Ecological Engineering 68 (2014) 9096

Contents lists available at ScienceDirect

Ecological Engineering
journal homepage: www.elsevier.com/locate/ecoleng

Three-stage horizontal subsurface ow constructed wetlands

for organics and nitrogen removal: Effect of aeration
Fengmin Li a, , Lun Lu a , Xiang Zheng b , Xiuwen Zhang a
a
b

College of Environmental Science and Engineering, Key Laboratory of Ministry of Education, Ocean University of China, Qingdao 266100, China
School of Environment and Natural Resources, Renmin University of China, Beijing 100872, China

a r t i c l e

i n f o

Article history:
Received 3 September 2013
Received in revised form 13 January 2014
Accepted 25 March 2014
Keywords:
Horizontal subsurface ow constructed
wetland
Dissolved oxygen
Aeration
Nitrication
COD removal

a b s t r a c t
Improving nitrogen removal in wastewater treatment is necessary for the extensive and effective use of
constructed wetlands (CWs) in rural areas. Six different three-stage horizontal subsurface ow (HSSF)
CWs were designed to examine the effects of articial aeration and aeration location on the treatment
of organics and nitrogen. In a typical conguration (unaerated surface and subsurface ow), single-stage
CWs fail to support aerobic and anoxic environments simultaneously. To adjust the distribution of dissolved oxygen (DO) in CWs and to improve the efciency of sewage water purication, typical HSSF CWs
were divided into different sections with enhanced functions. The treatment efciency of the proposed
HSSF CWs was compared with that of typical HSSF CW. Results show that the removal efciencies of
chemical oxygen demand (COD), ammonium nitrogen (NH4 + -N), and total nitrogen (TN) in typical HSSF
CW were 70.3%, 18.4%, and 24.6%, respectively, whereas the removal efciencies of CW1 with front and
rear aeration and anoxic treatment were 90.1%, 99.7%, and 51.3%, respectively, at average inuent COD
and NH4 + -N concentrations of 184.6 and 21.7 mg L1 , respectively. Furthermore, front aeration greatly
improved organics and nitrogen removal compared with middle aeration and rear aeration. The oxidized
conditions in the HSSF CWs could facilitate efcient biological and biochemical reactions, thus resulting in
efcient organics and NH4 + -N removal. However, high DO level and lack of carbon resources might restrain
denitrication even though NH4 + -N removal was enhanced due to aeration. Therefore, TN removal has
potential for improvement.
2014 Elsevier B.V. All rights reserved.

1. Introduction
China is in the midst of a water shortage. The average amount
of water resources per capita in China is only 23002400 m3 /year,
which is approximately 1/4 of the world average level (Webber
et al., 2008). Given the rapid urbanization and economic growth of
China, water scarcity due to water pollution is prevalent in most
parts of the country, thus limiting the water supply for agricultural,
industrial, and municipal use (Changming et al., 2001). According to
the National Water Resources Development Plan (20112015), the
effective farmland irrigation areas were only 62.097 million ha by
the end of 2011, which account for only 51.05% of the arable land in
the country. The availability of tap water is only 72.1% in rural areas
in 2011 (National Bureau of Statistics of China, 2013). Rural water
pollution is characterized as non-point source pollution and mainly

Corresponding author. Tel.: +86 053266782780.

E-mail addresses: lifengmin@ouc.edu.cn, lifengmin@tsinghua.org.cn (F. Li).
http://dx.doi.org/10.1016/j.ecoleng.2014.03.025
0925-8574/ 2014 Elsevier B.V. All rights reserved.

