Journal of Contaminant Hydrology 182 (2015) 221–230

Contents lists available at ScienceDirect

Journal of Contaminant Hydrology

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

Driving mechanism and sources of groundwater nitrate

contamination in the rapidly urbanized region of south China
Qianqian Zhang ⁎, Jichao Sun, Jingtao Liu, Guanxing Huang, Chuan Lu, Yuxi Zhang
Institute of Hydrogeology and Environmental Geology, Chinese Academy of Geological Sciences, Shijiazhuang 050061, China

a r t i c l e i n f o a b s t r a c t

Article history: Nitrate contamination of groundwater has become an environmental problem of widespread
Received 14 March 2015 concern in China. We collected 899 groundwater samples from a rapidly urbanized area, in order
Received in revised form 15 September 2015 to identify the main sources and driving mechanisms of groundwater nitrate contamination. The
Accepted 17 September 2015 results showed that the land use has a significant effect on groundwater nitrate concentration
Available online 21 September 2015
(P b 0.001). Landfill leakage was an important source of nitrate in groundwater in the PRD (Pearl
River Delta) region, since landfill yielded the highest nitrate concentration (38.14 mg/L) and the
Keywords: highest ratio of exceeded standard (42.50%). In this study, the driving mechanism of groundwater
Nitrate contamination nitrate contamination was determined to be urban construction and the secondary and tertiary
Driving force
industrial development, and population growth. This study revealed that domestic wastewater
Urbanization
and industrial wastewater were the main sources of groundwater nitrate pollution. Therefore,
Socioeconomic development
Driving mechanism the priority method for relieving groundwater nitrate contamination is to control the random
discharge of domestic and industrial wastewater in regions undergoing rapid urbanization.
Capsule abstract.
The main driving mechanism of groundwater nitrate contamination was determined to be urban
construction and the secondary and tertiary industrial development, and population growth.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction Pastén-Zapata et al., 2014) and cancer through the formation of

Nitrate (NO− 3 ) is a primary contributor to groundwater
pollution, due to its stability, high solubility and mobility.
Nowadays, NO− 3 contamination in groundwater has become a
widely concerned environmental problem (Xue et al., 2009).
Several studies have shown that increased levels of NO− 3 in
groundwater are primarily caused by anthropogenic activities
such as the overuse of nitrogen fertilizers and animal manures,
the discharge of domestic and industrial sewage, and the
elevated atmospheric N deposition (Dubrovsky et al., 2010;
Gu et al., 2012; Kaushal et al., 2011; Xue et al., 2009).
Groundwater NO− 3 contamination is a threat to human
health (Gu et al., 2013). Drinking water containing elevated
levels of nitrate has been associated with the risk of methemo-
globinemia or ‘blue baby syndrome’ (Fan and Steinberg, 1996;
⁎ Corresponding author.
URL: z_qqian@163.com (Q. Zhang).
carcinogenic N-nitroso compounds (Weyer et al., 2001). In
addition, the export of nitrate into surface water may lead to
many ecological and environmental problems, including eutro-
phication and seasonal hypoxia (Li et al., 2010; Li et al., 2013). As
a result of these issues, the World Health Organization (WHO)
has set an upper limit of 50 mg/L for drinking water (WHO,
2011).
Nitrate concentrations above the WHO's maximum con-
tamination level are relatively common in some regions,
especially in the emerging developing countries (Burow et al.,
2010; Gu et al., 2013). In China, the occurrence of serious
groundwater nitrate pollution has accompanied the fast
socioeconomic development in the country (Fu et al., 2007).
According to the Ministry of Land and Resources, 90% of China's
shallow groundwater is polluted, and nitrate being one of the
main pollutants (Qiu, 2011).
The key step to effectively control nitrate pollution is to
identify the factors affecting groundwater nitrate contamination.

