Chemosphere 64 (2006) 497–502

www.elsevier.com/locate/chemosphere

Technical Note

Application of a constructed wetland for industrial

wastewater treatment: A pilot-scale study
T.Y. Chen a, C.M. Kao a,*
, T.Y. Yeh b, H.Y. Chien a, A.C. Chao c

a
Institute of Environmental Engineering, National Sun Yat-Sen University, Kaohsiung 80424, Taiwan
b
Department of Civil and Environmental Engineering, National University of Kaohsiung, Kaohsiung, Taiwan
c
Department of Civil, Construction and Environmental Engineering, North Carolina State University, Raleigh, NC, USA

Received 27 June 2005; received in revised form 26 November 2005; accepted 30 November 2005
Available online 18 January 2006

Abstract

The main objective of this study was to examine the efficacy and capacity of using constructed wetlands on industrial pollutant
removal. Four parallel pilot-scale modified free water surface (FWS) constructed wetland systems [dimension for each system: 4-m
(L) · 1-m (W) · 1-m (D)] were installed inside an industrial park for conducting the proposed treatability study. The averaged influent
contains approximately 170 mg l 1 chemical oxygen demand (COD), 80 mg l 1 biochemical oxygen demand (BOD), 90 mg l 1 suspend
solid (SS), and 32 mg l 1 NH3–N. In the plant-selection study, four different wetland plant species including floating plants [Pistia strati-
otes L. (P. stratiotes) and Ipomoea aquatica (I. aquatica)] and emergent plants [Phragmites communis L. (P. communis) and Typha ori-
entalis Presl. (T. orientalis)] were evaluated. Results show that only the emergent plant (P. communis) could survive and reproduce
with a continuous feed of 0.4 m3 d 1 of the raw wastewater. Thus, P. communis was used in the subsequent treatment study. Two dif-
ferent control parameters including hydraulic retention time (HRT) (3, 5, and 7 d) and media [vesicles ceramic bioballs and small gravels,
1 cm in diameter] were examined in the treatment study. Results indicate that the system with a 5-d HRT (feed rate of 0.4 m3 d 1) and
vesicles ceramic bioballs as the media had the acceptable and optimal pollutant removal efficiency. If operated under conditions of the
above parameters, the pilot-plant wetland system can achieve removal of 61% COD, 89% BOD, 81% SS, 35% TP, and 56% NH3–N. The
treated wastewater meets the current industrial wastewater discharge standards in Taiwan.
 2006 Elsevier Ltd. All rights reserved.

Keywords: Constructed wetland; Industrial wastewater; Free water surface; Emergent plant

1. Introduction 2004). The wetland system is energetically sustainable

Over the past decade, there has been a growing appreci-
ation of the multiple values and functions of constructed
wetlands, which are increasingly used for treating a variety
of wastewaters (Green et al., 1997; Goulet et al., 2001;
Kadlec and Reddy, 2001). Moreover, use of constructed
wetlands is now recognized as an accepted eco-technology,
especially beneficial to small towns or industries that can-
not afford expensive conventional treatment systems (Jing
et al., 2001; Karathanasis et al., 2004; Karim et al.,
*
Corresponding author. Tel.: +886 7 525 4413; fax: +886 7 525 4449.
E-mail address: jkao@mail.nsysu.edu.tw (C.M. Kao).
because it uses only natural energy to reduce pollutants.
Compared with the conventional wastewater treatment sys-
tem currently in use, it only requires low construction and
low operational costs (Lin et al., 2002; Lloyd et al., 2004;
Ran et al., 2004). The constructed wetland, know as free
water surface (FWS) system, can mimic natural systems
as the water flows over the bed surface and is filtered
through a dense stand of aquatic plants (Li and Jiang,
1995; Cooper et al., 1999; Jing et al., 2002). An alternative
system, known as the subsurface flow wetland, is also a
constructed system consisting of an excavated but usually
lined shallow basin containing gravel media and emergent
aquatic plants (Gopal, 1999; Bachand and Horne, 2000;
Mashauri et al., 2000; Kao et al., 2002; Garcia et al.,