originates from agriculture and rural regions (overuse of fertilizers,

animal husbandry, and domestic raw sewage). Rural water pollution is the leading source of water pollution (Reddy and Behera,
2006; Shao, 2010). The most common pollutants in rural water
bodies (rivers and lakes) are classied as organics (chemical oxygen demand (COD)) and nutrients (N, P) (Wang et al., 2008). Large
quantities, high nutrient concentrations, and high portion of organics could be expected from rural setting. Considering the lack of
funding and technology, sewage is generally discharged directly to
rivers and lakes without any treatment (Ongley et al., 2010).
Constructed wetlands (CWs) are articial biomimetic wetlands that generally consist of gravel or sand as lter matrix,
rooted aquatic plants, and soil sustaining macrophytes, specifically microbes that propagate and metabolize by interacting
with pollutants in sewage (Brix, 1994; Cooper, 2009; Kadlec and
Wallace, 2009). CWs can be classied as surface ow and subsurface
ow systems depending on water level. Subsurface ow CWs are
divided into two groups: vertical ow and horizontal ow systems
(Vymazal, 2007). Compared with conventional energy-intensive

F. Li et al. / Ecological Engineering 68 (2014) 9096

treatment technologies, CWs have been proven as efcient and

environment-friendly treatment technologies with low operation
and maintenance costs (Babatunde et al., 2008; Lee et al., 2009;
Vymazal, 1996). Physical, chemical, and biological processes occur
simultaneously in CWs to reduce the amount of contaminants in
sewage (Rousseau et al., 2004).
Biological processes play an important role in the treatment of
organics and nitrogen in wastewater because microbial degradation is the dominant factor of organics and nitrogen removal (Truu
et al., 2009; Saeed and Sun, 2012). Furthermore, oxygen content
is important because proper oxygen distribution creates aerobic,
anoxic, and anaerobic zones that facilitate the propagation and
performance of various bacteria (Cooper, 2005; Dong et al., 2012;
Noorvee et al., 2007; Tanner and Kadlec, 2003).
Aquatic NH4 + -N decreases via nitrication and increases
through ammonication (Eqs. (1) and (2)) (Vymazal, 2007):
Organic N

Ammonication

Ammonium N

Ammonium N

Nitrication

Nitrite N

(1)

Nitrication

Nitrate N

(2)

In horizontal subsurface ow (HSSF) CWs, NH4 + -N removal

mostly relies on nitrication, which is highly dependent on environmental parameters such as dissolved oxygen (DO), pH, and
temperature. These parameters greatly affect nitriers, which
biologically oxidize NH4 + -N to NO3 -N with NO2 -N as the intermediate in the reaction sequence. Denitrication is extremely

91

important and is illustrated by the following equation (Hauck,

1984):
6(CH2 O) + 4NO3 6CO2 + 2N2 + 6H2 O

(3)

Denitrication occurs only under anaerobic or anoxic conditions

and in the presence of carbon source. Denitrifying bacteria utilize
nitrate instead of oxygen as terminal electron acceptors and organic
compounds as electron donors in the respiratory process.
Incomplete nitrication due to low oxygen availability results
in incomplete total nitrogen (TN) elimination. Considerable efforts
have been made to improve nitrogen removal in subsurface ow
CWs, such as the use of coarse grained substrates instead of
ne grained substrates to improve oxygen transfer (Saeed and
Sun, 2012); preaeration (Noorvee et al., 2007); articial aeration,
mainly classied as intermittent aeration (Dong et al., 2012; Fan
et al., 2013; Hu et al., 2012); continuous aeration (Butterworth
et al., 2013; Cottingham et al., 1999; Maltais-Landry et al., 2009).
However, excessive oxygenation can negatively affect TN removal
and result in nitrate accumulation via incomplete denitrication.
Thus, proper spatial aerobic and anoxic (anaerobic) conditions are
required to achieve simultaneous nitrication and denitrication
in constructed wetland.
This study investigates how articial aeration affects the
purication processes in HSSF CWs and indicates differences in
purication efciency among six HSSF CWs from different articial aeration and anaerobism combinations. The effects of different
arrangements and operating strategies on the abatement of COD,
NH4 + -N, and TN are also analyzed.

Fig. 1. (a) Schematic of the laboratory HSSF CWs, take CW1 as an example. (b) Sketch map of the HSSF CWs.