http://dx.doi.org/10.1016/j.jconhyd.2015.09.009
0169-7722/© 2015 Elsevier B.V. All rights reserved.
222 Q. Zhang et al. / Journal of Contaminant Hydrology 182 (2015) 221–230
Several studies have examined nitrate contamination resulting
from different agricultural practices and hydrogeological settings
(Güler et al., 2012; Landon et al., 2011; Stigter et al., 2008).
Lockhart et al. (2013) described the distribution of nitrate
resulting from different land uses. Rankinen et al. (2008)
evaluated the leaching of nitrate into shallow groundwater
from a variety of agricultural fields under different cultivating
practices. Researchers also found that the manure management
(Lord et al., 2002), reduction–oxidation conditions (Landon et al.,
2011), soil texture (Wick et al., 2012), atmospheric temperature
(Wick et al., 2012), and precipitation surpluses (Salo and Turtola,
2006) influence the concentration of nitrate leached into
groundwater. However, we do not fully understand the relation
between groundwater nitrate concentration elevation and the
rapid urbanization and economic development (Gu et al., 2013),
which lead to a serious impact on management of groundwater
nitrate pollution.
One major region of China that has recently seen accelerated
urbanization is the Pearl River Delta (PRD), which is situated in
the southern part of Guangdong Province. Economic growth and
urbanization have accelerated in the last two decades around the
Pearl River Delta, as a result of the drastic economic reform
initiated in the early 1980s (Mai et al., 2002). The water supply
of this region relies almost entirely on surface water due to
abundant rainfall and the well-developed river network. How-
ever, rapid population and economic growth have led to
deterioration of the quality of surface water through pollution
and declining river discharge (Cheung et al., 2003). Therefore,
surface water alone can no longer meet the needs of the region,
and people are looking to groundwater as an alternative supply.
With the increasing exploitation of the groundwater in the PRD,
it has been found to be contaminated with total dissolved solids
(TDS) of over 10 g/L, seawater intrusion, As contamination, etc.
(Huang et al., 2013; Wang et al., 2012). However, the study of
groundwater nitrate pollution in this region has not been
reported.
This study expands on previous work using a large area and
sample size across wider land uses. The main aims of this study
are 1) to analyze the relationship between the groundwater
nitrate levels and socioeconomic development parameters by
using regressions analysis (RA), 2) to identify the main driving
mechanism of groundwater nitrate contamination in a rapidly
urbanized area by using principal component analysis (PCA),
and 3) to identify the most likely sources of nitrate by using
water chemistry theory.
2. Materials and methods
2.1. Description of study area
2.1.1. Geographical conditions
The PRD covers an area of 41,600 km2 in south China,
extending between longitudes of 111°59′42″–115°25′18″E and
latitudes of 21°17′36″–23°55′54″N, and as of 2008 had a
permanent population of over 47 million. The study area
includes 9 major cities: Guangzhou, Shenzhen, Dongguan,
Zhongshan, Foshan, Huizhou, Zhaoqing, Jiangmen and Zhuhai
(Wong et al., 2002). The area experiences humid subtropical
weather with an annual average temperature and rainfall of
22 °C and 1690 mm, respectively. The mild climate facilitates
agriculture, allowing an annual production of 10–15 crops of
vegetables and three crops of grains.
2.1.2. Hydrogeological setting
The PRD is largely covered with Quaternary sediments, and
has an elevation ranging from 6 to 9 m above sea level in the
north to 1–2 m near the coast (Wang et al., 2012). The basement
rocks include shale, sandstone, limestone, dolomite, granite, and
gneiss ranging in age from Cambrian to Tertiary. The types of
groundwater include loose stratum pore water, bedrock fissure
water and carbonate karst cave water. Hydrochemical dominant
types of groundwater are HCO3–Ca–Na, HCO3–Cl–Ca–Na, HCO3–
Ca, Cl–Na and Cl–HCO3–Na–Ca (Supplemental Fig. 1). Ground-
water is mainly recharged by vertical infiltration from precipi-
tation and agricultural irrigation, as well as by the lateral flow
from rivers and floods during the wet season. Aquifers in this
area are associated to fractures, with spring flowrate range from
4.32 m3/day to 86.4 m3/day. In most wells of this area, flowrate
can reach 1000 m3/day. In addition, at high tide, salt water from
the sea penetrates as a wedge underneath the freshwater of the
river, which is the reason for the seawater intrusion in this area
(Huang et al., 2013). The general direction of groundwater flow
in the aquitard and aquifer follows roughly the major river flow,
which is from northeast to the coast (GHST, 1981).
2.1.3. Land use
Since the reforms and opening-up policy of China, land use
has been changing in the PRD region for decades. For instance,
the previously plentiful cultivated lands, fish ponds, gardens
and woodland in the study area have been urbanized and
industrialized. Nowadays, land use in the PRD is complex,
where nearly half of the land (49.61%) is devoted to forest
and orchard (Table 1). The majority of the remaining area
can be classified as cropland (21.17%), paddy fields
(17.07%), urban area (6.11%), river (4.13%), and wasteland
and aquaculture (1.91%).
In this study, the land use was reclassified into six categories
according to the well surrounding environment:
• Cropland: Soybean, sugar cane, peanut, cassava, tea, vegeta-
bles, etc.
• Paddy field: Rice, lotus root, etc.
• Urban: Urban, residential areas, river, industrial areas
• Forest and orchard: Tea seed, Pine tree, Banyan tree, palm
trees, banana, citrus, pineapple, litchi, etc.
• Landfill: Domestic waste and construction waste sites
• Wasteland and aquaculture: Wasteland, fish pond, other
aquaculture sites
2.2. Socioeconomic data
In this study, the socioeconomic data (for 2008) for the
nine metropolitan areas of the PRD region were cited from
Guangdong Statistical Yearbook 2009 (GPBS, 2009). Data for
these urban areas, including permanent populations (PP),
population density (PD), urbanization, total volume of waste
water discharged (TWD), the volume of living waste water
discharged (LWD), the volume of industrial waste water
discharged (IWD), gross domestic product (GDP), primary
industry (farming, forestry, animal husbandry, aquaculture,
etc.) GDP (PIGDP), secondary industry (processing industry)
 Q. Zhang et al. / Journal of Contaminant Hydrology 182 (2015) 221–230 223

Table 1
Land use details for 9 cities of the PRD.

Region Cropland Urban area Forest and orchard Wasteland and aquaculture Paddy field River

Dongguan 46.91 13.54 28.35 1.61 1.46 8.12

Fushan 46.95 11.96 23.16 0.06 11.36 6.51
Guangzhou 30.71 7.26 46.06 0.39 11.06 4.51
Huizhou 10.58 1.20 67.42 0.28 18.26 2.27
Shenzhen 7.57 24.31 64.06 2.47 0.08 1.51
Zhaoqing 8.67 1.47 66.73 0.00 19.43 3.70
Jiangmen 5.76 2.59 50.16 2.30 35.49 3.70
Zhongshan 62.52 9.07 20.17 1.89 0.11 6.24
Zhuhai 33.97 11.72 25.69 23.93 0.44 4.26
Total area (km2) 8806.72 2541.76 20637.76 794.56 7101.12 1718.08
Percentage of total area (%) 21.17 6.11 49.61 1.91 17.07 4.13

