Urban land-use effects on groundwater phosphate distribution in a shallow aquifer, Nanfei River basin, China Jiazhong Qian & Lulu Wang & Hongbin Zhan & Zhou Chen Abstract Groundwater, surface water, soil and river sediment samples, and information on land use in the Nanfei River basin (NRB) of China have been analyzed to study the geochemistry, distribution, and mobilization of phosphorus. The distribution of phosphate (PO43–) and the relationships between PO43– and several constituents in groundwater were studied. Partial correlation analysis relating PO43– to types of land use was conducted using the data analyzing tool SPSS 15.0. The processes controlling the transport of PO43– are discussed. The conclusions from this study are: (1) urban land use has obvious impact on PO43– in groundwater, the average concentration of PO43– being 4.37mg/L, greater than that resulting from farmland and mixed land use, which have average PO43– concentrations of 0.10 and 0.18mg/L, respectively; (2) the partial correlation between PO43– and types of land use is significant with a coefficient of 0.760; (3) the PO43– concentrations in surface water are generally higher than those in groundwater, and the total phosphorus (TP) concentrations in river sediments are generally higher than those in soil samples; (4) groundwater is a carrier of PO43– and is likely responsible for the redistribution of PO43– in different regions of NRB.

Received: 6 May 2010 / Accepted: 16 July 2011 Published online: 6 August 2011 * Springer-Verlag 2011 J. Qian : L. Wang : Z. Chen School of Resources and Environmental Engineering, Hefei University of Technology, Hefei, Anhui Province 230009, People’s Republic of China e-mail: [email protected] H. Zhan ()) Department of Geology & Geophysics, Texas A&M University, College Station, TX 77843-3115, USA e-mail: [email protected] Tel.: +1-979-8627961 Fax: +1-979-8456162 H. Zhan Faculty of Engineering and School of Environmental Studies, China University of Geosciences, Wuhan, Hubei Province 430074, China Hydrogeology Journal (2011) 19: 1431–1442

Keywords Urban groundwater . Groundwater monitoring . Partial correlation analysis . Phosphate . China

Introduction Groundwater flow has played an important role in carrying nutrients to surface water. Excess input of nitrogen (N) and phosphorus (P) to surface water can lead to eutrophication which is a serious environmental concern (Belanger and Mikutel 1985; Kim et al. 2008; Lee et al. 2009). Some studies have shown that growth of algae in most of the eutrophic lakes is limited by the concentration of P rather than N (Schindler 1977; Kamaya et al. 2004). It is important to understand the pathways in which P may be delivered to such ecosystems by hydrological processes. Point-source P inputs are relatively easy to quantify and can be managed via legislation on P use in domestic products and/or by enhanced waste-water-treatment technology. Understanding non-point source or diffusive P inputs has proved to be more difficult. Very little attention has been paid to evaluating transport of P via groundwater due to the long-held belief that adsorption and metal-complex formation retain the majority of potentially mobile P (Sims et al. 1998; Holman et al. 2008). To date, the studies related to P in groundwater mainly fall into two categories. The first category is related to non-point source P from agricultural activity, with a focus on transfer of P from agricultural land to surface water via surface runoff and sub-surface discharge (Sawhney 1978; McDowell et al. 2001; Heathwaite et al. 2005; Oenema et al. 2005; Withers and Haygarth 2007). For instance, Jalali (2009) conducted a survey on P concentration, solubility and species in groundwater in a semi-arid basin, southern Malayer, western Iran. Jalali (2009) showed that large amounts of P fertilizer, inadequate management of P fertilizer use, and low irrigation efficiency, coupled with sandy soils in some parts of the study area, could be mainly responsible for the elevated P concentration in groundwater. The second category involves studying the transport mechanism of non-point source P based on hydrochemical analysis. For example, Kulabako et al. (2008) studied P attenuation and adsorption capacities of DOI 10.1007/s10040-011-0770-x

1432

soils in the laboratory and the field, and suggested that P dynamics was related to calcium (Ca), Iron (Fe) and organic carbon content of the soils. Carlyle and Hill (2001) studied groundwater P dynamics in a river riparian zone and showed that the sources of P associated with buried channel sediments could also influence subsurface reactive transport and release of P to streams. Holman et al. (2010) proposed that groundwater P contributions should not automatically be viewed as a source of dilution, but rather as having the potential to trigger and/ or maintain eutrophication, based on field studies in the UK and Ireland. To date, much less attention has been paid to investigating the urban land-use effects on groundwater P distribution and transport, which has been the primary focus of this study. The selected field site, Nanfei River basin (NRB), is located between Hefei City and Chaohu Lake (see Fig. 1) in China. Chaohu Lake is the fourth largest freshwater lake in China. Since the middle of 1980s, algae blooms have occurred each year in this lake (Zhu et al. 2006). It has been reported that Chaohu Lake has significant eutrophication as a result of the development of agriculture and industry near the lake (Jin et al. 2005). To reduce the eutrophication in Chaohu Lake, national and local governments have paid great attention to the control of point-source P inputs through regulating P use in domestic products and by enhancing waste-watertreatment technology. However, such efforts have not alleviated the problem of eutrophication in Chaohu Lake (Jin et al. 2005; Zhu et al. 2006). This has lead to the authors hypothesizing that non-point source P from groundwater discharged to Chaohu Lake, which has not been taken into account previously, could be considerably large and may be responsible for persistent eutrophication in the lake. This issue has, so far, not been studied at this site with the exception of a preliminary hydrochemical survey on groundwater and surface water in NRB by Qian et al. (2007). The objective of this study is to study the spatial distribution of PO43– in groundwater affected by land use and urbanization in NRB and the geochemical factors affecting the P mobilization and transport.