92

F. Li et al. / Ecological Engineering 68 (2014) 9096

2. Materials and methods

2.1. Experimental setup and operation
Laboratory-scale HSSF CWs were constructed by using polyvinyl
chloride (Fig. 1(a)). The length, width, and depth of the CWs were
1.25, 0.2, and 0.3 m, respectively. The HSSF CWs were lled 5 cm
deep with gravel with diameter ranging from 3 to 5 cm followed
by 15 cm of cinder (13 cm diameter) and 5 cm of soil. Each HSSF
CW had three cells and six sampled wells labeled 11, 12, 21, 22,
31, and 32, which were 10, 30, 50, 70, 90, and 110 cm from the
inlet, respectively. Forced aeration was provided by the aeration
system, which consisted of an air compressor and aeration tube.
Anoxic treatment was conducted through sealing the compartment
by plastic sheets to cut off oxygen diffusion from the atmosphere.
The typical HSSF CW did not include aerobic and anoxic treatment,
but partitioning was applied as the rest of other CWs (Fig. 1(b)).
Based on different treatment, six HSSF CWs were distinguished
and named as CW1CW6 (Fig. 1(b)). All wetland systems were
unplanted.
2.2. Operation conditions
By considering the characteristics of rural sewage water to
maintain the stability and continuity of the inlet, the sewage water
was mainly composed of glucose, starch, peptone, beef extract,
(NH4 )2 SO4 , NaHCO3 , Na2 CO3 , and CaCl2 . Table 1 shows the water
quality parameters. The initial COD and NH4 + -N concentrations
averaged 184.6 and 21.7 mg L1 , respectively. Nominal hydraulic
retention time (HRTn ) is calculated according to Kadlec and Wallace
(2009). Uniform HRTn among six CWs is assumed, since the ow
rate and the demotions of the CWs are same and the disturbance
of articial aeration could be ignored because the strength of them
is mild. As a result, the HRTn and the hydraulic loading rate (HLR)
were calculated as 5 d and 6 cm d1 , respectively.
2.3. Water sampling and analysis
To obtain stable water quality, all six HSSF CWs were preoperated for about one month. Water samples were collected once
a week. For each sample, DO and pH were immediately measured
in situ. The pH was measured by using a glass pH meter (Accumet
BASIC AB15, Singapore). DO was determined with WTW2B20-0017
DO tester. NH4 + -N, NO2 -N, and NO3 -N, and TN levels were analyzed according to standard analytical methods (Standard Method
for the Examination of Water and Wastewater Editorial Board,
2002). COD was determined by using a Hach DR/2400 spectrophotometer following the standard procedure.
2.4. Statistical analysis
All experimental data were expressed as an average of three
replicates with standard deviation. To compare the pollutant
removal efciencies of the six wetland systems, which had various DO levels resulting from different aeration strategies; statistical
analyses were performed through the statistical program SPSS 18.0
(SPSS Inc., Chicago, USA), including analysis of variance, Bartletts
and Levines test for homogeneity of variance and normality, and

Fig. 2. DO distribution along the sampling points following the ow direction in six
HSSF CWs. Error bar are standard deviations of the mean.

Tamhanes T2 test and Dunnett T3 test at 95% condence level

(a = 0.05) for one way ANOVA multiple comparisons for the mean
removal efciencies. The methods were selected for that equal variance between groups was not assumed.
3. Results and discussion
3.1. Variation of DO in the water ow in different stages of the
HSSF CWs
Appropriate DO distribution was in expectation along the ow
direction, for example, proving sufcient oxygen for nitrication
and organics degradation by added aeration, sealing some compartments to lower DO level for better denitrication. The aerobic
and anoxic conditions in the wetland reactor were distinguished by
using DO proles. Comparing the DO levels of the aeration sampling wells with those of the non-aeration wells, the DO levels in
the systems with aeration CWs (CW1, CW3, and CW5) were higher
than those with non-aeration CWs (CW2, CW4, and CW6) (p < 0.05).
Fig. 2 shows that the aerobic and anoxic conditions were well developed in CW1. The DO level along the ow direction in CW1 followed
a high-low-high tendency, with DO concentration decreasing from
3.5 mg L1 to about 1.4 mg L1 and then returning to 3.0 mg L1
and above. CW3 exhibited a DO concentration range of 1.0 mg L1 to
3.2 mg L1 . By contrast, the DO concentration in the non-aeration
CWs was low, indicating relatively anoxic conditions. With supplementary aeration and sealed anoxic treatment, the aerobic and
anoxic regions created in the aeration systems (CW1, CW3, CW5),
particularly in CW1 and CW3, strengthened the effectiveness of
the microbes in reducing degradable organics, thereby facilitating
nitrication and denitrication and achieving simultaneous organics reduction and nitrogen elimination.
3.2. COD removal
The inlet and outlet COD concentrations of HSSF CWs are shown
in Fig. 3(a). The inuent COD concentrations over the entire operational period ranged from 160.9 mg L1 to 248.3 mg L1 , with a
mean of 184.6 mg L1 . The average efuent COD concentration in