GDP (SIGDP), tertiary industry (service industry) GDP (TIGDP),
built-up urban area (BUA), cultivated land area (CLA), and green
area (GA), were assembled and examined in order to quantita-
tive describe the relationships between these factors and
groundwater nitrate contamination.
2.3. Sample collection and analysis
A total of 899 groundwater samples were collected mainly
between January–May and October–December in 2007 and
2008 (the only 56 groundwater samples were collected in July
and September, 2007 and 2008). All the wells chosen for
groundwater sampling were commonly used for domestic and/
or industrial purposes. Before sampling, the physico-chemical
parameters of the groundwater were measured on-site using a
WTW Multi 340i/SET multiparameter instrument (Germany),
equipped with conductivity (EC), electrode potential (Eh),
dissolved oxygen (DO), and pH probes. The probes were
calibrated using appropriate standard solutions or procedures
before measurement of the groundwater. Samples were collect-
ed at the depth of 50 cm below the water table using a stainless
steel sampler and filtrated through 0.45 μm filter membranes,
and then two 250-mL high-density polyethylene sampling
bottles were filled with these samples. Note that the buckets
and bottles had been rinsed three times with well water
before samples were collected. Immediately after collection,
samples were brought to the laboratory for analyses of
major ions analyses. The samples, without preprocessing,
were used for anions analyses. The samples, used for
cations, were acidified with HCL to pH b 2.
All analyses were carried out at the Groundwater Mineral
Water and Environmental Monitoring Center of the Institute of
Hydrogeology and Environmental Geology, Chinese Academy
of Geological Sciences. Analyses of cations (potassium [K+],
sodium [Na+], calcium [Ca2 +], magnesium [Mg2 +] and
ammonia [NH+ 4 ]) in the water samples were performed using
inductively coupled plasma-mass spectrometry (Agilent 7500ce
ICP-MS, Tokyo, Japan), while analyses of anions (nitrate[NO− 3 ],
nitrite[NO− − − −
2 ], chloride[Cl ], bromide[Br ], iodide[I ] and sulfate
[SO2−
4 ]) were carried out using spectrophotometry (Perkin-
Elmer Lambda 35, United States).
2.4. Well information
In our study, the sampling well mainly includes open hole
well (845 wells), manually operated pumping well (33 wells)
and well (21 wells). Well construction information was sup-
plied by the well owners and field inspection. The open hole
well mean depth was 4.60 m (ranged from 0.38 to 16.00 m)
and mean groundwater table was 1.84 m (ranged from 0.10 to
8.50 m). The mean depth of manual-operated pumping wells
was 3.33 m (ranged from 1.00 to 10.00 m) and mean depth of
groundwater table was 1.09 m (ranged from 0.10 to 4.00 m).
The mean depth of well was 77.65 m (ranged from 30.00 to
120.00 m) and mean depth of groundwater table was 9.78 m
(ranged from 1.20 to 30.0 m). Most open hole well and manual-
operated pumping wells are not cased. The mean screen length
for the well is 43 m (ranged from 22.00 to 54.00 m).
2.5. Multivariate data analysis
The differences in groundwater nitrate concentration of the
different land use types were evaluated using the nonpara-
metric Kruskal–Wallis test and the nonparametric Mann–
Whitney U-test, as the data failed to meet the assumption of
normality for parametric statistical analyses. Correlation anal-
ysis (Spearman's Rank Correlation Coefficient) was used to
determine the degree of association between water quality
parameters. The degree of association between the levels of
groundwater nitrate and socioeconomic parameters for differ-
ent regions were determined by using regression analysis. The
main driving mechanism of groundwater nitrate contamina-
tion was identified by using the principal component analysis
(PCA). The Kruskal–Wallis test, Mann–Whitney U-test, and PCA
were performed using SPSS software (version 19.0; SPSS Inc.,
Chicago, IL, USA). The RA was performed using OriginPro
(version 9.0, OriginLab Corp., Northampton, MA, USA).
3. Results
3.1. NO−
3
The median NO− 3 level was 23.82 mg/L (ranged from 0
to 1683.46 mg/L) and 238 samples among the total 899
analyzed samples exceeded the WHO limit (50 mg/L). Therefore,
the groundwater has been seriously affected by the human
activities and could limit its sustainability as drinking water
resource in future for the PRD regions. The NO− 3 level showed
significant spatial variation (P b 0.05). The highest median
nitrate level (31.33 mg/L) and the highest ratio of exceeded
WHO standard (29.42%) were found in the Shenzhen. However,
the lowest median nitrate level (14.87 mg/L) and the lowest ratio
224 Q. Zhang et al. / Journal of Contaminant Hydrology 182 (2015) 221–230