Description of the study area Geology and hydrogeology of the site Nanfei River basin is located in the Jianghuai hilly country, near Hefei City, the capital of Anhui Province of China, as shown in Fig.1. Nanfei River, as one of the largest rivers discharging into Chaohu Lake in this basin, originates from south of the Dabieshan Mountain at the center of Anhui Province of China. Nanfei River is 70 km long with a catchment area of 1,446 km2. The climate of the area is sub-tropic and humid. Nanfei River basin is an important industrial and agricultural base in Anhui Province. Groundwater in NRB is mainly recharged by rainfall and surface runoff. According to data collected at North Gate Hydrological Station of Hefei, the average depth to the water table at NRB is 11.56 m, compared to Hydrogeology Journal (2011) 19: 1431–1442

the water-table depth of 12.4~29.5 m at Hefei City. Fluctuations of the water table in NRB are closely related to rainfall events, and are partially related to groundwater withdrawals. Surface runoff is greatest during May to August annually, which is also the growth period of rice paddy in this region. Hilly monadnock and flood plains are the main landforms of this area. The hilly monadnock plain stretches across the northern and eastern parts of the study area with elevation about 15.0–28.0 m above sea level (a.s.l.) with geographic gradients about 3–5°. It is formed by long-term cutting of intermittent running water. The flood plain is along Nanfei River, and in the southern part of the study area near Chaohu Lake, with elevation about 6.5–9.5 masl. It is mainly formed by periodic flooding of Nanfei River and Chaohu Lake, and is mostly used for agriculture. A Quaternary stratum nearly covers the entire study area, as shown in Fig. 1. Late Archeozoic–early Proterozoic stratum is the basement at the study area. Various geological layers, including Cretaceous (K), Tertiary (E), Quaternary (Q) and Igneous rock (γ) can be found at the study site. There are two hydro-stratigraphic groups, namely the loose rock group and the clastic rock group. For the loose rock group, four types of hydro-stratigraphic units can be identified. One is the Quaternary pore water (Qhw) with a single-well yield of 10~100 m3/day, and a thickness of 4–30 m. The second is the fracture-pore water (Qp3x) with a single-well yield less than 10 m3/day, and a thickness of 10–25 m. The third is the carbonate fracturekarst water (Qp 1–2 g) with a single-well yield of 10~500 m3/day, and a thickness of 2–18 m. The fourth is the fracture water (E1–2dn) with a single-well yield of 10~100 m3/day, and a thickness of greater than 20 m. For the clastic rock group, the single-well yield is less than 100 m3/day. This stratigraphic group is divided into several units including mild sandy soil, muddy silty sand rock, mild clay, clay, muddy clay and sand (see Fig. 1). Decline of the water table at the study site is caused by pumping for irrigation and municipal usage. There is also close interaction between groundwater and surface water in the study area. Equipotential lines of water-table elevation are shown (as dashed lines) in Fig. 1 and groundwater generally flows from northwest to southeast before discharging to Nanfei River and Chaohu Lake (Qian et al. 2007).

Sampling and analysis To understand the chemical characteristics of groundwater and the fate and transport of P, groundwater samples were collected from 39 wells between Hefei City and Chaohu Lake in NRB. The geographical locations of the wells were determined with a GARMIN handheld global positioning system (GPS) and are shown in Fig. 1. The collection of groundwater samples was carried out from 5–10 December 2006. DOI 10.1007/s10040-011-0770-x

1433

N S1 S2 S3

Rive

8

11

S4

r

Qhw

S5

Qp3x

S6

Nan

14

17

A

fei

S7 Qhw

14

Qhw

11

Qp3x

8

S8

A’

(a) Elevation (masl)

A 40 30 20 10 0 -10 -20 -30 -40 -50

129 Qp3x Qp3x

Qhw

Qp3x

Qp1-2g

E1-2dn

E1-2dn

A’ 40 30 20 10 0 -10 -20 -30 -40 -50

(b)

Legend 1. Single-well water yield (m 3/d) <10 Holocene series

10-100 Holocene series

10-100 Early-mid Pleistocene series

100-500 Early-mid Pleistocene series

2. Stratigraphic unit Sediment sample Holocene alluvial Wuhu Group

Qp1-2g Early-mid Pleistocene Lacustrine Gunzihe Group Qp3x

Late Pleistocene Aeolian Xiashu Group

Mild sandy soil

Clay

Mild clay

Muddy clay

Surface water sample Soil sample Groundwater sample

Sand

Water-table isobath (masl) E1-2dn Dingyuan Group

Muddy silty sand rock

0

2.5

5 km

Fig. 1 a Location and hydrogeologic map of the study area, showing locations where groundwater, surface water, soil- and riversediments samples were collected. b Cross section A–A’

Hydrogeology Journal (2011) 19: 1431–1442

DOI 10.1007/s10040-011-0770-x

1434

It must be pointed out that due to the lack of hydrogeological infrastructure at the site, including piezometers (with narrow screen intervals), no point measurements of groundwater quality could be taken. Nearly all the groundwater sampling wells at the study site are open boreholes. The groundwater sampling depths range from 4.5 to 60.0 m below ground surface (see Table 1). In each groundwater sampling well, the sample is taken from the middle point between the water table and bottom of the wellbore. Be aware that although such a sample is taken from a point, it may better represent quality of mixed groundwater over a vertical distance between the water table and the bottom of the wellbore provided that water inside the open borehole is well mixed. Although the open borehole may be regarded as a disadvantage for groundwater sampling because it cannot yield precise point values of groundwater quality; it may serve the purpose of this study which is of more concern regarding regional distribution of P. The wells are located at various sites including urban residential

land, farmland and other mixed land use. The locations of the 39 groundwater wells are classified according to land-use types. This includes eight samples in the urban land, nine samples in the farmland and 22 samples in other mixed land use. In addition, 35 surface water samples, 54 soil samples and 8 river sediments samples were collected in the study area during 5–10 December 2006. Figures 1 and 2 show the locations of the sample collection for this study. The pH values of groundwater and surface-water samples were measured using portable pH meters in situ (Orion 290 A). The groundwater and surface-water samples were sealed in acid-washed polyethylene bottles in situ, and sent to the geochemical laboratory at Hefei University of Technology for analysis. All the water samples were collected and preserved following the standard procedure outlined in the groundwater collection and analysis manual published by the Ministry of Land and Resources of China (DZT 1993). Water samples were filtered using a 0.45-μm Millipore filter, then analyzed

Table 1 Chemical compositions of groundwater from shallow aquifers in NRB of China No.

pH

Na+ Ca2+ Mg2+ Cl– PO43– HCO3– Fe NO3– SO42– TDS Water type (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L)

Sampling Type of depth(m) land use

G1 G2 G3 G4 G5 G6 G7 G8 G9 G10 G11 G12 G13 G14 G15 G16 G17 G18 G19 G20 G21 G22 G23 G24 G25 G26 G27 G28 G29 G30 G31 G32 G33 G34 G35 G36 G37 G38 G39

7.40 7.49 7.40 7.16 7.44 8.42 7.42 7.53 7.21 7.53 7.93 7.13 7.77 7.90 7.99 7.32 7.84 7.86 8.38 7.68 7.87 7.54 8.09 7.95 7.72 8.00 7.45 7.45 7.23 7.23 7.28 7.91 7.72 7.43 7.50 7.26 7.07 7.76 7.55