Table 1
Inuent parameters (mean standard deviation) in constructed wetland.
Parameters

COD (mg L1 )

TN (mg L1 )

NH4 + -N (mg L1 )

NO3 -N (mg L1 )

NO2 -N (mg L1 )

pH

T ( C)

Inuent

184.6 33.1

23.8 2.6

21.7 2.8

0.25 0.17

0.02 0.10

7.2 0.3

24.7 0.3

F. Li et al. / Ecological Engineering 68 (2014) 9096

93

Fig. 3. (a) COD efuent concentrations among six HSSF CWs. Error bar are standard
deviations of the mean. (b) COD removal efciencies among six HSSF CWs (along
the ow direction).

CW1 during the operational period was 11.8 mg L1

while the average efuent COD was 50.6 mg L1 in CW2 without aeration during
the same period. This nding indicates that HSSF CWs with alike
structures had signicant differences in COD removal (p < 0.01)
because regardless of aeration. The same results, which the average
COD removal of aerated CWs outperformed non-aerated CWs, were
observed in concentrations of 32.7 mg L1 for CW3 and 52.9 mg L1
for CW4, as well as in 35.8 mg L1 for CW5 and 44.4 mg L1 for
CW6. The average COD removal efciencies of six different HSSF
CWs all exceeded 65%, that is, 90.1%, 69.4%, 72.8%, 65.4%, 76.5%,
and 70.3% for CW1, CW2, CW3, CW4, CW5, and CW6, respectively. The aerated wetlands (CW1, CW3, and CW5) outperformed
the non-aerated wetlands in the removal of organics, indicating
that articial aeration can improve COD removal efciency. Similar results were also demonstrated by Ouellet-Plamondon et al.
(2006), who tested the contribution of articial aeration on pollutant removal with a combination of planted, unplanted, aerated, and
non-aerated mesocosms in treating reconstituted sh farm efuent. Results showed that the COD removal in aerated mesocosms
was more effective than that in non-aerated mesocosms in both
planted and unplanted units.
The effects of aeration sites on COD removal rates along
the direction of the water ow are shown in Fig. 3(b). COD
was effectively reduced at the front areas of the CW systems
and then steadily improved along the ow direction. The fast
degradation rate of organics might be attributed to the easily
biodegraded sucrose as carbon source and the relatively high temperature (Table 1), which promoted the decomposition of organic

Fig. 4. (a) NH4 + -N removal efciencies among six HSSF CWs. (b) NH4 + -N removal
efciencies among six HSSF CWs (along ow direction).

compounds (Fan et al., 2012; Kayser and Kunst, 2005). CW1 (front
and rear aeration) facilitated COD removal more effectively than
CW2, CW4, CW6 (non-aeration), and CW3 (rear aeration) did, indicating that aeration in the front part of HSSF CW had a better effect
on the COD removal efciency compared with middle-aeration

Fig. 5. TN removal efciencies among six HSSF CWs.