0045-6535/$ - see front matter  2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chemosphere.2005.11.069
498 T.Y. Chen et al. / Chemosphere 64 (2006) 497–502
2003; Kaseva, 2004). There are currently thousands of con-
structed wetlands worldwide receiving and treating a vari-
ety of municipal, industrial, and urban runoff wastewaters
(Green et al., 1997; Cameron et al., 2003; Solano et al.,
2004). Recently, there has been increasing interest in Tai-
wan in utilizing the constructed wetlands as an alternative
method for wastewater treatment because of its operational
simplicity and cost efficiency (Kao et al., 2001; Jing et al.,
2002; Lin et al., 2002; Lee et al., 2004). Although the con-
structed wetland technology is well established, its applica-
tion for treating specific industrial effluents has not been
well documented (Worall et al., 1997; Kao and Wu, 2001;
Garcia et al., 2004).
The recent economy recession has caused many factories
to trim down production leading to reduced wastewater
flow thus resulting in operational problems of conventional
wastewater treatment plant (WWTP) and sometimes seri-
ous violations of the effluent standards. To minimize the
operational and maintenance cost of the conventional
wastewater treatment utilities, some medium- and small-
scale factories have been found to illegally dump untreated
wastewater effluent directly into the water bodies causing
more serious environmental problems in Taiwan. Thus,
the less expensive wetland system has been proposed to
replace the existing WWTP. The industrial park selected
for this study is located in the southern part of Taiwan;
its central WWTP receives mixed wastewaters from a vari-
ety of industries (electronic, steel, chemical, food-process-
ing, etc.). Major components of the Park’s central
WWTP include bar racks, grit chamber, equalization basin,
primary sedimentation tank, aeration tank, final clarifier,
and chlorine contact chamber. The average influent chem-
ical oxygen demand (COD), biochemical oxygen demand
(BOD), and suspended solids (SS) are approximately 170,
80, and 90 mg l 1, respectively. With proper maintenance
and operation, the treatment facility is capable of treating
the Park’s wastewater effectively to meet the industrial
wastewater discharge standards in Taiwan (COD =
100 mg l 1, BOD = 30 mg l 1, and SS = 30 mg l 1). How-
ever, most factories inside the Park are closed due to recent
economic recession that causes a significant decrease of the
influent flow rate thus resulting in serious treatment prob-
lems. Because the constructed wetland is a low cost and low
risk alternative system, it is especially suitable for treating
the greatly reduced wastewater flow. The Park manage-
ment is considering incorporating a constructed wetland
system into the existing WWTP facilities to replace the sec-
ondary treatment system in order to meet the effluent stan-
dards while reducing the costs of wastewater treatment.
The main objectives of this study were to (1) examine the
efficacy and capacity of using a modified FWS constructed
wetland system on industrial wastewater treatment at this
studied industrial park, (2) determine the appropriate plant
species for the industrial wastewater treatment (plant-selec-
tion study), and (3) evaluate the optimal design factors,
e.g., plant species, type of media, and hydraulic retention
time (HRT) on the efficiency of the constructed wetland
system. Pollutant removal efficiencies were calculated from
the flow and concentration data. The removal efficiencies,
combined with information of HRT, may be useful in
the design of constructed wetlands for future practical
application.
2. Materials and methods
Four parallel pilot-scale FWS constructed wetland
tanks (labeled as T1 to T4) were built inside the industrial
park. Each FWS system consists of a continuous-flow
experimentation tank with dimensions of 4-m (L) · 1-m
(W) · 1-m (D). The pilot-scale system received and treated
the effluent from a reservoir that stores the effluent from
the Park’s WWTP equalization basin. The equalized waste-
water was delivered to the pilot system continuously at a
rate of approximately 0.4 m3 d 1 to yield a HRT of 5 d.
In the plant selection study, four different wetland plants
species including emergent plants [Phragmites communis
L. (P. communis) (reed) and Typha orientalis Presl. (cattail)]
and floating plants [Pistia stratiotes L. (P. stratiotes) (water
lettuce) and Ipomoea aquatica (I. aquatica) (water spinach)]
were planted in T1 to T4, respectively, to cover 40% of each
tank. These four plants had been successfully applied for
domestic and livestock wastewater treatment (Kao et al.,
2001; Jing et al., 2002). The tanks were then filled with ves-
icles ceramic bioballs [1 cm (D)] to a depth of 0.5 m, and
the water level was maintained at 0.8 m in all tanks. The
vesicles ceramic bioballs (Everglory Resource Technology,
Taiwan) are made of recycled materials (e.g., glass, con-
struction waste, clay) at an acceptable cost (US$50 m 3).
They have high specific surface areas (2360 m2 m 3), and
thus are appropriate for the growth of bacteria. The effects
of industrial wastewater on plant growth and reproduction
were evaluated. The selected plant species from the first
part of this study (plant-selection study) were used in the
following media selection study.
In the media selection phase, two different types of
media (1-cm (D) vesicles ceramic bioballs and 1-cm (D)
gravels) were examined. Tanks T1 and T2 have plants
but filled with different media; ceramic bioballs in T1 and
gravels in T2. Without plant growth, T3 and T4 are filled
with ceramic bioballs and gravels, respectively. The plant
grown in T1 and T2 tanks was the species selected from
the previous plant selection study. Four water quality mon-
itoring piezometers (labeled as P1–P4) was placed one in
each tank to monitor the variations in dissolved oxygen
(DO), redox potential (Eh), and pH. It had openings at
0.2–0.4 m above the bottom, and was placed at 2 m hori-
zontally from the center inlet of the tank. Moreover, other
water quality indicators (DO, Eh, and pH) were also mea-
sured at 0.1 m below the water surface in all four tanks so
that variations in DO, Eh, and pH above and below the
surface of the media could be evaluated. There were 133
samples collected for the measurements of pH, COD, and
SS. As for BOD, TP, TKN, NO3–N, NO2–N, and NH3–
N, 37 samples were collected for their analyses. The stan-
 T.Y. Chen et al. / Chemosphere 64 (2006) 497–502 499