73.0 160.9 124.1 202.6 149.9 125.8 140.7 42.5 163.5 95.3 85.7 54.7 95.7 96.0 104.3 56.3 35.7 424.0 83.3 31.9 113.4 128.0 59.9 536.5 101.4 80.3 41.3 95.6 67.3 132.3 125.1 170.4 137.0 122.6 82.5 151.4 90.8 65.8 166.7

24.7 14.0 11.0 10.0 20.0 60.0 16.0 12.0 8.0 20.0 40.0 15.0 40.0 15.0 22.0 22.0 35.0 55.0 25.0 18.9 28.0 7.9 6.0 30.0 50.0 50.0 40.0 10.0 20.0 36.0 7.2 36.0 30.0 12.6 10.0 38.0 24.0 4.5 38.0

66.9 120.9 154.6 115.4 58.1 29.6 120.9 90.5 145.8 110.5 42.1 92.6 56.1 92.7 94.1 73.7 55.3 31.8 53.2 64.8 76.9 68.9 70.3 62.7 51.5 50.8 54.1 101.8 84.2 85.2 106.7 77.2 117.4 134.5 69.9 149.7 54.3 58.5 77.8

30.1 52.7 79.6 53.4 20.8 15.3 47.2 24.8 61.4 50.8 22.7 39.4 27.9 34.1 28.1 23.9 21.8 19.7 18.6 16.6 33.0 29.4 21.9 26.4 23.9 23.6 18.9 34.5 26.0 35.7 54.6 32.9 38.1 50.1 32.5 60.7 26.2 20.5 28.6

60.7 123.8 120.3 210.9 50.4 22.9 154.7 26.4 133.0 135.2 13.8 47.0 18.3 149.0 151.3 88.3 35.5 60.7 55.0 33.2 105.4 114.6 75.6 95.1 22.9 16.1 6.9 107.7 68.8 57.3 73.4 114.6 144.4 241.8 25.2 105.4 21.8 11.5 56.2

3.50 0.29 1.27 11.63 2.77 0.09 0.34 0.04 9.65 0.14 0.13 0.00 0.01 0.22 0.15 0.16 0.18 0.04 0.16 0.17 0.56 0.440 0.34 0.12 0.09 0.11 0.25 0.05 0.09 0.12 0.36 0.18 3.38 2.43 0.23 0.10 0.04 0.03 0.04

352.1 594.8 758.9 437.6 376.0 399.9 372.6 311.1 451.2 338.4 413.6 376.0 512.8 153.8 222.2 307.7 218.8 461.5 256.4 262.5 396.9 418.2 232.0 492.2 451.2 423.9 348.7 345.3 280.3 413.6 583.1 417.0 471.7 358.9 417.0 683.7 444.4 403.4 454.6

0.33 0.07 0.01 0.04 0.03 0.02 0.04 0.07 0.05 0.01 0.02 0.01 0.03 0.03 0.03 0.19 0.03 0.09 0.02 0.65 0.03 0.01 0.04 0.07 0.05 0.05 0.179 0.040 0.017 0.529 0.037 0.022 0.014 0.012 0.544 0.021 0.259 0.060 1.518

3.68 10.20 14.91 161.85 0.99 1.98 166.79 46.42 216.36 8.42 1.86 11.07 <0.2 4.86 62.27 2.45 15.98 0.89 17.14 17.79 9.78 16.03 17.93 4.74 3.17 2.71 <0.2 78.97 7.77 <0.2 3.10 72.07 36.71 92.43 0.08 88.00 0.53 0.58 0.24

58.4 165.4 131.7 297.9 144.3 16.6 211.2 70.1 228.7 183.0 14.2 96.9 5.6 206.7 100.3 7.3 52.5 612.7 63.6 28.2 126.4 88.5 100.3 788.6 21.1 15.3 0.5 75.0 113.0 210.4 131.6 121.5 197.8 105.1 92.3 148.4 25.0 7.8 142.5

647 1278 1443 1650 802 621 1269 618 1449 973 603 721 729 740 746 576 445 1652 592 508 881 875 577 2115 717 666 526 907 710 965 1086 986 1200 1187 780 1432 688 586 899

Ca–Na–Mg–HCO3 Na–Ca–HCO3 Ca–Mg–Na–HCO3 Na–HCO3–SO4–Cl Na–Ca–HCO3–SO4 Na–HCO3 Na–Ca–HCO3–SO4–Cl Ca–HCO3 Na–Ca–HCO3 Ca–Na–Mg–HCO3–Cl–SO4 Na–Ca–HCO3 Ca–Mg–HCO3 Na–Ca–HCO3 Ca–Na–SO4–Cl Ca–Na–Cl–HCO3 Ca–Na–HCO3–Cl Ca–Mg–Na–HCO3 Na–SO4–HCO3 Na–Ca–HCO3 Ca–HCO3 Na–Ca–HCO3 Na–Ca–HCO3–Cl Ca–Na–HCO3–Cl–SO4 Na–SO4–HCO3 Na–Ca–HCO3 Na–Ca–HCO3 Ca–Na–Mg–HCO3 Ca–Na–HCO3–Cl Ca–Na–HCO3–SO4 Na–Ca–HCO3–SO4 Na–Ca–Mg–HCO3 Na–Ca–HCO3–Cl Na–Ca–HCO3–SO4–Cl Na–Ca–Cl–HCO3 Na–Ca–Mg–HCO3 Ca–Na–Mg–HCO3 Na–Ca–HCO3 Ca–Na–HCO3 Na–HCO3

3 3 3 3 3 1 1 1 3 1 1 1 2 1 1 2 1 1 1 1 1 1 1 1 2 2 1 1 1 2 1 2 3 3 1 2 2 1 1

Note: 3 urban residential land; 2 farmland; 1 mixed land use Hydrogeology Journal (2011) 19: 1431–1442

DOI 10.1007/s10040-011-0770-x

1435

Phosphate content in surface water (mg/L) (0.31,1] (1,2] (2,3]

Hefei

N

G1

G2

G4

City

9.08.0 7.0 6.0 5.0 G5 4.0

G3

3.0 2.0 1.0

G11

G12

G9

5.0 3.0

Guan Town G13

0

1 G15 .0

0.0

G10

(5,6]

6.0

G8

2.

(4,5]

G7

4.0

0.0 G6

(3,4]

Na

G16

nfe

G14

G24

G21

Surface water samples

Isoline of groundwater phosphate Main towm Study area boundary Nanfei River Land Use Mixed land use Urban residential land 0

5

G30 0.0

G32 0 1. G33 3.0 Dawei Town

G31

Legend Groundwater samples

0

G22

G25

G18

G23

G20

G26

0.0

G17

0.

i R iveG28 G27 r G29

G19

Farmland

G35

G34

2.

