Ecological Engineering 120 (2018) 116–125

Contents lists available at ScienceDirect

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

Performance of ornamental plants in monoculture and polyculture T

horizontal subsurface flow constructed wetlands for treating wastewater
⁎
Ana María Leiva, Romina Núñez, Gloria Gómez, Daniela López, Gladys Vidal
Engineering and Biotechnology Environmental Group, Environmental Science Faculty & Center EULA–Chile, University of Concepción, P.O. Box 160-C, Concepción, Chile

A R T I C LE I N FO A B S T R A C T

Keywords: The aim of this study was to evaluate the effect of two ornamental plants in monoculture and polyculture
Cyperus papyrus horizontal subsurface flow (HSSF) constructed wetlands for treating wastewater. Two pilot-scale HSSF systems
Zantedeschia aethiopica each with a surface area of 4.5 m2 were operated over two years (2015 and 2016); a) one was planted with a
Biomass production mixture of Cyperus papyrus and Zantedeschia aethiopica (HSSF-Cyp/Zant), and b) the other was planted only with
Nutrient uptake
Cyperus papyrus (HSSF-Cyp). To compare the performance between monoculture and polyculture systems, in situ
Plant competition
parameters and organic matter (chemical oxygen demand (COD) and biological oxygen demand (BOD5)), total
suspended solids (TSS), nutrients (total nitrogen (TN) and total phosphorus (TP)) and pathogens (fecal coliform
(FC) and total coliform (TC)) removal efficiencies were evaluated. Moreover, growth characteristics, biomass
production and nutrients uptake of different plants used in HSSF systems were also compared. The removal
efficiencies of organic matter, suspended solids, nutrients and pathogens during the operational years were
above 60%, 90%, 10% and 1.8 Log most probable number (MPN)/100 mL, respectively, without significant
differences between HSSF-Cyp/Zant and HSSF-Cyp. The biomass production and the density of both HSSF
systems fluctuated between 19.7 and 21.5 kg dry weight (DW)/m2 and 454–684 individuals/m2. Regarding the
nutrient content of different plants used, Zantedeschia aethiopica, which was planted in the polyculture system,
had the highest TN and TP content in all plants tissues (59.6 g N/kg·DW and 8.28 g P/kg·DW, respectively).
However, TN and TP mass balances determined that the effect of monoculture and polyculture systems was not
significant. Despite these results, polyculture CWs represent a good alternative of treatment system because they
provide social benefits to the community such the improving of the system landscape and a better habitat
quality. Moreover, some authors reported that polyculture system enhance the resistance to environmental stress
and disease and the system landscape.

1. Introduction landscape and their commercial values through the production of

Constructed wetlands (CWs) are engineered systems designed for
treating industrial and domestic wastewater. They are considered an
attractive solution to use in rural areas with a population of up to 2000
equivalent habitants (Carballeira et al., 2016; Vymazal, 2011). The
advantages of this technology are their simplicity, good performance
and maintenance cost between 13 and 101 USD/hab (Vera et al., 2011).
Plants are considered to be an essential component of CWs. Their
most important roles are related to their physical effects, the uptake of
nutrients, the release of oxygen to the rhizosphere and the micro-
organism hosting (Brix, 1997; Shelef et al., 2013). Phragmites spp.,
Typha spp. and Schoenoplectus spp. are the most frequent plant species
used in these systems (Vymazal, 2011). However, it is known that the
use of ornamental plants in CWs is an attractive alternative for small
communities due to their capacity to improve the treatment system
flowers and fibers (Zurita et al., 2009). In fact, these systems achieve
removal efficiencies of total suspended solids (TSS), chemical oxygen
demand (COD) and biological oxygen demand (BOD5) of over 60%
(Burgos et al., 2017; Calheiros et al., 2015; Zurita et al., 2011). In ad-
dition, ammoniacal nitrogen (NH4+-N) and total phosphorus (TP) re-
moval efficiencies have been documented in the ranges of 48–55% and
24–50%, respectively (Belmont and Metcalfe, 2003; Merino-Solís et al.,
2015). In the cases of fecal coliform (FC) and total coliform (TC), some
studies have reported removal efficiencies above 95% (Abou-Elela and
Hellal, 2012). These results suggest that it is possible to use ornamental
plants in CWs without reducing the efficiency of treatment system.
One aspect that has been controversial is the effect of monoculture
and polyculture in CWs. A comparative study between a monoculture
system planted with Canna indica and a polyculture system planted with
Canna indica and Lythrum salicaria showed that removal efficiencies of