of exceeded WHO standard (14.29%) were found in the Jiangmen Wallis test and Mann–Whitney U-test (Table 3). The results
and Zhuhai, respectively. The NO− 3 level showed no seasonal showed that land use has a significant effect on groundwater
difference in the PRD region (P N 0.05). NO− + −
3 , NH4 and NO2 concentration (P b 0.001).
The highest median NO− 3 concentration (38.14 mg/L), and
−
3.2. NH+
4 and NO2 the highest ratio of exceeded WHO standard (42.50%), were
found in the landfill. However, there were no significant
The groundwater NH+ −
4 and NO2 concentration ranged from differences between NO− 3 concentration resulting from landfill
0 to 800.00 mg/L and was 0 and 155.35 mg/L and the median and from other land uses (urban (32.16 mg/L) (P = 0.869),
values were 0.04 and 0.019 mg/L, respectively. The NH+ 4 and paddy field (31.61 mg/L) (P = 0.766), cropland (19.69 mg/L)
NO− 2 level showed significant spatial variation (P b 0.05). The (P = 0.718) and forest and orchard (9.41 mg/L) (P = 0.390)),
highest median NH+ 4 level (0.08 mg/L) and the highest ratio of except for that from the wasteland and aquaculture areas
exceeded (56.03%) quality standard for groundwater in China (2.23 mg/L) (P = 0.000). High median NO− 3 concentrations and
(0.257 mg/L) (MGMRC, 1993) were found in the Shenzhen and the ratio of exceeded WHO standard (32.20%, 30.77%, and
Dongguan, respectively. However, the lowest median NH+ 4 level 25.21%) were found in the urban areas, paddy fields and
(0.02 mg/L) was found in Jiangmen. The highest median NO− 2 cropland, respectively. The median NO− 3 concentration and the
level (0.39 mg/L) and the highest ratio of exceeded (63.16%) ratio of exceeded WHO standard (6.25%) were found to be
quality standard for groundwater in China (0.066 mg/L) relatively low in the forest and orchard areas, and significantly
(MGMRC, 1993) were found in Zhongshan. However, the lower than in paddy fields (P = 0.000), cropland (P = 0.000)
lowest median NO− 2 level (0.003 mg/L) was found in Jiangmen and urban areas (P = 0.000). The lowest median nitrate
−
and Zhaoqing. The NH+ 4 and NO2 levels showed no seasonal concentration (2.23 mg/L) and the lowest ratio of exceeded
difference in the PRD region (P N 0.05). WHO standard (0%) were found in the wasteland and aqua-
culture areas.
−
3.3. Correlation between water quality parameters The highest median NH+ 4 and NO2 concentration (2.90 and
0.06 mg/L), and the highest ratio of exceeded GB/T14848-93
Correlation analyses between the water quality parameters standard (77.50% and 47.50%), were found in the landfill and
−
(Table 2) show very significant positive correlation (P b 0.01) there were significant differences between NH+ 4 and NO2
between Cl−, NO− − 2− + + +
3 , NO2 , SO4 , NH4 , K , Na , Ca
2+
and Mg2+, concentration resulting from landfill and other land use types
− − − +
except NO3 and NO2 (0.021) and NO3 and NH4 (−0.119). The (Table 3) (P = 0.000). There were no significant differences
highest significant correlation is found between Cl− and Na+, between wasteland and aquaculture, urban, cropland, forest
NO− − +
3 , NO2 and NH4 . There is a significant negative correlation and orchard and paddy field (Table 3).
(P b 0.01) between DO, Eh and the following parameters: NO− 2 ,
−
SO2− + + +
4 , Cl , NH4 , K , Na , Ca
2+
and Mg2+, however, there is a 3.5. Nitrate and regional socioeconomic development
significant positive correlation (P b 0.01) between NO− 3 and Eh.
In addition, there is a significant positive correlation (P b 0.01) The socioeconomic development maybe a driving factor of
between pH and the following parameters: NO− 2− −
2 , SO4 , Cl , groundwater nitrate contamination. We analyzed the nitrate
+ + 2+ 2+
K , Na , Ca and Mg . concentrations and the level of regional socioeconomic devel-
opment by using regression analysis. As shown in Fig. 1, the
3.4. Correlation between NO− + −
3 , NH4 , NO2 and land use types concentration of nitrate was significantly correlated with the
PP, PD, urbanization, TWD and LWD, but not with IWD. The
Human activity has dramatically increased NO− +
3 , NH4 and concentration of nitrate and the BUA and GA showed also
NO−
2 levels in groundwater. In order to further explore the significant positive relationships (forest and grassland), but not
effects of human activities on groundwater NO− +
3 , NH4 and with CLA (Fig. 2). In addition, there was a significant positive
NO− − + −
2 , we compared NO3 , NH4 and NO2 concentrations under relationship between the concentration of nitrate and GDP
different land use types by using the nonparametric Kruskal– (Fig. 3).

Table 2
Spearman correlation coefficients between the water quality parameters.

Parameters pH DO Eh NO−
3 NH+
4 NO−
2 Cl− SO2−
4 K+ Na+ Ca2+ Mg2+

pH 1.000
DO 0.045 1.000
Eh −0.460** 0.427** 1.000
NO− 3 −0.054 −0.028 0.094** 1.000
NH+ 4 0.046 −0.423** −0.265** −0.119** 1.000
NO− 2 0.222** −0.319** −0.315** 0.021 0.443** 1.000
−
Cl 0.128** −0.458** −0.375** 0.930** 0.869** 0.824** 1.000
SO2−
4 0.330** −0.307** −0.372** 0.232** 0.231** 0.305** 0.604** 1.000
K+ 0.283** −0.293** −0.323** 0.350** 0.268** 0.274** 0.666** 0.594** 1.000
Na+ 0.201** −0.476** −0.457** 0.889** 0.895** 0.858** 0.943** 0.646** 0.680** 1.000
Ca2+ 0.550** −0.387** −0.596** 0.128** 0.288** 0.338** 0.605** 0.715** 0.524** 0.639** 1.000
Mg2+ 0.319** −0.347** −0.431** 0.080* 0.293** 0.259** 0.676** 0.593** 0.523** 0.647** 0.642** 1.000

Note: ⁎P b 0.05; ⁎⁎P b 0.01.

 Q. Zhang et al. / Journal of Contaminant Hydrology 182 (2015) 221–230 225

Table 3
The median concentration of NO− + − ⁎
3 , NH4 and NO2 for different land use in the PRD .

Parameters Cropland Paddy field Urban Forest and orchard Landfill Wasteland and P-value
aquaculture

Number of samples 234 78 413 112 40 22 −

NO−
3 NO− 3 level (mg/L) 19.69a 31.61a 32.16a 9.41b 38.14ab 2.23c 0
Number of exceeded WHO standard 59 24 133 7 17 0 −
Rate of exceeded WHO standard (%) 25.21 30.77 32.20 6.25 42.50 0 −
NH+
4 NH+ 4 level (mg/L) 0.04b 0.02b 0.08b 0.02b 2.90a 0.11b 0
Number of exceeded GB/T14848-93 standard 58 9 136 22 31 4 −
Rate of exceeded GB/T14848-93 standard (%) 24.79 11.54 32.93 19.64 77.50 18.18 −
NO−
2 NO− 2 level(mg/L) 0.01b 0.01b 0.04b 0.01b 0.06a 0.01b 0
Number of exceeded GB/T14848-93 standard 57 8 142 17 19 2 −
Rate of exceeded GB/T14848-93 standard (%) 24.36 10.26 34.38 15.18 47.50 9.01 −

Note: ⁎The values of NO− + −

3 , NH4 and NO2 levels refer to median. The urban areas include residential building, commercial and industrial area. Significant levels from
nonparametric Kruskal–Wallis test and Mann–Whitney U-test for comparison of NO− + −
3 , NH4 and NO2 concentrations for different land use in the PRD. The Kruskal–
Wallis test was used to determine if there was a significant difference between six groups of data. The Mann–Whitney U-test was used to determine if there was a
difference between two groups of data.