0

G36 G37 G38

G39

0.0

10 km

Fig. 2 The distribution of phosphate in groundwater and surface water, and land use at the study site

using ion chromatography (LC-10A, Shimadzu) to measure nitrate (NO3–), chloride (Cl–) and sulfate (SO42–) concentrations. Concentrations of sodium (Na+), calcium (Ca2+) and magnesium (Mg2+) were measured by inductively coupled plasma atomic emission spectrometry (ICPS-1000 Ш C, Shimadzu). Fe2+ and Fe3+ were analyzed with a flame atomic absorption spectrophotometer (AAS) (220FS, Varian, Palo Alto, CA). Bicarbonate (HCO3–) was determined by titration with hydrochloric acid (HCl) (Rowell 1994). Total dissolved solids (TDS) concentration was computed by multiplying the electrical conductivity (EC, dS/m) by a factor of 640 (Ayers and Westcot 1985). Concentrations of PO43– in groundwater and surface-water samples were determined with a Dionex DX-120 ion chromatograph. Ammonium concentrations were measured by using the Nessler’s method (Wei 2003). Soil samples were collected at two different depths in each borehole. The shallower depth is from 0.6 to 0.9 m and the deeper depth is from 1.6 to 1.9 m. Thus, the change of total phosphorus (TP) with sampling depth could be found. Each soil sample was collected by inserting a 10-cm-long polyvinyl chloride (PVC) cylinder with an inside diameter of 5 cm laterally into a designated soil horizon, and then placed in a sealed bag. Each soil Hydrogeology Journal (2011) 19: 1431–1442

sample was subsequently fully mixed until it appeared to be homogeneous. After that, 5 g (dry weight) of the sample was removed and analyzed using the molybdate blue method after dissolution in sulfuric acid and perchloric acid to measure the TP content (Liu 1996). Eight river sediments samples were collected from the upper 0~10 cm sediments at different locations: near Gujing Bridge (S1), Yangtze River Road Bridge (S2), Tunxi Road Bridge (S3), Shipyard (S4), Newport Dock (S5), Daxing Port (S6), the inlet of Twenty-li River (S7) and Sanli River (S8), which are shown in Fig 1. The river sediment samples were taken using a sediment core sampler, stored in sealed polyethylene bags, and transported within 8 h to an ice locker in which the temperature is kept at –10°C. The TP contents in the river sediment samples was measured in the same way as that for the soil samples, right after they were air-dried. It is worthwhile to point out that the PO 4 3– concentrations were measured for all the groundwater and surface-water samples, but only TP contents were measured for all the soil and river sediments samples. Partial correlation analyses were performed using Statistical Package for the Social Sciences (SPSS) version 15.0 (SPSS GmbH Software, Munich, Germany). DOI 10.1007/s10040-011-0770-x

1436

Results and discussions

80

Groundwater chemistry

60

Groundwater chemistry results are shown in Table 1, which includes the pH values, the concentrations of Na+, Ca2+, Mg2+, Cl–, PO43–, HCO3–, Fe (including Fe2+ and Fe3+), NO3–, SO42– and TDS, water-chemistry type, groundwater sampling depth, and types of land use. The pH values of these samples are slightly alkaline (7.07– 8.42). TDS of groundwater ranges greatly from 508 to 2,115 mg/L. The Ca2+ and Mg2+ concentrations in groundwater range from 29.56 to 154.60 mg/L and 15.32 to 61.42 mg/L, respectively. Ca2+ and Mg2+ probably come from dissolution of magnesium carbonate or dolomite. The Na+ concentration in groundwater is between 36.80 and 336.20 mg/L, with an average of 120.39 mg/L; while the Cl– concentration in groundwater is from 6.88 to 241.83 mg/L. The Na/Cl ratios for more than half of the groundwater samples are approximately equal to unity, an indication of halite dissolution. For groundwater samples whose Na/Cl ratios are greater than unity, the Na+ is typically interpreted as released from silicate weathering reaction (Meybeck 1987). If silicate weathering is indeed a source of Na+, the concentration of HCO3– should be higher than that of Na+. This is found to be true at the study site. Another source of Na+ is probably irrigation return flow. The Na+ concentration in groundwater may also increase due to evaporation. The concentration of HCO3– varies between 153.83 and 683.67 mg/L. Natural processes such as dissolution of carbonate minerals in the presence of atmospheric or soil carbon dioxide (CO2) and the dissolution of CO2 (Eq. 1) are probably responsible for the occurrence of HCO3–.  CaCO3 þ CO2 þ H2 O ¼ Ca2þ þ 2HCO 3 ; 2CO2 þ H2 O ¼ 2HCO3 :

ð1Þ The SO 4 2– concentration ranges from 0.18 to 788.60 mg/L, possibly derived from sulfide minerals such as gypsum, or from rainfall and fertilizers. The major ions of groundwater are plotted on a trilinear diagram in Fig. 3. Three types of land use are identified in the study area including urban residential land, farmland and mixed land use, which are represented in Fig. 3 using yellow circles, green diamonds and red square, respectively. It is shown in this figure that, overall, there are more alkaline-earth metal ions than alkaline metal ions; the total concentration of weak acids is higher than that of strong acids; and the carbonate hardness exceeds 50% in the study area. Meanwhile, two groundwater samples collected from urban residential land and two collected from mixedland-use areas mainly contain alkali and strong acid, and the two highest groundwater SO42– concentrations were both collected from mixed-land-use areas. Detailed compositions are shown in Table 1. A total of 20 chemical types of water are obtained from the samples, and the Hydrogeology Journal (2011) 19: 1431–1442

80 60

40

40

20

20

Mg

SO4

80

80

60

60

40

40

20

20 80

60

Ca Mixed land use

40

20

20 Na HCO 3

40

Urban residential land

60

80 Cl Farmland

Fig. 3 Piper trilinear diagram of the chemical type of groundwater at the study site

most frequently found is Na–Ca–HCO3. From Table 1, it is shown that Ca2+ and Mg2+ are the most abundant cations, while HCO3– is the most abundant anion.