⁎
Corresponding author.
E-mail address: glvidal@udec.cl (G. Vidal).

https://doi.org/10.1016/j.ecoleng.2018.05.023
Received 7 January 2018; Received in revised form 7 April 2018; Accepted 15 May 2018
Available online 01 June 2018
0925-8574/ © 2018 Elsevier B.V. All rights reserved.
A.M. Leiva et al. Ecological Engineering 120 (2018) 116–125

COD were 1.2 times higher in the polyculture system (Zhou et al., Table 1
2017). This difference was attributed to temporal and spatial com- Operational conditions and design parameters of HSSF-Cyp/Zant and HSSF-
pensation, root distribution and nutrient preferences that predominated Cyp.
in polyculture configurations (Karathanasis et al., 2003). However, Operational parametersa
Liang et al. (2011) observed that the monoculture wetland had sig-
nificantly higher COD and NH4+-N removal rates than the polyculture Year Season OLR HLR HRT (d) ET P
(gBOD5/m2·d) (mm/d) (mm/d) (mm/d)
wetland during the first year of operation (p < 0.05). Another com-
parative study reported that monoculture and polyculture wetlands 2015 F/W 5.7 ± 1.8 26.7 ± 2.0 6 1.5 2.3
planted with Canna indica and Schoenoplectus validus achieved NH4+-N S/Sm 4.1 ± 0.5 28.9 ± 1.5 6 3.5 1.3
and phosphate (PO4-3-P) removal efficiencies above 90% and 70%, 2016 F/W 5.2 ± 2.0 24.4 ± 2.3 6 1.5 3.5
S/Sm 5.1 ± 2.3 22.2 ± 3.2 5 3.5 0.6
respectively, without a significant difference (p > 0.05) (Zhang et al.,
2007). Regarding the effect of these systems on plant growth char- a
Operational and design parameters are the same for both constructed
acteristics, Liang et al. (2011) also found that a polyculture wetland had wetlands. OLR: organic loading rate; HLR: hydraulic loading rate; HRT: hy-
a density between 85.6 and 123.8 individuals/m2, whereas the mono- draulic retention time; ET: evapotranspiration; P: precipitation; F/W: Fall/
culture presented a density between 56 and 86.3 individuals/m2. Winter season and S/Sm: Spring/Summer season.
Nonetheless, during the first year, the monoculture wetland had a
larger biomass (2.0 kg of dry weight (DW)/m2) compared to the poly- Fig. 1a) shows a schematic diagram of the system. After the distribution
culture system (1.7 kg·DW/m2). In this same study, this behavior tank, the influent was conducted to two parallel HSSF CWs. One of
changed during the latter three years, where biomass production was them was planted with Cyperus papyrus (HSSF-Cyp), and the other was
1.2–1.7 times higher in the polyculture system. These results indicate planted with a mixture of Cyperus papyrus (11 plantlets) and Zante-
that the differences between monoculture and polyculture wetlands are deschia aethiopica (7 plantlets) (HSSF-Cyp/Zant). These two ornamental
still unclear. To improve the performance of polyculture in constructed plant species are commonly used in CWs applications (Vymazal, 2011).
wetlands, some authors suggest a good plant selection for avoiding In addition, Fig. 1b) explained the cross section of each pilot-scale
species competition that may affect nutrient removal and vegetation HSSF, which was divided into three zones separated by three sampling
stability in CWs (Zhang et al., 2007). tubes: Zone A (the inlet zone), 0.65 m from the inlet; Zone B (the middle
Taking the above into account, the aim of this study was to compare zone), 1.4 m from the inlet; and Zone C (the outlet zone), 2.25 m from
the effect of two ornamental plants, Zantedeschia aethiopica and Cyperus the inlet. The surface area of each zone was 1.5 m2 (López et al., 2015).
papyrus, in monoculture and polyculture horizontal subsurface flow Table 1 summarizes the operational conditions and design para-
(HSSF) CWs for treating wastewater. meters of HSSF systems. Each HSSF constructed wetland had an area of
4.5 m2, a water level of 0.4 m, an average depth of 0.57 m and a the-
2. Materials and methods oretical volume of 1.28 m3. The support medium used was gravel with a
size of 19–25 mm and a porosity of 0.6% (Sepúlveda-Mardones et al.,
2.1. Design of HSSF systems 2017). The gravel used in this study has the same characteristics de-
scribed by Andrés et al. (2018). This support medium was composed by
The wetland system consists in a HSSF pilot plant located in Hualqui quartz (SiO2), muscovite mica (KAl2(Si3Al)O10(OH)2) and feldspar
(36°59′26.93″ south latitude and 72°56′47.23″ west longitude), Biobío ((K,Na,Ca,Ba,NH4)(Si,Al)4O8). During the operation, the organic
Region, Chile. The influent used corresponded to the wastewater from a loading rates (OLRs) were in the range of 4.1–5.7 gBOD5/m2 d during
rural community of 20,000 inhabitants. The wastewater was subjected 2015–2016. For the hydraulic loading rates (HLRs) and the hydraulic
to a pre-treatment and then to a primary treatment, which were de- retention times (HRTs), these values varied between 22.2 and 28.9 mm/
scribed by López et al. (2015) and Sepúlveda-Mardones et al. (2017). d and 5–6 d, respectively. The evapotranspiration (ET) presented