4. Discussion level drop to the ground, so that the underground environment

4.1. Factors affecting groundwater nitrate contamination
4.1.1. Nitrate and land use types
Land use, as the direct reaction of human activity, is closely
related to the groundwater pollution. Previous researches
found that land use has a significant effect on groundwater
nitrate concentration (Lockhart et al., 2013; Qin et al.,
2013). In this study, the highest median nitrate concentra-
tion and the highest ratio of exceeded WHO standard were
found in the landfill and the median nitrite and ammonia
level of the landfill were significantly higher than the other land
use, which indicates that landfill leakage was an important
source of nitrate in groundwater in PRD (Gu et al., 2011, 2012,
2013). However, there were no significant differences between
nitrate concentration resulting from landfill and cropland, paddy
field, urban and forest and orchard, which may be due to the
different rubbish types and hydrogeology condition of landfill
and the landfills carry the measures to prevent leakage and
without doing (there were 7 landfills that carry the measures to
prevent leakage in our investigated the 40 landfills in this study),
lead to the higher difference of the nitrate concentration in dif-
ferent landfills.
High nitrate concentrations and the ratio of exceeded WHO
standard were found in the urban areas, paddy fields and
cropland, respectively. These high nitrate concentrations are
primarily due to the application of fertilizers and wastewater
irrigation in paddy field and cropland areas, and domestic and
industrial sewage directly discharged into the nearby river in
the urban areas (Lockhart et al., 2013; Qin et al., 2013). It is
worth noting that the median nitrate level was higher in
paddy fields than in cropland, however, the median nitrite and
ammonia level was higher in paddy fields than in cropland.
Kumazawa (2002) found that the paddy fields are often being
in a water-saturated state, thus, groundwater is often in
reductive environment and is beneficial to denitrification.
Therefore, the groundwater of the paddy fields is less
susceptible to nitrate pollution. However, this result is not
consistent with previous researches, which is most probably
because the time of sampling was mainly in dry season and
the paddy fields are in the idle state leading to the groundwater
is changed into beneficial to nitrification. In addition, the
groundwater table is lower in the PRD, resulting in nitrate
easily migrating into the shallow groundwater.
In the forest and orchard areas, the median nitrate con-
centration and the ratio of exceeded WHO standard were
relatively low, which is mainly due to less fertilizer being
applied and effects of human activity in those areas. According
to field investigation, the fertilizing amount (nitrogen fertilizer)
of farmland is 1500–2250 kg ha−1year−1, however, the
fertilizing amount (nitrogen fertilizer) is 1000–1500 kg ha−1-
year−1 in orchard areas and the forest areas in not fertilized. In
addition, the retain fertilizer capacity is stronger to the trees
than crops. The lowest median nitrate concentration and the
lowest ratio of exceeded WHO standard were found in the
wasteland and aquaculture areas. This low value appears to be
mainly due to the less fertilizer using in these areas.
4.1.2. Nitrate and regional socioeconomic development
Nitrate levels in groundwater have gradually increased with
rapid urbanization and economic development. Therefore, it is
important to study the forces that drive the generation of
nitrate pollution in order to help devise ways to regulate
groundwater nitrate contamination. As shown in Fig. 1, the
concentration of nitrate was significantly correlated with the
PP, PD, urbanization, TWD and LWD (Gu et al., 2013), but not
with IWD, indicating that domestic waste discharge was the
main source of pollution during rapid urbanization. The high
levels of urbanization, PP and PD would appear to be causing
high levels of domestic wastewater being discharged into the
water environment, and resulting in the increased nitrate con-
centration in the groundwater. During the study period, we
found that huge amount of domestic sewage was directly
drained into the nearby river or ditches without treatment, and
finally the domestic sewage infiltrates into groundwater.
Bilsborough and Mann (2006) also found that the per capita
N leakage via excretion is stable because the human's basic
metabolic need on protein is stable. Therefore, it was expected
that the nitrate levels of groundwater would be highly correlated
with the permanent populations and population density. Gu
et al. (2013) found that the higher nitrate concentrations usually
 226
Q. Zhang et al. / Journal of Contaminant Hydrology 182 (2015) 221–230
Fig. 1. Relation between permanent populations (PP), population density (PD), urbanization, total volume of waste water discharged (TWD), the volume of living waste water discharged (LWD), the volume of industrial waste
water discharged (IWD) and nitrate concentration in groundwater of Pearl River Delta areas.
Q. Zhang et al. / Journal of Contaminant Hydrology 182 (2015) 221–230 227

occur in the more urbanized level regions, such as Beijing,

Shanghai, and Tianjin.
There were significant positive relationships between the
concentration of nitrate and the BUA and GA (forest and
grassland) (Fig. 2). The BUA and GA are indexes that reflect the
level of urbanization. The larger the built-up urban area, the
more it could accommodate a larger population, and the
greater its potential to produce more domestic wastewater

Fig. 2. Relation between the built-up urban area (BUA), green area (GA), cultivated land area (CLA) and nitrate concentration in groundwater of Pearl River Delta areas.
and hence groundwater nitrate pollution. Gu et al. (2012) also
reported that the nitrogen input significantly correlated with
the built-up urban area. However, the concentration of nitrate
was not significantly correlated with the CLA, which differs
from the results of a previous study. Gu et al. (2013) concluded
that the N leakage was significantly related with cropland area
in each province of China. The difference with our study is that
our research mainly involved regions undergoing rapid urban-
ization, and the most important source for nitrate in groundwa-
ter maybe was domestic wastewater instead of agriculture non-
point pollution sources.
As shown in Fig. 3, there was a significant positive
relationship between the concentration of nitrate and GDP. This
relationship is consistent with the result reported by Gu et al.
(2013) in 31 provinces of China. Interestingly, in our study, there
was no significant relationship between the concentration of
nitrate and the primary industry (agriculture) GDP, but, there
were significant positive relationships between the concentra-
tion of nitrate and the secondary industry (industrial) GDP and
the tertiary industry (services) GDP. These became most possible
because the economic development of the PRD area relies mainly
on manufacturing (a part of the secondary industry) and tertiary
industry, which could lead to the discharge of relatively high
amounts of industrial and domestic wastewater, and then give
rise to the relatively high concentration of groundwater nitrate.