Spatial distribution of PO43– and related ions in groundwater In order to look for the relationship between land-use and PO43– concentration in groundwater, a geographic information system (GIS) and the kriging interpolation method of MapGIS were used to exhibit the land-use types and the iso-concentration lines of groundwater PO43– (see Fig. 2). To compare the distribution of PO43– between surface water and groundwater, the sites of the surface-water samples have been marked along with the distribution of PO43– in surface water. Figure 2 shows that higher concentration of PO43– in groundwater mainly appeared in the area along the upstream of Nanfei River (Hefei City to Guan town) and near Dawei town. A similar pattern of PO43– distribution was also found in surface water. From the iso-concentration lines of PO43–, it can be concluded that the highest concentration of PO43– in groundwater is in the north, northeast and southwest of the study area, among which the PO43– concentration in the north and northeast area varies greatly. According to the Chinese National Standard for Drinking Water, less than 0.1 mg/L TP is classified as level III water, which cannot be used as drinking water directly. If one assumes that all P comes from PO43–, according to the mass ratio of P (atomic weight of 31) to PO43– (molecular weight of 95), there should be less than 0.31 mg/L (0.1×95/31=0.31) PO43– in level III water. In fact, besides PO43–, other P-related substances such as superphosphate may exist as well. Therefore, if the concentration of PO43– in a sample is found to be greater than 0.31 mg/L, one can assure that the TP content of this sample is equal to or above 0.1 mg/L, exceeding the limit of level III water. Through a comparison DOI 10.1007/s10040-011-0770-x

1437

PO43–

of concentrations and the land-use types shown in Table 2, the highest concentration of PO43– is found to be in the urban residential land (11.63 mg/L). Among the three types of land use, the average value of PO43– in the urban residential land is the highest (4.37 mg/L). The PO43– concentrations in seven out of eight samples (87.5%) from urban residential land exceed 0.31 mg/L. The samples collected from farmland have an average PO43– concentration of 0.10 mg/L, and none of these samples exceeds the PO43– concentration of 0.31 mg/L. The average PO43– concentration of samples collected from the mixed land use is 0.18 mg/L, and 17.4% of these samples have concentrations exceeding 0.31 mg/L. Figure 4 demonstrates the variation of groundwater PO43– with two different well depths and three land-use types, in which the top and bottom of each box represent the 25th and 75th percentiles, respectively, and the upper and lower boundaries of the bars represent the 10th and 90th percentiles; additionally, the middle thick horizontal black line is the median, and the black circle is an outlier. Groundwater in urban residential land at shallower well depth (<30 m) has the highest PO43– concentration; a large residential district is located near the area with the densest iso-concentration lines (Fig. 2), so anthropogenic activities including the use of P-containing products by the urban residents (such as laundry powder, detergents etc.) and the wastewater discharged from factories to ditches and the rivers are probably responsible for the high PO43– concentration in groundwater there. There are several different pathways for transporting P to groundwater in urban residential land. First, precipitation on the urban land may bring P to groundwater through vertical recharge. Second, wastewater from residential areas and factories is often discharged to rivers after different degrees of treatment. For instance, in the major city of Hefei, with a population about four million (see Fig. 2), wastewater is routed to a network of underground pipelines to the treatment plant before it is discharged into the Nanfei River. During this process, leakage of wastewater to the subsurface is inevitable because of broken or malfunctioning pipelines. Third, because of close interaction of groundwater and surface water, wastewater discharged into rivers may end in groundwater, particularly during the rainy seasons when water levels in rivers are higher than the water tables in the adjacent aquifers. Anthropogenic activities may also contribute to high concentration of PO43– in ground-

water in the southwest area because of heavily applied fertilizers in the farmland there. The PO43– concentrations and some relevant constituents such as Fe (including Fe2+ and Fe3+), Ca2+ and NO3– are plotted against bottom depths of wells in Fig. 5. This figure shows that the PO43– concentrations of most groundwater samples are less than 1.0 mg/L. A few higher PO43– concentrations (>1.0 mg/L) are observed at the wells whose bottom depths are less than 30.0 m, and two highest concentrations are at the depths of 8.0 and 10.0 m. Concentrations of Fe in groundwater are between 0.012 and 1.518 mg/L, and most of the samples except one have Fe concentrations below the Chinese National Standard for Drinking Water (0.3 mg/L), but the distribution of Fe is rather random without a clear trend (see Fig. 5b). The distributions of both Ca2+ and NO3– show the trend of decreasing with depth (see Fig. 5c and d). In order to identify the correlation between PO43– and Fe, Ca2+, NO3– concentrations in groundwater, 158 additional groundwater samples in the neighborhood of NRB within the same catchment were collected and analyzed together with the samples in NRB, and are represented in Fig. 6 (Xia 2007). The distribution of data is rather scattered but weak negative correlations between PO43– concentration and Fe, Ca2+, NO3– concentrations can be generally observed in Fig. 6a–d, indicating that one is more likely to find higher concentrations of Fe, Ca2+, NO3– when the concentration of PO43– is lower, or vice versa. The PO43– concentration appears not to be correlated with the pH value, as reflected in Fig. 6d. Boers et al. (1994) and Smolders et al. (1995) found that Fe could bind to PO43–, which might lead to lower P availability. In organic or peaty soils, organic matter may aid the transport of P together with organic acids and Fe and Al (Jalali 2009). Jalali (2009) pointed out that dissolved PO43– in calcareous soils would be expected to not only be adsorbed to minerals but interact with dissolved Ca2+ to form phosphate minerals, and this is well supported by Fig. 6b. N and P are well-known to be primarily responsible for the algal growth. Xu et al. (2008) found that phytoplankton and bacterial growth was actually P-limited in Pearl River which lies in the south of China, in late summer. The weak negative correlation between NO3– and PO43– is probably due to the P release and denitrification competitive relation (Chuang et al. 1996). A higher groundwater PO43– concentration is found to appear when the pH value ranges between 7.00 and 8.00 (Fig. 6d), and this indicates that a weakly alkaline

Table 2 Land use and corresponding phosphate concentration in groundwater Land use

Groundwater PO43– concentration (mg/L) Average Maximum Minimum

The distribution of groundwater samples by PO43– concentration <0.31 mg/L 0.31–5 mg/L >5 mg/L

Urban residential land Farmland Mixed land use

4.37

11.63

0.29

1(12.5%)

5(62.5%)

2(25%)

0.10 0.18

0.18 0.56

0.01 0.00

8(100%) 19(82.6%)

0 4(17.4%)

0 0

Hydrogeology Journal (2011) 19: 1431–1442

DOI 10.1007/s10040-011-0770-x

1438 12.00

10.00

PO4 3- (mg/L)