Fig. 1. a) Schematic diagram of pilot-scale wastewater treatment plant using horizontal subsurface flow constructed wetlands (HSSFs). b) Cross section of each pilot-
scale HSSF showing dimensions and in situ measurement points (cylindrical tubes).

117
A.M. Leiva et al. Ecological Engineering 120 (2018) 116–125

Table 2
Physicochemical characterization of the influent of HSSF-Cyp/Zant and HSSF-Cyp.
Years

2015 2016

Parameters Units F/W S/Sm F/W S/Sm

BOD5 mg/L 215.3 ± 67.2 142.9 ± 16.6 210.7 ± 81.2 229.0 ± 102.1
COD mg/L 297.6 ± 97.2 278.5 ± 90.4 262.4 ± 93.9 342.6 ± 82.1
TSS mg/L 412.8 ± 380.3 257.3 ± 109.7 245.7 ± 94.8 144.0 ± 15.6
VSS mg/L 309.4 ± 203.6 235.2 ± 89.2 245.7 ± 94.8 144.0 ± 15.6
NH4+-N mg/L 76.7 ± 9.3 73.6 ± 16.2 64.7 ± 13.9 60.9 ± 10.4
TN mg/L 97.3 ± 17.6 102.2 ± 31.8 84.9 ± 13.9 87.5 ± 14.5
PO4−3-P mg/L 10.7 ± 2.7 12.2 ± 0.1 11.4 ± 2.1 10.4 ± 2.7
TP mg/L 11.7 ± 3.0 13.5 ± 1.2 14.4 ± 2.4 13.1 ± 2.5
TC Log MPN/100 mL ND 8.1 ± 0.1 7.3 ± 0.7 7.4 ± 0.6
FC Log MPN/100 mL ND 8.0 ± 0.1 6.7 ± 0.9 7.1 ± 0.8

All values are expressed as mean + standard deviation. For F/W 2015, n = 7; for S/Sm 2015, n = 3; for F/W 2016, n = 5 and for S/Sm 2016, n = 4. ND: no detected.

variations between 1.5 and 3.5 mm/d. Additionally, the precipitation
values were obtained from meteorological data supplied by The Water
General Direction of Chilean government (DGA).
HSSFs were implemented in December 2014, and they were oper-
ated for 844 days with a stabilization period of 114 days. During the
operational time (2015–2016), physicochemical parameters, i.e., COD,
BOD5, NH4+-N, total nitrogen (TN), TP, PO4−3-P, FC, TC, total sus-
pended solids (TSS) and volatile suspended solids (VSS), were measured
monthly considering the average between fall-winter (F/W) and spring-
summer (S/Sm) seasons in the analysis. In situ parameters, such as pH,
temperature, oxidation reduction potential (ORP) and dissolved oxygen
(DO), were also monitored monthly in the three sampling points
(Fig. 1b), except for DO, due to the lack of variation along the HSSF
zones and the low concentrations measured, implying a significant
error (García et al., 2004; Sepúlveda-Mardones et al., 2017).
(466 nm). All plants analyses were done in summer (S) (January 2015),
fall (F) (May 2016) and spring (S) (September 2016).
2.4. Mass balance
Mass balances were performed in terms of nutrients (TN and TP).
Eq. (1) explains the procedure for calculating balances as follows:
(Ci·Qi·Do)/·A−Ci + r −Cp = (Ce ·Qe·De )/ A (1)
where Ci = influent concentration of TN or TP (g/L); Qi = input stream
wastewater (L/d); Do = operation time (d); A = surface area (m2);
Ci + r = intake by microorganisms or retention in HSSF (g/m2);
Cp = plant uptake (g/m2); Ce = effluent concentration of TN or TP (g/
L); and Qe = effluent output stream (L/d). This procedure was modified
according to Kadlec and Wallace (2009).