Table 1
Averaged water quality monitoring data from the media selection study
Parameter Inlet T1 outlet T2 outlet T3 outlet T4 outlet T1a (%) T2 (%) T3 (%) T4 (%)
1 b
SS (mg l ) 90 ± 22 17 ± 3 24 ± 8 43 ± 13 48 ± 15 81 73 52 47
TP (mg l 1) 69 ± 12 45 ± 9 49 ± 10 53 ± 12 57 ± 16 35 29 23 17
NH3–N (mg l 1) 32 ± 14 14 ± 4 16 ± 5 23 ± 9 25 ± 8 56 50 34 22
BOD (mg l 1) 80 ± 23 9±3 14 ± 3 35 ± 12 39 ± 7 89 83 56 51
TKN (mg l 1) 35 ± 11 16 ± 4 19 ± 6 26 ± 8 28 ± 7 46 31 17 9
COD (mg l 1) 170 ± 36 67 ± 14 70 ± 17 110 ± 18 119 ± 24 61 59 35 30
NO3–N (mg l 1) 13 ± 4 6±2 7±3 4±1 3 ± 0.9 85 62 46 38
NO2–N (mg l 1) 0 0.3 ± 0.1 0.4 ± 0.2 0.2 ± 0.1 0.1 ± 0.1 –c – – –
a
% of removal = [(averaged inlet concentration averaged outlet concentration)/averaged inlet concentration] · 100%.
b
Mean ± standard deviation.
c
Not available.