8.00

6.00

4.00

2.00

0.00 <30m

30-60m

<30m

Urban residential land

30-60m

<30m

30-60m

Mixed land use

Farmland

Fig. 4 Diagram showing the groundwater PO43– concentration variation with well-bottom depth and land-use types. The top and bottom of each box represent the 25th and 75th percentiles, respectively, whereas the upper and lower boundaries of the bars represent the 10th and 90th percentiles. The middle thick horizontal black line is the median

environment favors P release. This finding agrees with the experimental results of Jin et al. (2006) and also agrees PO4 3- (mg/L)

a

with several previous studies which show that increase of the pH value can free P from its binding with ferric complexes due to competition between hydroxyl ions and the bound P ions (Andersen 1974; Kim et al. 2003). Partial correlation analysis in SPSS 15.0 (GmbH Software, Munich, Germany) was selected to analyze the correlation relationship between groundwater PO43– contents and types of land use from 39 groundwater samples in Table 1. The statistical method used was the Mann– Whitney U-test which is a non-parametric test used to compare two independent groups of sampled data (Corder and Foreman 2009). The significant level of 0.05 was commonly used in the test (Corder and Foreman 2009). A higher partial correlation coefficient represents a closer correlation between two selected variables. Several ions such as NO3–, HCO3–, Cl–, Fe (Fe2++Fe3+), Ca2+, K+, Na+, Mg+2 and the depths of wells may influence the groundwater PO43– concentration, and were selected as control variables. The statistical analysis showed that the partial correlation between PO43– and types of land use was relatively significant with a correlation coefficient of 0.760.

Distribution of TP in soils and river sediments The TP of the soil samples collected from 27 locations and 8 river sediments in the study area were investigated.

b

Fe (mg/L) 0 0

10

10

20

20

Depth (m)

Depth (m)

-1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 0

30 40

20

40

60

30 40 50

50

60 60

70 70

c 0

50

100

150

-

NO3 (mg/L)

200

0

0

0

10

10

20

20

30 40

Depth (m)

Depth (m)

d

Ca2+ (mg/L)

100

200

300

30 40

50

50

60

60

70

70

Fig. 5 Vertical profiles of a PO43–; b Fe (including Fe2+ and Fe3+); c Ca2+; and d NO3– with well-bottom depth in groundwater samples at the study site Hydrogeology Journal (2011) 19: 1431–1442

DOI 10.1007/s10040-011-0770-x

1439

a

b

4.00

600.00

3.50

500.00

3.00

Ca 2+ (mg/L)

Fe (mg/L)

4.50

2.50 2.00 1.50 1.00

400.00 300.00 200.00 100.00

0.50 0.00

0.00 0.00

2.00

4.00

6.00

8.00

10.00

0.00

12.00

2.00

4.00

6.00

8.00

10.00

12.00

PO43- (mg/L)

PO43- (mg/L)

c

d 900.00

9.00 8.50

700.00 600.00 500.00

pH

NO 3- (mg/L)

800.00

400.00

8.00 7.50

300.00 200.00

7.00

100.00 0.00

6.50 0.00

2.00

4.00

6.00

8.00

10.00

12.00

0.00

2.00

4.00

6.00

8.00

10.00

12.00

14.00

PO43- (mg/L)

PO43- (mg/L)

Fig. 6 Variation of a Fe (including Fe2+ and Fe3+), b Ca2+, c NO3–, and d the pH value with PO43– in shallow groundwater in the study area. Red crosses represent the measured data collected by the authors during 5–10 December 2006; blue circles represent the data collected by Xia (2007) during the same period of time

As mentioned in the previous, the soil samples come from two distinctive depths for a given location: one is at 0.6~0.9 m (shallower), and the other is at 1.6~1.9 m (deeper), with a total of 54 samples collected. Figure 7 shows the comparison of TP at these two depths. By comparing TP at different depths for a given land-use type, one can see that TPs at shallower depths have a wider range of variation than that at deeper depths.

600.0

TP (mg/kg)

500.0

400.0

300.0

200.0

100.0 shallow part

deep part

Urban residential land

shallow part

deep part shallow part deep part

Farmland

Mixed land use

Fig. 7 Diagrams showing the soil total phosphorus (TP) contents variation with different sampling depths and land-use types Hydrogeology Journal (2011) 19: 1431–1442

It is very interesting to compare the results of Figs. 4 and 7. From Fig. 7, one can see that for samples collected at a given depth (shallow or deep) but with different landuse types, the distributions of TP in the soil samples are more or less similar to each other. However, one can see from Fig. 4 that there are considerably higher PO43– concentrations in urban land than those in farmland and mixed land use. This suggests that the urban land has received considerable contribution of P from other sources in addition to soil P than the farmland and mixed land use. Those sources of P for the urban land are most likely from anthropogenic activities such as the use of P-containing laundry powder or detergent by the urban residents and the discharge of treated municipal wastewater or some illegal untreated wastewater from various sources. The PO43– concentrations in groundwater and surfacewater samples and the TP contents in soil and river sediment samples are compared in Fig. 8 (where black circles represent outliers). Note that Fig. 8 is not used to compare the TP contents in soil and river sediments with the PO43– concentrations in groundwater and surface water. Instead, it is used to compare the PO43– concentrations in groundwater and surface water (see the unit of mg/L on the left vertical axis) and the TP contents in soil and river sediments (see the unit of mg/kg on the right vertical axis). In Fig. 8, the median PO43– in surface-water samples is greater than that in groundwater samples. This may be explained from two perspectives. First, surface runoff DOI 10.1007/s10040-011-0770-x

30

3000

25

2500

20

2000

15

1500

10

1000

5

500

0

0 Groundwater

Surface water

Soil

TP (mg/kg) (Soil and Sediment)

PO4 3- (mg/L)

(Groundwater and Surface water)

1440

Sediment

Fig. 8 Diagram showing the concentrations of PO43– in groundwater and surface water, and TP in river sediments and soil, in the study area

carrying P-rich wastewater discharged from the industrial factories and the urban sewage generally flows to the rivers and other surface-water bodies, resulting in accumulation of PO43– in surface water. Second, groundwater in the study area generally discharges to various surface water bodies in dry seasons; thus, PO43– carried by groundwater generally accumulates in the surface water as well. These two phenomena will lead to higher PO43– concentrations in surface water than those in groundwater. The PO43– concentrations in surface water range from 0.3 to 11.87 mg/L with an average of 2.44 mg/L, while they vary between detection limit and 11.63 mg/L with an average of 1.02 mg/L in groundwater samples. The TP contents in river sediment samples are found to be much higher than those in soil samples. The TP contents in the river sediment samples vary from 760 to 2,420 mg/kg with an average of 1,909 mg/kg, whereas the TP contents in the soil samples vary from 183 to 576 mg/L with an average of 326 mg/L. It seems that P in river sediments is also linked to anthropogenic activities. Many years of sewage disposal will lead to accumulation of chemicals including P in the river sediments. Experimental results show that river sediments can release chemical ions including P to overlying water.