2.2. Analytical methods 2.5. Statistical analyses

To evaluate the influent and effluent of each HSSF, samples were
filtered using a 0.45 μm pore size membrane. The physicochemical
parameters of COD (colorimetric method, 5210-B), BOD5 (modified
Winker azide method, 5210-B), NH4+-N (distillation methods by TKN
equipment), TN, TP (Spectroquant-Nova 60, Merck kits), PO4−3-P
(colorimetric method), FC, TC (multiple-tube fermentation technique,
9221-TC) and TSS and VSS (gravimetric method, 2540-D and 2540-E,
respectively) were measured based on the protocols described in
Standard Methods (APHA, 1998). In situ parameters, such pH, tem-
perature and OPR, were determined using portable equipment
OAKTON (PC650–480485). The DO concentration was measured using
a portable oxygen sensor (HANNA OXI 330i/set HI 9146-04).
2.3. Proximate analyses of ornamental plants
To compare the growth of Zantedeschia aethiopica and Cyperus pa-
pyrus in each HSSFs, the coverage percentage (%) and the density (in-
dividuals/m2) were calculated from counting the number of individuals
in a PVC quadrant of 0.0625 m2 (Neubauer et al., 2012; López et al.,
2016).
The biomass above and belowground was divided into flowers, roots
and stems, and then they were measured after drying in an oven be-
tween 70 and 80 °C for 12–24 h (López et al., 2016). According to the
protocol described by Sadzawka et al. (2007), the separated biomasses
were powdered and analyzed for nutrient content. For nitrogen ana-
lysis, the samples were digested with sulfuric acid (H2SO4), salicylic
acid (C7H6O3) and hydrogen peroxide (H2O2), and then the nitrogen
content was measured by colorimetry (650 nm). The phosphorus con-
tent was determined by calcination (500 °C) and then by colorimetry
Statistical analyses were performed using the statistical program
Statistica v.7.0 with a significance level of p = 0.05. First, data were
subjected to a normality test (the Shapiro–Wilk test). Then, to compare
influent concentrations, removal efficiencies and in situ parameters
during different seasons (F/W and S/Sm) and years (2015 and 2016)
between HSSF-Cyp and HSSF-Cyp/Zant, the following tests were per-
formed: a) for data without a normal distribution, an ANOVA test, and
b) for data without a normal distribution, a Kruskal-Wallis test. To
compare the nutrients (nitrogen and phosphorus) contents and biomass
between different zones (A, B and C), the data were analyzed with the
same tests previously presented. Finally, to analyze the difference of
biomass production between HSSFs and the nutrient content between
plants species, the following tests were performed: a) for data with a
normal distribution, a paired t-test, and b) for data without a normal
distribution, a Wilcoxon test.
3. Results and discussion
3.1. Influent characteristics
Table 2 shows the physicochemical characterization of the influent
of HSSF-Cyp/Zant and HSSF-Cyp. For all parameters, non-significant
differences among the average concentrations between seasons were
observed (p > 0.05). The mean BOD5 and COD concentrations varied
between 142.9–229.0 mg/L and 262.4–342.6 mg/L, respectively. These
values are similar to those found by López et al. (2015), where the
average concentrations of BOD5 and COD fluctuated between
167–240 mg/L and 210–420 mg/L, respectively. Regarding the COD/
BOD5 ratios of this influent during the F/W and S/Sm of 2015 and