4.2. Driving mechanism of groundwater nitrate contamination

In order to identify the main factors causing groundwater

nitrate concentration, we used PCA to reduce the original
eleven variables into three key independent factors (Table 4).
On the basis of eigenvalues N 1, PCA generated three principal
components (PCs) comprising 93.69% of the total variance.
As shown in Table 4, PC1 accounted for 68.15% of the total
variance and had strong and positive loadings on GA, BUA,
TIGDP, urbanization, LWD, NO− 3 and SIGDP. The GA, BUA and
urbanization are the indexes that reflect the level of urban
construction and the TIGDP and SIGDP represent the secondary
and tertiary industrial development. There are strong relation-
ships between the levels of NO− 3 and GA, BUA, TIGDP,
urbanization, LWD and SIGDP. Thus, PC1 can be identified as
the driving force of the urban construction and the secondary
and tertiary industrial development for groundwater nitrate
contamination.
PC2 accounted for 17.34% of the total variance and had
strong and positive loadings on PP and PD, strongly negative
loading on the CLA, and moderately positive loadings on NO− 3 .
The PP, PD and CLA are the indexes that reflect population
growth, and had a strong relationship with NO− 3 . Thus, PC2 can
be identified as the driving force of population growth for
groundwater nitrate contamination. PC3 accounted for 8.20% of
the total variance and had strong and positive loadings on
PIGDP and IWD. Thus, PC3 can be identified as the driving force
228 Q. Zhang et al. / Journal of Contaminant Hydrology 182 (2015) 221–230

Fig. 3. Relation between the Gross Domestic Product (GDP), primary industry GDP (PIGDP), secondary industry GDP(SIGDP), tertiary industry GDP(TIGDP) and nitrate
concentration in groundwater of Pearl River Delta areas.

of the primary industrial. However, there is no relationship
between PC3 and NO− 3 . Thus, primary industrial develop-
ment is not a primary driving force for groundwater nitrate
contamination.
4.3. Source of groundwater nitrate pollution
To solve groundwater nitrate contamination problems, it is
first necessary to identify the source(s) of the contamination.
Table 4
The halides Cl−, Br−, and I− are conservative anions because
they have minimal interactions with surrounding substrates in
most situations as they are transported with ground water
(Panno et al., 2006). Therefore, they are good indicators of
pollution sources.
Generally, the concentration of Cl− in sewage and manure is
relatively high, and Cl− is thus an indicator of sewage. Plotting
NO− −
3 versus Cl is widely used to identify the sources of nitrate
pollution (Lee et al., 2008; Li et al., 2010; Widory et al., 2005).
As shown in Fig. 4, there was a strong positive relationship
between NO− − 2
3 and Cl (R = 0.976, P b 0.001), which indicated