Implications for P mobilization The distribution and hydrochemical characteristics of groundwater PO43– in NRB reveal that P released from soil and river sediments is probably another source of PO43– for groundwater and surface water (Macrae et al. 2005). The weakly alkaline and anaerobic environment provides a favorable condition for the release of P from river sediments, as demonstrated experimentally by Kim et al. (2003) who showed that P released from river sediments was a function of pH and dissolved oxygen concentration. The groundwater environment at the study site is also favorable for P desorption. Gjettermann et al. (2007) found that the dissolved organic P increased strongly with increasing pH values. Kim et al. (2004) indicated that P tended to be Hydrogeology Journal (2011) 19: 1431–1442

released from river sediments when the P concentration of river water was less than 1.4 mg/L. Previous studies show that a high N content may restrict the P release (Chuang et al. 1996; Roy et al. 1986). However, Chuang et al. (1996) found that simultaneous P release and denitrification were two competitive kinetic processes under anaerobic conditions in the presence of available organic substrate (acetate). Moreover, the sludge with a high P content has a higher P release rate in accordance with a lower denitrification rate. As can be seen from the equipotential lines in Fig. 1, groundwater in the study area is generally discharged to Chaohu Lake and Nanfei River. The transport of P associated with such a groundwater flow scheme may be a potential contributor to the eutrophication of Chaohu Lake, which agrees with studies by Correll (1998) and Reynolds and Davies (2001) at other sites. However, a precise quantification of the amount of P transported to the Chaohu Lake from groundwater discharge is only possible after a detailed numerical model of P transport can be established. This is a subject that needs further investigation. Comparing Figs. 1 and 3, one can see that the mobilization of groundwater PO43– is closely related to the interaction between groundwater and surface water. From Fig. 3, there is a higher PO43– zone in groundwater along the Nanfei River close to Guan town, while there is a relatively higher PO43– concentration in surface water along Nanfei River in the same area. That is partially because the groundwater was recharged by the Nanfei River in the up-stream section, whereas the Nafei River was recharged by groundwater through base flow in the down-stream section (see Fig. 1).

Future study Due to the limits of time and funding, this study still lacks a few important data sets which will be collected in a future study. For instance, there are no data on the PO43– carried by the suspended river sediment, which is also a potential source of PO43–. If the suspended river sediment is proven to be an important source of PO43–, one is likely to observe higher PO43– concentration during storm events when more sediment is suspended. Another interesting research project that would be helpful is to observe the PO43– concentration in surface water and groundwater along a cross-sectional view of a hydrostratigraphic unit. In this way, one can gain more insight into the transport process controlling the PO43– interaction between surface water and groundwater. It would also be interesting to collect data for a potentially conservative ion (such as Cl–) simultaneously with the PO43– data along a cross-sectional view of a hydro-stratigraphic unit. By doing so, one can have a better understanding of the role of dilution on both non-reactive and reactive ions. One issue that has not been addressed quantitatively in this study is the relative contributions of PO43– to Chaohu Lake through the surface water versus the groundwater discharges. This issue will be pursued in the future as well. DOI 10.1007/s10040-011-0770-x

1441

Conclusions This report presents the first systematic hydrochemical survey of the area between Hefei City and Chaohu Lake in NRB of China. The compositional characteristics of groundwater are analyzed and the distribution of PO43– and the relationships between PO43– and several constituents in groundwater are studied. The processes related to the transport of PO43– are also discussed. The main conclusions can be drawn as follows: 1. The chemical composition of groundwater in NRB varies greatly and can be classified into 20 chemical types, with the most frequently observed being Na–Ca– HCO3. The pH values of groundwater in NRB show it to be slightly alkaline (7.07–8.42). 2. The concentrations of PO43– in groundwater vary greatly with land use. Of these land uses, the urban land plays an important role in contributing PO43– to groundwater and has the highest average PO43– concentration. Anthropogenic activities, including the use of P-containing products by the urban residents and the treated and untreated wastewater discharged from factories and other sources, are probably responsible for the high PO43– concentration in groundwater there. Anthropogenic activities may also contribute to high concentration of PO43– in groundwater in the southwest of the study area because of applied fertilizers in the farmland there. 3. The partial correlation between PO43– and types of land use is relatively significant with a correlation coefficient of 0.760. Somewhat weak negative correlations are found between PO43– in groundwater and several typical constituents such as Fe (including Fe2+ and Fe3+), Ca2+ and NO3– in groundwater. It shows that an environment with lower concentrations of Fe, Ca2+, NO3– favors PO43– accumulation in groundwater. Higher groundwater PO43– concentration is found to appear when the pH value ranges between 7.00 and 8.00, but a general correlation between the PO43– concentration and the pH value is not identified. 4. The PO43– concentration in surface water is generally higher than that in groundwater at the study site. The TP contents in river sediments are generally higher than those in soil samples. 5. In general, groundwater flows from the northwest to the southeast in the study area, discharging to the Nanfei River and the Chaohu Lake. The transport of P associated with groundwater flow may be a potential contributor to the eutrophication of Chaohu Lake. Quantification of the amount of P transported to the Chaohu Lake from groundwater discharge is a subject that needs further investigation.

Acknowledgements The research was supported by The Funds for Creative Research Groups of Hefei University of Technology (No. 2009HGCX0233), National Natural Science Foundation of China (No. 40872166) and National “Water Special” Program of China (No. 2008ZX07103-003-01). We thank the reviewers and the editor Hydrogeology Journal (2011) 19: 1431–1442

for their constructive comments which helped us improve the manuscript substantially.

