F., 2006.Groundwater recharge history and hydrogeochemical evolution in the MinqinBasin, North West China. Applied Geochemistry 21 (12), 2148–2170.

Feng, A.P., Gu, D.Q., Xia, D.X., 2006. Developments and causes of seawater intrusionin the south coast area of the Laizhou Bay. Coastal Engineering 25 (3), 7–13 (inChinese).

Ferguson, G., Gleeson, T., 2012. Vulnerability of coastal aquifers to groundwater useand climate change. Nature Climate Change 2, 342–345.

Fontes, J.C., Matray, J.M., 1993. Geochemistry and origin of formation brines fromthe Paris Basin, France. 1. Brines associated with Triassic salts. ChemicalGeology 109, 149–175.

Green, T.R., Taniguchi, M., Kooi, H., Gurdak, J.J., Allen, D.M., Hiscock, K.M., Treidel, H.,Aureli, A., 2011. Beneath the surface of global change: impacts of climatechange on groundwater. Journal of Hydrology 405, 532–560.

Han, Y.S., Meng, G.L., Wang, S.Q., 1996. Quaternary Underground Brine in theCoastal Areas of the Northern China. Science Press, Beijing, pp. 36–54 (Chapter4, in Chinese).

Han, D.M., Kohfahl, C., Song, X.F., Xiao, G.Q., Yang, J.L., 2011. Geochemical andisotopic evidence for Palaeo-Seawater intrusion into the south coast aquifer ofLaizhou Bay, China. Applied Geochemistry 26 (5), 863–883.

Horita, J., Gat, J.R., 1988. Procedure for the hydrogen isotope analysis of water fromconcentrated brines. Chemical Geology: Isotope Geoscience Section 72, 85–88.

Huang, M.Y., Zhi, T.Y.K., Liu, S.R., 1993. Comparison and analysis of the chemicalcharacteristics of precipitation in China and Japan. Chinese Journal ofAtmospheric Sciences 17 (1), 27–38 (in Chinese with English abstract).

IAEA/WMO, 2006. Global network of isotopes in precipitation. The GNIP Database.Vienna. <http://www-naweb.iaea.org/napc/ih/IHS_resources_gnip.html> (citedSeptember 2010).

Ji, M.C., 1991. Some changes of resources and environment caused by seawaterintrusion. Environmental Sciences 12 (5), 62–66 (in Chinese).

Jiang, W.M., 1992. Exploitation and utilization of brine water resources in the coastarea of Laizhou Bay. Resources Development and conservation 4, 305–307 (inChinese).

Jorgensen, N.O., Andersen, M.S., Engesgaard, P., 2008. Investigation of a dynamicseawater intrusion event using strontium isotopes (87Sr/86Sr). Journal ofHydrology 348, 257–269.

Kreuzer, A.M., Rohden, C.V., Friedrich, R., Chen, Z., Shi, J., Hajdas, I., Aeschbach-Hertig, W., 2009. A record of temperature and monsoon intensity over the past40 kyr from groundwater in the North China Plain. Chemical Geology 259, 168–180.

Landmeyer, J.E., Belval, D.L., 1996. Water-chemistry and chloride fluctuations in theUpper Floridan aquifer in the Port Royal Sound area, Beaufort and JasperCounties, South Carolina: US Geological Survey Water-Resources Investigations.Report 96–4102, 106p.

Li, D.G., Zhao, M.H., Han, M., Jiang, A.X., Zhang, Z.L., 2000. A study of the shallowly-buried paleochannel zones in the south coast plain of the Laizhou Bay. MarineGeology & Quaternary Geology 20 (1), 23–29 (In Chinese with English abstract).

Liu, S.M., Huang, W.W., Zhang, J., Wang, J.H., Ji, X.W., Wang, J.Y., 1993. Chemicalcomposition of meteoric precipitation in Qingdao area. Marine EnvironmentalScience 12 (3–4), 89–98 (in Chinese).

Liu, E.F., Zhang, Z.L., Shen, J., Song, J.F., 2004. Origin and characteristics of saltwaterintrusion disaster in the downstream of Wei River on the south coast of LaizhouBay. Journal of Earth Sciences and Environment 26 (3), 78–87 (in Chinese).

McCaffrey, M.A., Lazr, B., Holland, H.D., 1987. The evaporation path of sea water andthe coprecipitation of Br� and K+ with halite. Journal of Sedimentary Research57, 928–937.