Driving mechanism of groundwater nitrate contamination in the PRD. that groundwater nitrate in the PRD region could mainly come
from anthropogenic sources such as sewage and manure. This
Parameters Principal components
result differs from that of the previous studies of Widory et al.
PC1 PC2 PC3 (2005) and Lee et al. (2008), who found no such relationship
GA 0.951 0.069 0.170 between the NO− −
3 and Cl . This difference could be due to the
BUA 0.940 0.038 0.165 pollution sources of nitrate being more complex in their study
TIGDP 0.921 0.215 0.293 area. In the PRD region, the average annual volume of living
Urbanization 0.897 0.316 0.266
LWD 0.892 0.189 0.399
waste water discharged was 28.01 × 108 in 2000–2007,
NO−3 0.862 0.468 −0.046 however, there were only 20 urban sewage treatment plants
SIGDP 0.765 0.517 0.287 in the PRD region and the processing rate of urban sewage is
CLA 0.051 −0.970 0.196 only 14% (the processing rate of village sewage is less 1%)
PP 0.060 0.906 0.349
(GPBS, 2009). During the study period, we found that a great
PD 0.622 0.728 −0.138
PIGDP 0.181 −0.125 0.870 deal of domestic sewage was directly discharged into the
IWD 0.282 0.051 0.848 nearby ground or ditches without treatment, and of the 40
Eigenvalue 9.54 2.43 1.15 landfills we investigated, only 7 were made impermeable. In
% total variance 68.15 17.34 8.20 addition, the groundwater table is very low (mean values is
Cumulative % 68.15 85.49 93.69
2.03 m) in the PRD region, and sewage therefore infiltrates
 Q. Zhang et al. / Journal of Contaminant Hydrology 182 (2015) 221–230
Fig. 4. Relationship between the Cl− and NO−
3 concentration of the groundwater in Pearl River Delta areas.
relatively easily into the groundwater. In addition, there was a
significant positive correlation between CL− and NO− −
3 , NO2 ,
SO2−
4 , NH +
4 , K+
, Na +
, Ca2+
and Mg 2+
. In summary, we think
that sewage is the main source of groundwater nitrate in the
PRD region.
In order to further differentiate between the sources of
nitrate pollution, we used the system for I/Na against Br by
Panno et al. (2006) which was developed with ranges of
sources affecting groundwater for tracing Na+ and Cl−. Pastén-
Zapata et al. (2014) found that this system was a very effective
tool that allowed for a complete separation between different
sources. As shown in Fig. 5, the main pollution sources of
groundwater nitrate in the PRD region determined using this
method are sewage, manure and chemical fertilizer, accounting
for 93.8% of samples, followed by landfill leachate (2.7%), and
seawater intrusion (2.2%). In addition, 1.3% of the samples were
pristine water.
Nitrate contamination in groundwater has become an
environmental problem of widespread concern. The key step
in an effective management strategy is to identify the sources of
NO− 3 in groundwater. The system for I/Na against Br was by
Panno et al. (2006) developed with ranges of sources affecting
groundwater for tracing Na+ and Cl−. In our study, we used
Fig. 5. Relationship between the I/Na ratios and Br− concentrations showing separation of sources according to Panno et al. (2006).
229
this system to differentiate between the sources of nitrate
pollution and get an approximate result. However, we cannot
determine the endmember values of the potential nitrate
pollution sources and the system failed to distinguish the
pollution sources from sewage, manure and chemical fertilizer,
which may have an effect for tracking nitrate sources. Therefore,
further studies are needed to determine the endmember value of
all potential nitrate sources and differentiate pollution sources of
the sewage, manure and chemical fertilizer.
5. Conclusion
The results of this study show that human activity has
dramatically increased the groundwater nitrate levels in the PRD.
The median concentration of groundwater nitrate (23.82 mg/L)
and ratio of exceeded WHO standard (26.47%) were shown to be
high in the rapidly urbanized area of the PRD based on 899
samples. By comparing nitrate concentrations under different
land use types, we found the landfill with the highest mean
nitrate concentration (38.14 mg/L) and the highest ratio of
exceeded WHO standard (42.50%), therefore, the landfill leakage
was an important source of nitrate found in groundwater in PRD.
The ratio of exceeded WHO standard in paddy fields was higher
230 Q. Zhang et al. / Journal of Contaminant Hydrology 182 (2015) 221–230
than in cropland due to the nitrate easily leaching from paddy
fields into the shallow groundwater.
RA results show that domestic wastewater and industrial
wastewater were the main sources of groundwater nitrate
pollution in the rapidly urbanized region since the concentra-
tion of nitrate was determined to be significantly related to the
PP, PD, urbanization, TWD, LWD, BUA, GA, GDP, SIGDP, TIGDP
and Cl−. PCA results show that the main driving mechanisms of
groundwater nitrate contamination were urban construction
and the secondary and tertiary industrial development, and
population growth.
The water chemistry analysis results show that the system
for I/Na against Br is a very effective tool that allowed for a
complete separation between different sources. Using this tool,
the main pollution sources of groundwater nitrate in the PRD
region were determined to be sewage, manure and chemical
fertilizer, accounting for 93.8% of samples, followed by landfill
leachate (2.7%), seawater intrusion (2.2%), and pristine water
(1.3%). This tool, however, did not distinguish between waste-
water, chemical fertilizer and manure. Therefore, in order to
mitigate the contamination of groundwater nitrate in the PRD,
further study is necessary to identify these unknown sources.
Supplementary data to this article can be found online at
http://dx.doi.org/10.1016/j.jconhyd.2015.09.009.
Acknowledgments
The study has been funded by the National Natural Science
Foundation of China (no. 41401593) and the projects of China
Geological Survey (no. 12120115048001 and 1212011220982).
References
Bilsborough, S., Mann, N., 2006. A review of issues of dietary protein intake in
humans. Int. J. Sport Nutr. Exerc. Metab. 16, 129.
Burow, K.R., Nolan, B.T., Rupert, M.G., Dubrovsky, N.M., 2010. Nitrate in
groundwater of the United States, 1991–2003. Environ. Sci. Technol. 44,
4988–4997.
Cheung, K., Poon, B., Lan, C., Wong, M., 2003. Assessment of metal and nutrient
concentrations in river water and sediment collected from the cities in the
Pearl River Delta, south China. Chemosphere 52, 1431–1440.
Dubrovsky, N.M., Burow, K.R., Clark, G.M., Gronberg, J., Hamilton, P.A., Hitt, K.J.,