References Andersen JM (1974) Nitrogen and phosphorus budgets and the role of sediments in six shallow Danish Lakes. Arch Hydrobiol 74:528–550 Ayers RS, Westcot DW (1985) Water quality of agriculture. FAO Irrigation and Drainage Paper 29, Rev. 1. Rome, pp 29–31 Belanger TV, Mikutel DF (1985) On the use of seepage meters to estimate groundwater nutrient loading to lakes. J Am Water Resour Assoc 21(2):265–272 Boers P, Van der Does J, Quaak M, Van der Vlugt J (1994) Phosphorus fixation with iron (III) chloride: a new method to combat internal phosphorus loading in shallow lakes? Arch Hydrobiol 129(3):339–351 Carlyle GC, Hill AR (2001) Groundwater phosphate dynamics in a river riparian zone: effects of hydrologic flowpaths, lithology and redox chemistry. J Hydrol 247(3–4):151–168 Chuang SH, Ouyang CF, Wang YB (1996) Kinetic competition between phosphorus release and denitrification on sludge under anoxic condition. Water Res 30(12):2961–2968 Corder GW, Foreman DL (2009) Nonparametric statistics for nonstatisticians: a step-by-step approach. Wiley, Hoboken, NJ Correll DL (1998) The role of phosphorus in the eutrophication of receiving waters: a review. J Environ Qual 27:261–266 DZT (1993) Collection and preservation methods of water samples for groundwater test (in Chinese). DZT 0064.2-1993, Ministry of Land and Resources of China, Beijing, pp 1–10 Gjettermann B, Styczen M, Hansen S, Borggaard OK, Hansen HCB (2007) Sorption and fractionation of dissolved organic matter and associated phosphorus in agricultural soil. J Environ Qual 36(3):753–763 Heathwaite AL, Dils RM, Liu S, Carvalho L, Brazier RE, Pope L, Hughes M, Phillips G, May L (2005) A tiered risk-based approach for predicting diffuse and point source phosphorus losses in agricultural areas. Sci Total Environ 344(1–3):225–239 Holman IP, Whelan MJ, Howden NJK, Bellamy PH, Willby NJ, Rivas-Casado M, McConvey P (2008) Phosphorus in groundwater: an overlooked contributor to eutrophication? Hydrol Process 22(26):5121–5127 Holman IP, Howden NJK, Bellamy PH, Willby NJ, Whelan MJ, Rivas-Casado M (2010) An assessment of the risk to surface water ecosystems of groundwater P in the UK and Ireland. Sci Total Environ 408(8):1847–1857 Jalali M (2009) Phosphorous concentration, solubility and species in the groundwater in a semi-arid basin, southern Malayer, western Iran. Environ Geol 57(5):1011–1020 Jin X, Xu Q, Huang C (2005) Current status and future tendency of lake eutrophication in China. Sci China C Life Sci 48(12):948– 954 Jin X, Wang S, Pang Y, Wu F (2006) Phosphorus fractions and the effect of pH on the phosphorus release of the sediments from different trophic areas in Taihu Lake, China. Environ Pollut 139 (2):288–295 Kamaya Y, Takada T, Suzuki K (2004) Effect of medium phosphate levels on the sensitivity of Selenastrum capricornutum to chemicals. Bull Environ Contam Toxicol 73 (6):995–1000 Kim LH, Choi E, Stenstrom MK (2003) Sediment characteristics, phosphorus types and phosphorus release rates between river and lake sediments. Chemosphere 50(1):53–61 Kim LH, Choi E, Gil KI, Stenstrom MK (2004) Phosphorus release rates from sediments and pollutant characteristics in Han River, Seoul, Korea. Sci Total Environ 321(1–3):115–125 Kim G, Ryu JW, Hwang DW (2008) Radium tracing of submarine groundwater discharge (SGD) and associated nutrient fluxes in a highly-permeable bed coastal zone, Korea. Mar Chem 109(3– 4):307–317 DOI 10.1007/s10040-011-0770-x

1442 Kulabako NR, Nalubega M, Thunvik R (2008) Phosphorus transport in shallow groundwater in peri-urban Kampala, Uganda: results from field and laboratory measurements. Environ Geol 53(7):1535–1551 Lee YW, Hwang DW, Kim G, Lee WC, Oh HT (2009) Nutrient inputs from submarine groundwater discharge (SGD) in Masan Bay, an embayment surrounded by heavily industrialized cities, Korea. Sci Total Environ 407(9):3181–3188 Liu GS (1996) Soil physical, chemical analysis and description of soil profiles (in Chinese). Standards Press of China, Beijing, pp 33–37 Macrae ML, Redding TE, Creed IF, Bell WR, Devito KJ (2005) Soil, surface water and ground water phosphorus relationships in a partially harvested Boreal Plain aspen catchment. For Ecol Manage 206(1–3):315–329 McDowell R, Sharpley A, Brookes P, Poulton P (2001) Relationship between soil test phosphorus and phosphorus release to solution. Soil Sci 166(2):137–149 Meybeck M (1987) Global chemical weathering of surficial rocks estimated from river dissolved loads. Am J Sci 287(5):401–428 Oenema O, van Liere L, Schoumans O (2005) Effects of lowering nitrogen and phosphorus surpluses in agriculture on the quality of groundwater and surface water in the Netherlands. J Hydrol 304(1–4):289–301 Qian JZ, Zhao WD, Hong TQ, Liu Y, Tang CY (2007) Spatial variability in hydrochemistry of groundwater and surface water: a case study in Nanfei River catchment, China. In: Bullen DT, Wang Y (eds) Water–rock interaction, vols 1 and 2, proceedings. Taylor and Francis, London, pp887–890 Reynolds CS, Davies PS (2001) Sources and bioavailability of phosphorus fractions in freshwaters: a British perspective. Biol Rev 76(1):27–64

Hydrogeology Journal (2011) 19: 1431–1442

Rowell DL (1994) Soil science: methods and applications. Longman, Harlow, UK; Wiley, New York Roy WR, Hassett JJ, Griffin RA (1986) Competitive coefficients for the adsorption of arsenate, molybdate, and phosphate mixtures by soils. Soil Sci Soc Am J 50(5):1176–1182 Sawhney BL (1978) Leaching of phosphorus from agricultural soils to groundwater. Water Air Soil Pollut 9(4):499–505 Schindler DW (1977) Evolution of phosphorus limitation in lakes. Science 195:260–262 Sims JT, Simard RR, Joern BC (1998) Phosphorus loss in agricultural drainage: historical perspective and current research. J Environ Qual 27:277–293 Smolders AJP, Nijboer RC, Roelofs JGM (1995) Prevention of sulphide accumulation and phosphate mobilization by the addition of iron (II) chloride to a reduced sediment: an enclosure experiment. Freshw Biol 34(3):559–568 Wei F (2003) Method of Analyzing and Monitoring the Water and Wastewater, 4th edn., State Environmental Protection Administration of China. China Environmental Science Press. Beijing, pp 276–280 Withers PJA, Haygarth PM (2007) Agriculture, phosphorus and eutrophication: a European perspective. Soil Use Manage 23:1–4 Xia Q (2007) Hydrochemical Characteristics and Experimental Study of Nitrate Removal in New Lakeshore District of Hefei City. MSc Thesis, Hefei University of Technology, China, pp 1–72 Xu J, Yin K, He L, Yuan X, Ho AYT, Harrison PJ (2008) Phosphorus limitation in the northern South China Sea during late summer: influence of the Pearl River. Deep-Sea Res 55 (10):1330–1342 Zhu GW, Qin BQ, Zhang L, Luo LC (2006) Geochemical forms of phosphorus in sediments of three large, shallow lakes of China. Pedosphere 16(6):726–734

DOI 10.1007/s10040-011-0770-x