dard deviations of the analytical results are presented in
Table 1.
Water quality analyses were conducted in accordance
with the methods described in Standard Methods (APHA,
1995). The treatment efficiency was determined using the
averaged influent and effluent concentrations of the above
major water quality parameters. Specific surface area was
measured following the procedures described in American
Society of Testing and Materials (ASTM) method
(D4820-96a). The first-order decay model for plug flow
concentration evaluation was used to determine the first-
order decay rates (d 1) for BOD and NH3–N in the wet-
land system (Kadlec and Reddy, 2001; Jing and Lin, 2004).
3. Results and discussion
In the plant-selection study, average concentrations of
the influent water quality parameters are shown in Table
1. Nitrite and all heavy metals were not detected in the
influent (data not shown). Results from the six-month
investigation show that the emergent plant (P. communis)
was the only species among the four tested plants to survive
and reproduce subject to continuous feed of 0.4 m3 d 1 of
the equalized raw wastewater. The other three plants did
not grow well in the industrial wastewater; two plants (I.
aquatica and T. orientalis Presl.) died after one month
and the growth of P. stratiotes L. was significantly inhib-
ited. High concentrations of COD and nutrient might were
thought to cause the death and growth inhibition of those
three plant species. Thus, P. communis was selected for the
following media and HRT selection experiments.
Results of the media selection study are listed in Table 1
by showing the averaged water quality monitoring data
during the one-year investigation period. The standard
deviations for the influent and effluent data are also
included. The results indicate that the wetland system using
P. communis as the wetland plant and vesicles ceramic bio-
balls as the media is effective in improving the effluent qual-
ity as seen by removal of 81% SS, 35% TP, 56% NH3–N,
89% BOD, and 61% COD, such that the effluents from
both T1 and T2 met the wastewater discharge standards.
Thus, from the regulation point of view, both T1 and T2
systems are acceptable treatment schemes. Because the
major routes for TP removal might be through plant
uptakes and precipitation, the removal for TP was only
moderate comparing with the removal efficiencies for other
water quality indicators. The low oxygen in the system
would also cause the relatively lower TP removal efficiency.
Less treatment efficiencies were observed in the systems
(T3 and T4) with media only indicating that wetland
plants play an important role in contaminant removal.
Although the effluents from T3 and T4 cannot meet the dis-
charge standards, significant amounts of contaminants were
removed in those systems in the absence of wetland plants.
This suggests that the microbial and physical/chemical
processes may also contribute to the removal of pollutants
because the media created suitable environment for
microbial growth, thus leading to the observed pollutant
removal. Therefore, biodegradation, sedimentation, and
sorption were the major pollutant removal mechanisms in
the systems with media only. Moreover, the results also
show that T3 system had slightly higher contaminant
removal efficiencies than T4 system. This might be due to
the effects that vesicles ceramic bioballs had higher effective
surface areas (2360 m2 m 3) than gravels (990 m2 m 3) that
favor microbial growth. Thus, the T3 system has more
attached biological growth than T4 for degrading the
wastewater. Although vesicles ceramic bioballs cost more
than gravels (US$50 m 3 vs. US$20 m 3), they are more
effective in removing pollutants, and thus, were selected
as the media for the HRT determination study.
Fig. 1 shows variations of the influent and effluent pH,
COD, and BOD while Fig. 2 presents variations of the
influent and effluent SS, TP, and NH3–N for T1. In this
study, 133 samples have been collected for the measure-
ments of pH, COD, and SS. As for BOD, TP, and NH3–
N, 37 samples have taken for the measurements. The calcu-
lated standard deviations for the influent are 36 mg l 1 for
COD, 22 mg l 1 for SS, 23 mg l 1 for BOD, 14 mg l 1 for
NH3–N, and 12 mg l 1 for TP. The calculated standard
deviations for the effluent in the T1 system are 14 mg l 1
for COD, 3 mg l 1 for SS, 3 mg l 1 for BOD, 4 mg l 1
for NH3–N, and 9 mg l 1 for TP. These results show that
BOD had the highest removal efficiency (89%) followed
by SS with a removal efficiency of 81% (Table 1) and more
significantly that the water quality of wetland effluent is
500 T.Y. Chen et al. / Chemosphere 64 (2006) 497–502

9.0 400
Influent Effluent Influent Effluent
350
8.0 300

250

SS (mg/l)
pH

7.0
200

150
6.0
100

50
5.0
10 35 60 85 110 135 160 185 210 0
10 35 60 85 110 135 160 185 210

300
Influent Effluent 120
Influent Effluent
250
100
200 80
COD (mg/l)

TP (mg/l)
150 60

100 40

50 20

0 0
10 35 60 85 110 135 160 185 210 10 35 60 85 110 135 160 185 210

140 60
Influent Effluent Influent Effluent
120 50
NH3–N (mg/l)

100 40
BOD (mg/l)

80 30

60 20

40 10

20 0
10 35 60 85 110 135 160 185 210
0 Days
10 35 60 85 110 135 160 185 210
Days Fig. 2. Variations in SS, TP, and NH3–N measurements for T1 system.

Fig. 1. Variations in pH, COD, and BOD measurements for T1 system.