94

F. Li et al. / Ecological Engineering 68 (2014) 9096

(CW5) and rear-aeration (CW3) (p < 0.01). The intensity of the

metabolic activities of organics degradation microorganisms is
largely dependent on oxygen supply. Consequently, sufcient oxygen supply by articial aeration would favor microbial activities,
resulting in high COD removal.
3.3. Removal efciency of NH4 + -N by six HSSF CWs
The effects of aeration on the removal of NH4 + -N in HSSF CWs
were investigated, and the results are shown in Fig. 4(a). The

NH4 + -N removal performance in the aerated CWs (CW1, CW3,

and CW5) was signicantly higher than that in non-aerated CWs
(CW2, CW4, and CW6) (p < 0.01), while there is no signicant difference among the NH4 + -N removal in the aerated CWs (p > 0.05).
The NH4 + -N removal efciency levels of CW1, CW3, and CW5 averaged nearly 100% and remained quite stable during the whole
experiment period. On the contrary, NH4 + -N removal in the nonaeration systems (CW2, CW4, CW6) were far from satisfactory, as
considerable NH4 + -N accumulation could be found in the efuent
as illustrated in Fig. 6(b), (d) and (f). There was a steep decline

Fig. 6. Variations of NH4 + -N, NO3 -N, and NO2 -N in the HSSF CWs (along ow direction). (a) CW1; (b) CW2; (c) CW3; (d) CW4; (e) CW5; (f) CW6.

F. Li et al. / Ecological Engineering 68 (2014) 9096

in NH4 + -N removal at the non-aerated CWs between day 14 and

day 21 as shown in Fig. 4(a), and as much as minus 20.5% NH4 + N removal for CW6 was also observed. This phenomenon may be
unexpected; nevertheless, it is not an exception in this experiment.
Similarly, Akratos and Tsihrintzis (2007) observed even 63.8%
NH4 + -N removal in an unplanted wetland tank with medium gravel
as media, and much uctuated NH4 + -N removal occasionally was
observed within a small period time. The result in this study,
together with the result in the study of Akratos and Tsihrintzis
(2007), indicated that nitrogen removal in constructed wetland
may be quite complicated and unstable CW performance may occasionally occur for a small time period. The 20.5% NH4 + -N removal
may be attributed to the following reasons. (1) NH4 + -N was converted from organic nitrogen, which was in the form of peptone
and beef extract in synthetic domestic wastewater (Section 2.2),
via ammonication and consumption of a large amount of DO at
the same time. (2) Nitrication was remarkably limited due to the
limited DO available (Fig. 2). The nitrifying bacteria obtained energy
from oxidizing NH4 + -N and used oxygen as an electron acceptor.
However, most of DO was consumed by ammonication and organics degradation, limited DO thus hindered nitrication, resulting in
NH4 + -N accumulation (Sun et al., 2005).
The location of aeration had a great effect on NH4 + -N removal
along the direction of the water ow, as shown in Fig. 4(b). Aerated HSSF CWs (CW1, CW3, and CW5) outperformed non-aerated
HSSF CWs (CW2, CW4, and CW6). The NH4 + -N removal of aerated
HSSF CWs distinctly differed from non-aerated HSSF CWs in terms
of DO supply. DO was reduced in non-aerated HSSF CWs, accompanied by the degradation of organics (COD). Thus, the initial low DO
(Fig. 2) could not meet the DO demand for nitrication. Similarly,
the NH4 + -N removal in CW3 with rear-aeration was unsatisfactory
at a level below 10% for the two-thirds part of the bed. By contrast,
the NH4 + -N removal efciency in CW3 reached nearly 100% as the
DO concentration increased due to articial aeration (Fig. 2). The
strategy of front-aerations in CW1 and CW5 ensure sufcient DO
and improved the oxidation of wetland matrix, ending with nearly
complete NH4 + -N removal in the efuent. Additional articial aeration appeared to be an effective alternative to improve substantially
the oxidized condition within the bed, which is in accordance with
a previous study (Maltais-Landry et al., 2009). On the basis of the
ultimate efuent NH4 + -N concentration, aeration strategies (frontaeration, middle-aeration, and rear-aeration) seemed to have no
difference (p > 0.05) among each other.
3.4. Total nitrogen removal
Generally, the transformation and removal of nitrogen in CWs
could be accomplished by ammonication (Amino acids NH4 N), classical route of nitrication (NH4 -N NO2 -N NO3 -N),
and classical denitrication (NO3 -N NO2 -N NO N2 O N2 )
(Vymazal, 2007). Nitrication is the second step of nitrogen transformation if the wastewater is primarily composed of organic
nitrogen. However, if NH4 + -N predominates, the nitrication process becomes the rst step to convert the existing form (Saeed
and Sun, 2012). In this study, the organic N only account for a
small part of the inuent TN, whereas NH4 + -N account for most
of the inuent TN (Table 1). Besides, it has to mention that peptone
and beef extract where the organic N comes from are both easily
biodegradable. Hence, the total removal of nitrogen is equivalent
to the abatement of NH4 + -N, NO2 -N, and NO3 -N.
As shown in Fig. 5, CW1 with aeration showed the highest TN
removal efciency among the six CWs. On the other hand, CW6
without articial aeration nor anoxic treatment performed the
most inefcient TN removal. The results revealed that articial
aeration and/or anoxic treatment by sealing some compartments