Mueller, D.K., Munn, M.D., Nolan, B.T., Puckett, L.J., 2010. The Quality of Our
Nation's Water: Nutrients in the Nation's Streams and Groundwater, 1992–
2004. US Department of the Interior, US Geological Survey.
Fan, A.M., Steinberg, V.E., 1996. Health implications of nitrate and nitrite in
drinking water: an update on methemoglobinemia occurrence and
reproductive and developmental toxicity. Regul. Toxicol. Pharmacol. 23,
35–43.
Fu, B.-J., Zhuang, X.-L., Jiang, G.-B., Shi, J.-B., Lu, Y.-h., 2007. Feature:
environmental problems and challenges in China. Environ. Sci. Technol.
41, 7597–7602.
GHST (Guangdong Hydrogeological Second Team), 1981. Regional
Hydrogeological Survey Report. Guangdong Hydrogeological Second
Team, Guangzhou [in Chinese].
GPBS (Compiled by Guangdong Provincial Bureau of Statistics), 2009.
Guangdong Statistical Yearbook. China Statistics Press.
Gu, B., Zhu, Y., Chang, J., Peng, C., Liu, D., Min, Y., Luo, W., Howarth, R.W., Ge, Y.,
2011. The role of technology and policy in mitigating regional nitrogen
pollution. Environ. Res. Lett. 6, 014011.
Gu, B., Dong, X., Peng, C., Luo, W., Chang, J., Ge, Y., 2012. The long-term impact of
urbanization on nitrogen patterns and dynamics in Shanghai, China.
Environ. Pollut. 171, 30–37.
Gu, B., Ge, Y., Chang, S.X., Luo, W., Chang, J., 2013. Nitrate in groundwater of
China: sources and driving forces. Glob. Environ. Chang. 23, 1112–1121.
Güler, C., Kurt, M.A., Alpaslan, M., Akbulut, C., 2012. Assessment of the impact of
anthropogenic activities on the groundwater hydrology and chemistry in
Tarsus coastal plain (Mersin, SE Turkey) using fuzzy clustering, multivar-
iate statistics and GIS techniques. J. Hydrol. 414, 435–451.
Huang, G., Sun, J., Zhang, Y., Chen, Z., Liu, F., 2013. Impact of anthropogenic and
natural processes on the evolution of groundwater chemistry in a rapidly
urbanized coastal area, south China. Sci. Total Environ. 463, 209–221.
Kaushal, S.S., Groffman, P.M., Band, L.E., Elliott, E.M., Shields, C.A., Kendall, C.,
2011. Tracking nonpoint source nitrogen pollution in human-impacted
watersheds. Environ. Sci. Technol. 45, 8225–8232.
Kumazawa, K., 2002. Nitrogen fertilization and nitrate pollution in groundwater
in Japan: present status and measures for sustainable agriculture. Nutr.
Cycl. Agroecosyst. 63 (2–3), 129–137.
Landon, M.K., Green, C.T., Belitz, K., Singleton, M.J., Esser, B.K., 2011. Relations of
hydrogeologic factors, groundwater reduction–oxidation conditions, and
temporal and spatial distributions of nitrate, central-eastside San Joaquin
Valley, California, USA. Hydrogeol. J. 19, 1203–1224.
Lee, K.-S., Bong, Y.-S., Lee, D., Kim, Y., Kim, K., 2008. Tracing the sources of nitrate
in the Han River watershed in Korea, using δ 15 N-NO 3− and δ 18 O-NO
3− values. Sci. Total Environ. 395, 117–124.
Li, S.-L., Liu, C.-Q., Li, J., Liu, X., Chetelat, B., Wang, B., Wang, F., 2010. Assessment
of the sources of nitrate in the Changjiang River, China using a nitrogen and
oxygen isotopic approach. Environ. Sci. Technol. 44, 1573–1578.
Li, S.-L., Liu, C.-Q., Li, J., Xue, Z., Guan, J., Lang, Y., Ding, H., Li, L., 2013. Evaluation
of nitrate source in surface water of southwestern China based on stable
isotopes. Environ. Earth Sci. 68, 219–228.
Lockhart, K., King, A., Harter, T., 2013. Identifying sources of groundwater
nitrate contamination in a large alluvial groundwater basin with highly
diversified intensive agricultural production. J. Contam. Hydrol. 151,
140–154.
Lord, E., Anthony, S., Goodlass, G., 2002. Agricultural nitrogen balance and water
quality in the UK. Soil Use Manag. 18, 363–369.
Mai, B.-X., Fu, J.-M., Sheng, G.-Y., Kang, Y.-H., Lin, Z., Zhang, G., Min, Y.-S., Zeng,
E.Y., 2002. Chlorinated and polycyclic aromatic hydrocarbons in riverine
and estuarine sediments from Pearl River Delta, China. Environ. Pollut. 117,
457–474.
Ministry of Geology and Mineral Resources of China (MGMRC), 1993. Quality
Standard for Ground Water (GB/T14848-93). The State Bureau of Quality
and Technical Supervision (in Chinese).
Panno, S., Hackley, K., Hwang, H., Greenberg, S., Krapac, I., Landsberger, S.,
O'Kelly, D., 2006. Characterization and identification of Na–Cl sources in
ground water. Groundwater 44, 176–187.
Pastén-Zapata, E., Ledesma-Ruiz, R., Harter, T., Ramírez, A.I., Mahlknecht, J.,
2014. Assessment of sources and fate of nitrate in shallow groundwater of
an agricultural area by using a multi-tracer approach. Sci. Total Environ.
470, 855–864.
Qin, R., Wu, Y., Xu, Z., Xie, D., Zhang, C., 2013. Assessing the impact of natural
and anthropogenic activities on groundwater quality in coastal alluvial
aquifers of the lower Liaohe River Plain, NE China. Appl. Geochem. 31,
142–158.
Qiu, J., 2011. China to spend billions cleaning up groundwater. Science 334,
745-745.
Rankinen, K., Salo, T., Granlund, K., 2008. Simulated nitrogen leaching, nitrogen
mass field balances and their correlation on four farms in south-western
Finland during the period 2000–2005. Agric. Food Sci. 16, 387–406.
Salo, T., Turtola, E., 2006. Nitrogen balance as an indicator of nitrogen leaching
in Finland. Agric. Ecosyst. Environ. 113, 98–107.
Stigter, T., Ribeiro, L., Dill, A., 2008. Building factorial regression models to
explain and predict nitrate concentrations in groundwater under agricultural
land. J. Hydrol. 357, 42–56.
Wang, Y., Jiao, J.J., Cherry, J.A., 2012. Occurrence and geochemical behavior of
arsenic in a coastal aquifer–aquitard system of the Pearl River Delta, China.
Sci. Total Environ. 427, 286–297.
Weyer, P.J., Cerhan, J.R., Kross, B.C., Hallberg, G.R., Kantamneni, J., Breuer, G.,
Jones, M.P., Zheng, W., Lynch, C.F., 2001. Municipal drinking water nitrate
level and cancer risk in older women: the Iowa Women's Health Study.
Epidemiology 12, 327–338.
WHO (World Health Organization), 2011. Guidelines for Drinking-Water
Quality. 4th ed. World Health Organization.
Wick, K., Heumesser, C., Schmid, E., 2012. Groundwater nitrate contamination:
factors and indicators. J. Environ. Manag. 111, 178–186.
Widory, D., Petelet-Giraud, E., Négrel, P., Ladouche, B., 2005. Tracking the
sources of nitrate in groundwater using coupled nitrogen and boron
isotopes: a synthesis. Environ. Sci. Technol. 39, 539–548.
Wong, S., Li, X., Zhang, G., Qi, S., Min, Y., 2002. Heavy metals in agricultural soils
of the Pearl River Delta, south China. Environ. Pollut. 119, 33–44.
Xue, D., Botte, J., De Baets, B., Accoe, F., Nestler, A., Taylor, P., Van Cleemput, O.,
Berglund, M., Boeckx, P., 2009. Present limitations and future prospects of
stable isotope methods for nitrate source identification in surface-and
groundwater. Water Res. 43, 1159–1170.