stable and reliable because of insignificant variations in the
effluent quality observed.
Table 2 shows the average DO and Eh results and
ranges of pH values on the influent, 0.1 m below the water
surface, and at 0.2–0.4 m measured from the tank bottom
using the piezometers (P1–P4) installed one in each tank
(T1–T4). Low DO and Eh measurements in all four piez-
ometers indicate that decomposition and biodegradation
of the organic materials indicating that the pH value varied
on a daily basis. Higher DO measurements in T1 and T2
might be due to the occurrence of oxygen transportation
mechanism of plant vascular systems that would enhance
the aerobic biodegradation around the root zone area.
Higher DO and Eh values were also observed at the water
surface compared with the results obtained in the piezo-
meters (Table 2). This indicates that natural reaeration
effects at the water surface caused the significant variations
in DO and Eh above and below the surface of the media.
The occurrence of both aerobic and low oxygen conditions
suggests that the nitrification and denitrification processes
would occur simultaneously in all four tanks to result in
the observed significant decrease of both NH3–N and
NO3–N. Moreover, low NO2–N measurements (less than
0.4 mg l 1) were observed in all tanks suggesting that the
intermediate products of nitrification and denitrification
(NO2–N) were not accumulated in the systems. Results of
NH3 and TKN also indicate that organic nitrogen was
not the major nitrogen form.
In the third phase of this study, the optimal HRT was
determined for the constructed wetland systems using ves-
icles ceramic bioballs as the media and P. communis as the
wetland plant. Three different HRT (3, 5, and 7 d) were
examined and compared to evaluate the effectiveness of
HRT on contaminant removal. Table 3 shows the average
effluent concentrations and removal efficiencies for SS, TP,
NH3–N, BOD, and COD with different HRTs. The results
show that higher removal efficiencies were observed in the
wetland system with a 7-d HRT (Table 3) to achieve
removal of 83% of SS, 49% of TP, 75% of NH3–N, 95%
of BOD, and 69% of COD (Table 3). If the HRT is short-
 T.Y. Chen et al. / Chemosphere 64 (2006) 497–502 501

Table 2
Averaged DO and Eh monitoring data and ranges of pH values from four piezometers and surface water of four tanks
Parameter Inf.a P1 P2 P3 P4 S1b S2 S3 S4
pH 6–8.3 6.1–7.7 6.2–7.6 6.2–7.7 6–7.8 6.2–7.9 6.1–7.7 6.2–7.8 6.1–7.8
Eh (mV) 86 30 18 50 69 153 132 78 107
DO (mg l 1) 0.8 0.2 0.3 0.1 0.1 1.5 1.6 1.3 1.4
a
Influent.
b
Water quality parameters were measured at a 0.1 m depth below the water surface.

Table 3
Averaged effluent concentrations and removal efficiencies for SS, TP, AN, BOD, and COD with different HRT
HRT (d) NH3–N (mg l 1) %a BOD (mg l 1) % COD (mg l 1) % TP (mg l 1) % SS (mg l 1) %
b
3 21 ± 11 34 22 ± 9 73 107 ± 24 37 51 ± 12 26 27 ± 10 70
5 14 ± 4 56 9±3 89 67 ± 14 61 45 ± 9 35 17 ± 3 81
7 8±2 75 25 ± 5 94 52 ± 17 69 35 ± 5 49 15 ± 3 83
a
% of removal = [(averaged influent concentration averaged effluent concentration)/averaged inlet concentration] · 100%; the averaged influent
concentrations are shown in Table 1.
b
Mean ± standard deviation.