95

Fig. 7. Correlation plot between COD/NO3 -N and efuent TN concentration in CW.

indeed have an impact on TN removal. By supplying sufcient DO

for nitrication, articial aeration could highly improve NH4 + -N
removal in aerated CWs (Fig. 4(a), and (b)); but compared to CW1,
TN removal in CW3 and CW5 did not benet much from aeration
(Fig. 5). Even for CW1, TN removal averaged only 51.3% (Fig. 5),
though NH4 + -N removal nearly reached 100% (Fig. 4(a)). This result
can be caused by the unsatisfactory denitrication even though
nitrication was performed well. Fig. 6 illustrates the variation of
NH4 + -N, NO2 -N, and NO3 -N along the ow direction to determine the extent of nitrication and denitrication in HSSF CWs. As
indicated in Fig. 6(a), (c), and (e), efuent NH4 + -N of aerated CWs
decreased to a very low level, that is, 0.27 mg L1 , 0.36 mg L1 , and
0.04 mg L1 for CW1, CW3, and CW5, respectively, indicating that
nitrication was well performed. However, the levels of NH4 + -N
in CW2, CW4, and CW6 were all high, indicating that nitrication
was not well performed. The remarkable difference of TN removal
between aerated HSSF CWs and non-aerated HSSF CWs could be
attributed to the DO available in the matrix (Fig. 2).
Lack of biodegradable organics often hinders classic denitrication metabolism (due to dependency on organic carbon) in
wetland systems, resulting in unsatisfactory TN removal though
nitrication process is guaranteed by sufcient oxygen supply. To
determine the cause of the low TN removal even though the oxygen distribution is appropriate, the correlation between efuent TN
concentration and COD/NO3 -N value in CW1 was determined. As
observed in Fig. 7, it can be found that low efuent TN concentration
coincides with high values of COD/NO3 -N. The negative correlation coefcient of 0.78 strongly conrmed that COD/NO3 -N affects
denitrication. Thus, denitrication depends on the availability of
organic carbon, which is in agreement with previous studies (Her
and Huang, 1995; Kuschk et al., 2003; Schipper et al., 1993). Generally, it is inuent COD that play an important role in organic carbon
supply for denitricaton. Appropriate C/N ratio could be adjusted
by controlling of the carbon or nitrogen source concentration to
achieve the optimal nitrogen removal (Fan et al., 2013; Zhao et al.,
2010). Saeed and Sun, (2012) studied the inuence of organic C
availability (organic coco-peat) on denitrication, results showed
that higher efuent COD concentration coincided with lower ratio
of efuent and inuent NO3 concentration (Ce /Ci ), conrming the
contribution of internal C (leaching from the organic coco-peat
media) to facilitate NO3 reduction in VF wetland. Therefore, TN
removal in CW1 can be improved as long as organic substance is
available.