Ning, J.S., 2004. The Research of Chemical Characters of the Underground Brinealong Laizhou Bay. Masters Ocean Univ., China. pp. 37–38 (in Chinese).

Parkhurst, D.L., Appelo, C.A.J., 1999. User’s guide to PHREEQC – a computer programfor speciation, reaction-path, 1D-transport, and inverse geochemicalcalculation. US Geological Survey Water Resources Investigations Report , 99–4259.

Peng, Z.Ch., Han, Y., Zhang, X., Huang, B.H., 1992. The study of the changes ofsedimental environments in the Laizhou Bay area. Geological Review 38 (4),360–367 (in Chinese).

Pulido-Leboeuf, P., 2004. Seawater intrusion and associated processes in a smallcoastal complex aquifer (Castell de Ferro, Spain). Applied Geochemistry 19,1517–1527.

Shi, M.Q., 2012. Spatial distribution of population in the low elevation coastal zoneand assessment on vulnerability of natural disaster in the coastal area of China.Master thesis of Shanghai Normal University, 24–32.

Stumm, W., Morgan, J., 1981. Aquatic Chemistry, second ed. Wiley, p. 780.TJR, 2006. Report of ‘‘Study of Real-time Monitoring and Warning on sea-salt water

intrusion in Laizhou Bay——A case study of Changyi-Liutuan Profile’’. TianjinInstitute of Geology and Mineral Resources, 37–38 (unpublished in Chinese).

D.M. Han et al. / Journal of Hydrology 508 (2014) 12–27 27

UN Atlas, 2010. UN Atlas: 44 Percent of us Live in Coastal Areas. <http://coastalchallenges.com/2010/01/31/un-atlas-60-of-us-live-in-the-coastal-areas/>.

Vengosh, A., Gill, J., Davisson, M.L., Hudson, B., 2002. A multi-isotope (B, Sr, O, H andC) and age dating (3H–3He and 14C) study of groundwater from Salinas Valley,California: Hydrochemistry, dynamics and contamination processes. WaterResources Research 38 (1), WR000517.

Vengosh, A., Hendry, M.J., 2001. Chloride-bromide-d11B systematics of a thick clay-rich aquitard system. Water Resources Research 37 (5), 1437–1444.

Vengosh, A., Kloppmann, W., Marei, A., Livshitz, Y., Gutierrez, A., Banna, M., Guerrot,C., Pankratov, I., Raanan, H., 2005. Sources of salinity and boron in the Gazastrip: Natural contaminant flow in the southern Mediterranean coastal aquifer.Water Resources Research 41, W01013.

Wang, J., Liu, M.G., Wang, J.H., Zhang, J., 1993. Chemical compositions in meteoricprecipitation and their constraints. Ocean Science 3, 39–43 (in Chinese).

Werner, A.D., Simmons, C.T., 2009. Impact of sea-level rise on sea water intrusion incoastal aquifers. Ground Water 47 (2), 197–204.

Xue, Y.Q., Wu, J.C., Ye, S.J., Zhang, Y.X., 2000. Hydrogeological andHydrogeochemical Studies for Salt Water Intrusion on the South Coast ofLaizhou Bay, China. Ground Water 38 (1), 38–45.

Yang, M., 2005. Study on the coastal zone environment degradation in the southerncoast of the Laizhou Bay and its control policy. Ph.D degree thesis of OceanUniversity of China. p. 4–8 (in Chinese).

Zhang, Y.X., Xue, Y.Q., Chen, H.H., 1996. Deposit Seawater Characteristics in theStrata and its Formation Environment in the South Coastal Plain of Laizhou Baysince late Pleistocene. Acta Oceanologica Sinica 18 (6), 61–68 (in Chinese).

Zhang, Y.X., Xue, Y.Q., Chen, H.H., 1997. Salt-brine water intrusion and its chemicalcharacteristics in Weifang area of southern coast of Laizhou Bay. Earth Science-Journal of China University of Geosciences 22 (1), 94–98 (in Chinese).

Zhang, Z.L., Peng, L.M., 1998. The groundwater hydrochemical characteristics onseawater intruded in eastern and southern coasts of Laizhou Bay. ChinaEnvironmental Science 18 (2), 121–125 (in Chinese).

Zhao, D.S., 1996. Research on Disaster Protection for Seawater Intrusion. ShandongPress of Sciences and Technology, Jinan, pp. 3–4 (in Chinese).