ened to 3 d, lower removal efficiencies for those major
water quality indicators were observed; the effluent COD
cannot meet the discharge standard of the industrial waste-
water. Thus, the HRT of the wetland system affects the
treatment efficiency, and higher efficiency can be achieved
with a longer HRT. The pilot plant test results indicate that
a minimum of 5-d HRT is sufficient for the constructed
wetland system to treat the Park’s industrial wastewater.
Although further improvement of the treatment efficiency
could be obtained with the system operated at the 7-d
HRT, the 5-d HRT is recommended for the future practical
application because of cost-effectiveness considerations.
In this study, the first-order decay model was applied to
determine the decay rate (k). The calculated k values of
0.41 d 1 for BOD and 0.18 d 1 for NH3–N are close to
those reported in literature (Reed et al., 1995; Kadlec,
1997; Wittgren and Mahlum, 1997; IWA, 2000; Goulet
et al., 2001, 2004). This indicates that the industrial waste-
water tested in this study is as treatable in the constructed
wetland systems as the domestic wastewater and high
organic-contained wastewater used in other studies (IWA,
2000; Huang et al., 2000; Goulet et al., 2001; Kao et al.,
2001; Jing et al., 2002). Moreover, the average temperature
in southern Taiwan, located in the subtropical region, dur-
ing the investigation period was approximately 28 C,
which is much higher than the averaged temperature in
North America and European countries. Since higher tem-
perature favors the biodegradation rate, the wetland sys-
tems is more suitable for treating wastewater in this
studied area. Moreover, the observed high DO values in
the system above the media level also facilitate the nitrifica-
tion process. The resulting higher removal efficiencies
(Tables 1 and 3) also indicate that the application of FWS
constructed wetland system is an acceptable alternative
treatment scheme for the industrial wastewater treatment.
Results from this study also suggest the following activ-
ities to enhance the effectiveness of pollutant removal in the
future field application: (1) frequent plant harvesting to
maintain a vegetation with high metabolic uptake of nutri-
ents and pollutants, (2) effective pest control, (3) proper
preliminary treatment to prevent the accumulation of
sludges, (4) application of chemical precipitation process
for removing heavy metal from the influent prior to its dis-
charge into the constructed wetland system if significant
quantities of heavy metals are present in the wastewater,
(5) addition of an upstream oxidation pond for high nutri-
ent-containing wastewater, and (6) density control of the
vegetation to maintain its healthy growth.
4. Conclusions
The emergent wetland plant P. communis has been
shown to survive and reproduce well in a pilot-scale con-
structed wetland for receiving and treating a continuous
industrial wastewater input from an industrial park in
southern Taiwan. The pilot-scale wetland system that uses
vesicles ceramic bioballs as the media and P. communis as
the wetland plant achieves the optimal results if it is oper-
ated at a 5-d HRT. The overall removal efficiencies are 61%
for COD, 89% for BOD, 81% for SS, 35% for TP, and 56%
for NH3–N such that the treated wastewater meets the cur-
rent industrial wastewater discharge standards in Taiwan.
Results from this study indicate that a reliable and stable
effluent can be obtained using the constructed wetland sys-
tem for treating industrial wastewater treatment. From the
viewpoint of sustainable development, the vesicles ceramic
bioballs, which are made of recycled materials with accept-
able cost, are especially desirable as the media for similar
constructed wetland systems to be constructed in the
future.
Results from the pilot-scale study also indicate that the
constructed wetland system is a feasible and cost-effective
alternative technology to replace traditional secondary bio-
logical system for treating industrial wastewater to meet
502 T.Y. Chen et al. / Chemosphere 64 (2006) 497–502
the discharge standards currently enforced in Taiwan.
Thus, the constructed wetland scheme has the potential
to be developed into an environmentally and economically
acceptable industrial wastewater treatment technology for
developing countries. Results from this study will be useful
to assist environmental professionals in designing a scale-
up system for practical application.
Acknowledgments
This study was funded in part by Taiwan’s National Sci-
ence Council and Environmental Protection Bureau of Tai-
nan County, Taiwan. Additional thanks to Dr. M.S. Lee
and Ms. J.L. Hong of Environmental Protection Bureau
of Tainan County, Taiwan and Mr. E.H. Lai of Ping-
Nan Industrial Park, Taiwan for their assistance through-
out this project.
References
APHA (American Public Health Association), 1995. Standard Methods
for the Examination of Water and Wastewater, 19th ed. APHA-
AWWA-WEF, Washington, DC.