96

F. Li et al. / Ecological Engineering 68 (2014) 9096

4. Conclusions
The performances of different types of multi-stage aerobicanoxic HSSF CWs were investigated to determine the effects of
articial aeration and aeration location on the treatment of organics
and nitrogen. The results show that articial aeration can establish
aerobic conditions in HSSF CWs. High COD and NH4 + -N removal
was simultaneously obtained in the aerated HSSF CWs, which
remarkably outperformed the non-aeration HSSF CWs. For example, COD, NH4 + -N, and TN removal efciency levels were 70.3%,
18.4%, and 24.6%, respectively, in CW6, and 90.1%, 99.7%, and
51.3%, respectively, in CW1. Aeration in the front end was found to
improve organics and nitrogen removal better than middle aeration
and rear aeration. With proper application, the aeration strategy
could help increase the application and popularity of HSSF CWs
in the future. In addition, low COD/NO3 -N may restrain denitrication though nitrication performed well given sufcient oxygen
supply. Thus, TN removal may still be improved as long as proper
DO distribution and sufcient carbon are available.
Acknowledgements
This study was supported by Major Science and Technology Program for Water Pollution Control and Treatment
(2012ZX07203004) and Program for New Century Excellent Talents
in University (NCET-11-0468).
References
Akratos, C.S., Tsihrintzis, V.A., 2007. Effect of temperature, HRT, vegetation and
porous media on removal efciency of pilot-scale horizontal subsurface ow
constructed wetlands. Ecol. Eng. 29, 173191.
Babatunde, A.O., Zhao, Y.Q., ONeill, M., OSullivan, B., 2008. Constructed wetlands
for environmental pollution control: a review of developments, research and
practice in Ireland. Environ. Int. 34, 116126.
Brix, H., 1994. Use of constructed wetlands in water pollution control: historical
development, present status, and future perspectives. Water Sci. Technol. 30,
209224.
Butterworth, E., Dotro, G., Jones, M., et al., 2013. Effect of articial aeration on tertiary
nitrication in a full-scale subsurface horizontal ow constructed wetland. Ecol.
Eng. 54, 236244.
Changming, L., Jingjie, Y., Kendy, E., 2001. Groundwater exploitation and its
impact on the environment in the North China Plain. Water. Int. 26,
265272.
Cooper, P., 2005. The performance of vertical ow constructed wetland systems with
special reference to the signicance of oxygen transfer and hydraulic loading
rates. Water Sci. Technol. 51, 8190.
Cooper, P., 2009. What can we learn from old wetlands? Lessons that have been
learned and some that may have been forgotten over the past 20 years. Desalination 246, 1126.
Cottingham, P., Davies, T., Hart, B., 1999. Aeration to promote nitrication in constructed wetlands. Environ. Technol. 20, 6975.
Dong, H., Qiang, Z., Li, T., Jin, H., Chen, W., 2012. Effect of articial aeration on the performance of vertical-ow constructed wetland treating heavily polluted river
water. J. Environ. Sci. 24, 596601.
Fan, J., Liang, S., Zhang, B., Zhang, J., 2012. Enhanced organics and nitrogen
removal in batch-operated vertical ow constructed wetlands by combination
of intermittent aeration and step feeding strategy. Environ. Sci. Pollut. Res.,
18.

Fan, J., Zhang, B., Zhang, J., et al., 2013. Intermittent aeration strategy to enhance
organics and nitrogen removal in subsurface ow constructed wetlands. Bioresour. Technol. 114, 117122.
Hauck, R., 1984. Atmospheric nitrogen chemistry, nitrication, denitrication, and
their relationships. The Handbook of Environmental Chemistry, vol. 1., pp.
105127.
Her, J.J., Huang, J.S., 1995. Inuences of carbon source and C/N ratio on nitrate/nitrite
denitrication and carbon breakthrough. Bioresour. Technol. 54, 4551.
Hu, Y., Zhao, Y.Q., Zhao, X., Kumar, J.L., 2012. High rate nitrogen removal in an alum
sludge-based intermittent aeration constructed wetland. Environ. Sci. Technol.
46, 45834590.
Kadlec, R.H., Wallace, S., 2009. Treatment Wetlands, 2nd ed. Taylor & Francis Group,
FL, USA.
Kayser, K., Kunst, S., 2005. Processes in vertical-ow reed beds, nitrication, oxygen
transfer and soil clogging. Water Sci. Technol. 51, 177.
Kuschk, P., Wiessner, A., Kappelmeyer, U., Weibrodt, E., Kstner, M., Stottmeister,
U., 2003. Annual cycle of nitrogen removal by a pilot-scale subsurface horizontal ow in a constructed wetland under moderate climate. Water Res. 37,
42364242.
Lee, C., Fletcher, T.D., Sun, G., 2009. Nitrogen removal in constructed wetland systems. Eng. Life Sci. 9, 1122.
Maltais-Landry, G., Maranger, R., Brisson, J., 2009. Effect of articial aeration and
macrophyte species on nitrogen cycling and gas ux in constructed wetlands.
Ecol. Eng. 35, 221229.
National Bureau of Statistics of China, 2013. BRICS Joint Statistical Manual. China
Statistics Press, Beijing, pp. 73.
Noorvee, A., Poldvere, E., Mander, U., 2007. The effect of pre-aeration on the purication processes in the long-term performance of a horizontal subsurface ow
constructed wetland. Sci. Total Environ. 380, 229236.
Ongley, E.D., Xiaolan, Z., Tao, Y., 2010. Current status of agricultural and rural nonpoint source pollution assessment in China. Environ. Pollut. 158, 11591168.
Ouellet-Plamondon, C., Chazarenc, F., Comeau, Y., Brisson, J., 2006. Articial aeration
to increase pollutant removal efciency of constructed wetlands in cold climate.
Ecol. Eng. 27, 258264.
Reddy, V.R., Behera, B., 2006. Impact of water pollution on rural communities: an
economic analysis. Ecol. Econ. 58, 520537.
Rousseau, D.P., Vanrolleghem, P.A., De Pauw, N., 2004. Model-based design of horizontal subsurface ow constructed treatment wetlands: a review. Water Res.
38, 14841493.
Saeed, T., Sun, G., 2012. A review on nitrogen and organics removal mechanisms in
subsurface ow constructed wetlands, dependency on environmental parameters, operating conditions and supporting media. J. Environ. Manage. 112,
429448.
Schipper, L., Cooper, A., Harfoot, C., Dyck, W., 1993. Regulators of denitrication in
an organic riparian soil. Soil Biol. Biochem. 25, 925933.
Shao, W., 2010. Effectiveness of water protection policy in China, a case study of
Jiaxing. Sci. Total Environ. 408, 690701.
Standard Method for the Examination of Water and Wastewater Editorial Board,
2002. Standard Method for the examination of water and wastewater, 4th ed.
Environmental Science Press of China, Beijing, China.
Sun, G., Zhao, Y., Allen, S., 2005. Enhanced removal of organic matter and
ammoniacal-nitrogen in a column experiment of tidal ow constructed wetland
system. J. Biotechnol. 115, 189197.
Tanner, C.C., Kadlec, R., 2003. Oxygen ux implications of observed nitrogen removal
rates in subsurface-ow treatment wetlands. Water Sci. Technol. 48, 191198.
Truu, M., Juhanson, J., Truu, J., 2009. Microbial biomass, activity and community
composition in constructed wetlands. Sci. Total Environ. 407, 39583971.
Vymazal, J., 1996. The use of subsurface-ow constructed wetlands for wastewater
treatment in the Czech Republic. Ecol. Eng. 7, 114.
Vymazal, J., 2007. Removal of nutrients in various types of constructed wetlands.
Sci. Total Environ. 380, 4865.
Wang, M., Webber, M., Finlayson, B., Barnett, J., 2008. Rural industries and water
pollution in China. J. Environ. Manage. 86, 648659.
Webber, M., Barnett, J., Finlayson, B., Wang, M., 2008. Pricing Chinas irrigation water.
Global Environ. Change. 18, 617625.
Zhao, Y., Liu, B., Zhang, W., Ouyang, Y., An, S., 2010. Performance of pilot-scale
vertical-ow constructed wetlands in responding to variation in inuent C/N
ratios of simulated urban sewage. Bioresour.Technol. 101, 16931700.