Bachand, P.A., Horne, J.A., 2000. Denitrification in constructed free-
water surface wetland. II. Effects of vegetation and temperature. Ecol.
Eng. 14, 17–32.
Cameron, K., Madramootoo, C., Crolla, A., Kinsley, C., 2003. Pollutant
removal for municipal sewage lagoon effluents with a free-surface
wetland. Water Res. 37, 2803–2812.
Cooper, P., Griffin, P., Humphries, S., Pound, A., 1999. Design of a
hybrid reed bed system to achieve complete nitrification and denitri-
fication of domestic sewage. Water Sci. Technol. 40, 283–289.
Garcia, J., Ojeda, E., Sales, E., Chico, F., Piriz, T., Aguirre, P., Mujeriego,
R., 2003. Spatial variations of temperature, redox potential, and
contaminants in horizontal flow reed beds. Ecol. Eng. 21, 129–142.
Garcia, J., Aguirre, P., Mujeriego, R., Huang, Y., Ortiz, L., Bayona, J.,
2004. Initial contaminant removal performance factors in horizontal
flow reed beds used for treating urban wastewater. Water Res. 38,
1669–1678.
Gopal, B., 1999. Natural and constructed wetland for wastewater
treatment—potentials and problems. Water Sci. Technol. 40, 27–35.
Goulet, R.R., Pick, F.R., Droste, R.L., 2001. Test of the first-order
removal model for metal retention in a young constructed wetland.
Ecol. Eng. 17, 357–371.
Green, M.B., Griffin, P., Seabridge, J.K., Dhobie, D., 1997. Removal of
bacteria in subsurface flow wetland. Water Sci. Technol. 35, 109–116.
Huang, J., Reneau Jr., R.B., Hagedorn, C., 2000. Nitrogen removal in
constructed wetlands employed to treat domestic wastewater. Water
Res. 34, 2582–2588.
IWA Special Group on Use of Macrophytes in Water Pollution Control,
2000. Constructed wetlands for pollutant control. Scientific and
Technology Report No. 8. IWA Publishing, London, England.
Jing, S.R., Lin, Y.F., 2004. Seasonal effect on ammonia nitrogen removal
by constructed wetland treating polluted river water in southern
Taiwan. Environ. Pollut. 127, 291–301.
Jing, S.R., Lin, Y.F., Lee, D.Y., Wang, T.W., 2001. Nutrient removal
from polluted river water by using constructed wetlands. Bioresour.
Technol. 76, 131–135.
Jing, S.R., Lin, Y.F., Lee, D.Y., Wang, T.W., 2002. Microcosm wetland
for wastewater treatment with different hydraulic loading rates and
macrophytes. J. Environ. Qual. 31, 690–696.
Kadlec, R.H., 1997. Determination and stochastic aspects of constructed
wetland performance and design. Water Sci. Technol. 35, 149–156.
Kadlec, R.H., Reddy, K.R., 2001. Temperature effects in treatment
wetlands. Water Environ. Res. 73, 543–557.
Kao, C.M., Wu, M.J., 2001. Control of non-point source pollution by a
natural wetland. Water Sci. Technol. 43, 169–174.
Kao, C.M., Wang, J.Y., Lee, H.Y., Wen, C.K., 2001. Application of a
constructed wetland for non-point source pollution control. Water Sci.
Technol. 44, 585–590.
Kao, C.M., Wang, J.Y., Chen, K.F., Lee, H.Y., Wu, M.J., 2002. Non-
point source pesticide removal by a mountainous wetland. Water Sci.
Technol. 46, 199–206.
Karathanasis, A., Potter, C., Coyne, M., 2004. Vegetation effects on fecal
bacteria, BOD, and suspended solid removal in constructed wetlands
treating domestic wastewater. Ecol. Eng. 20, 157–169.
Karim, M., Manshadi, F., Karpiscak, M., Gerba, C., 2004. The
persistence and removal of enteric pathogens in constructed wetlands.
Water Res. 38, 1831–1837.
Kaseva, M., 2004. Performance of a sub-surface flow constructed wetland
in polishing pre-treated wastewater—a tropical case study. Water Res.
38, 681–687.
Lee, C.Y., Lee, C.C., Lee, F.Y., Tseng, S.K., Liao, C.J., 2004.
Performance of subsurface flow constructed wetland taking pretreated
swine effluent under heavy loads. Bioresour. Technol. 92, 173–179.
Li, X., Jiang, D., 1995. Constructed wetland systems for water pollution
control in North China. Water Sci. Technol. 32, 349–356.
Lin, Y.F., Jing, S.R., Lee, D.Y., Wang, T.W., 2002. Nutrient removal
from aquaculture wastewater using a constructed wetlands system.
Aquaculture 209, 169–184.
Lloyd, J., Klessa, D., Parry, D., Buck, P., Brown, N., 2004. Stimulation of
microbial sulfate reduction in a constructed wetland: microbiological
and geochemical analysis. Water Res. 38, 1822–1830.
Mashauri, D.A., Mulungu, D.M., Abdulhussein, B.S., 2000. Constructed
wetland at the university of Dar Es Salaam. Water Res. 34, 1135–1144.
Ran, N., Agami, M., Oron, G., 2004. A pilot study of constructed
wetlands using duckweed for treatment of domestic primary effluent in
Israel. Water Res. 38, 2241–2248.
Reed, S.C., Crites, R.W., Middlebrooks, E.J., 1995. Natural Systems for
Waste Management and Treatment, second ed. McGraw-Hill Inc.,
New York.
Solano, M., Soriano, P., Ciria, M., 2004. Constructed wetlands as a
sustainable solution for wastewater treatment in small villages. Biosyst.
Eng. 87, 109–118.
Wittgren, H.B., Mahlum, T., 1997. Wastewater treatment wetlands in cold
climates. Water Sci. Technol. 35, 45–53.
Worall, P., Peberdy, K.J., Millett, M.C., 1997. Constructed wetland and
natural conservation. Water Sci. Technol. 3, 205–213.