118
A.M. Leiva et al. Ecological Engineering 120 (2018) 116–125

2016, their values were between 1.2 and 1.9. These low ratios indicate
that the organic matter was simple to degrade (Henze et al., 2002).
Likewise, the influent showed mean TSS and VSS concentrations be-
tween 144.0–412.8 mg/L and 144.0–309.4 mg/L, respectively, during
F/W and S/Sm of 2015–2016. Sepúlveda-Mardones et al. (2017) pre-
sented TSS and VSS values of 131.9–412.8 mg/L and 95.2–309.5 mg/L,
respectively, during 2013–2015. These similarities are explained using
the same influent in both research studies. For NH4+-N and TN con-
centrations, these values were 4 and 3 times higher, respectively, than
those observed in the influent of HSSFs planted with ornamental plants
(Zurita et al., 2009). However, TP and PO4−3-P concentrations were
similar to the values reported by other authors, which ranged between
6.0–13.9 mg/L (Burgos et al., 2017; Konnerup et al., 2009; Zurita et al.,
2009). These concentrations of nutrients (nitrogen and phosphorus) are
possible due to the water use and the behavior of the rural population.
Additionally, the values of FC and TC varied between 6.7 and 8.0 Log
MPN/100 mL and 7.3–8.1 Log MPN/100 mL, respectively, during the
different seasons, which agree with values from Burgos et al. (2017).
3.2. In situ parameters of HSSF-Cyp/Zant and HSSF-Cyp
Fig. 2 shows the behavior of the ORP, pH, T and DO measured in
Zones A, B and C for HSSF-Cyp/Zant and HSSF-Cyp during 2015 and
2016. In this study, the effect of different zones on in situ parameters
behavior was not significant (p > 0.05). The ORP fluctuated from
−356.3 mV to −23.3 mV in both HSSF systems during the operational
time (Fig. 2a). Non-significant differences between HSSF-Cyp/Zant and
HSSF-Cyp were observed (p > 0.05). These ranges of ORP agree with
values described in HSSF CWs planted with ornamental and common
plants that fluctuated from −356.5 mV to −61 mV (Belmont et al.,
2004; Burgos et al., 2017; López et al., 2015). For both HSSF systems,
the ORP values measured in F/W of 2015 varied from −207.9 mV to
−201 mV, and during the subsequent seasons, they increased from
−342 mV to −23 mV (p < 0.05). This difference could be related to
OLR variations during the seasons; these variations decreased from
5.7 gBOD5/m2 d to 4.1 g DBO5/m2 between F/W of 2015 and S/Sm of
2015.
In the case of pH values, they ranged between 6.1 and 7.4 during
2015–2016, with an average value of 6.7 ± 0.2 in HSSF-Cyp/Zant and
HSSF-Cyp. A non-significant effect of HSSFs and seasons were reported
(p > 0.05). Additionally, similar results between 6.95 and 7.9 have
been found in the literature for HSSF CWs planted with ornamental
plants (Calheiros et al., 2015; Konnerup et al., 2009; Zurita et al.,
2009). Regarding water temperatures for HSSF-Cyp/Zant and HSSF-
Cyp, their values ranged between 8.2 and 22.9 °C (averaging
14.6 ± 4.1 °C and 14.8 ± 4.1 °C for HSSF-Cyp/Zant and HSSF-Cyp,
respectively) without significant effect between species (p > 0.05).
However, a seasonal effect on temperature was observed (p < 0.05).
The average temperatures of F/W during 2015 (13.9 ± 3.9 °C) and
2016 (11.3 ± 3.0 °C) were 18% and 39% lower than S/Sm during 2015
(16.8 ± 3.2 °C) and 2016 (18.4 ± 3.6 °C), respectively. In this study,
the DO concentrations were below 0.5 mg/L in both HSSF systems
(p > 0.05). These OD and ORP (averaging −250.0 ± 77.9 mV) values
confirm the anaerobic conditions of HSSF-Cyp/Zant and HSSF-Cyp,
which are similar to those described by Zurita et al. (2009) (0.4 mg/L)
and López et al. (2015) (−183.5 mV).
3.3. Organic matter and suspended solid removal efficiencies of HSSF-Cyp/
Zant and HSSF-Cyp
Fig. 3 shows the removal efficiencies and concentrations of BOD5,
COD, TSS and VSS for both HSSF systems during the operational period.
For HSSF-Cyp/Zant and HSSF-Cyp, the average removal efficiencies of
BOD5 were 70.0 ± 21.8% and 66.1 ± 17.9%, respectively, with mean

HSSF-Cyp/Zant HSSF-Cyp
2015 2016 2015 2016
0
a)
ORP (mV)

-100

-200

-300

-400
8
b)
7
pH

5
Temperature (°C)

25
c)
20
15
10
5
0
0,6
Dissolved Oxygen

d)

0,4
(mg/L)

0,2

0,0
Mar May Jul Sep Dec Feb Apr Jul Sep Nov Jan Mar May Jul Sep Dec Feb Apr Jul Sep Nov Jan
Zone A Zone B Zone C
Fig. 2. In situ parameters variation in Zone A, Zone B and Zone C of HSSF-Cyp/Zant and HSSF-Cyp, measured from April 2015 to December 2016. a) ORP; b) pH; c)
Temperature and d) DO.

119
A.M. Leiva et al.
Fig. 3. Removal efficiencies (bar chart) and effluent concentration (scatter plot) for a) COD; b) BOD5, c) TSS and d) VSS during monitored seasons in HSSF-Cyp/Zant
and HSSF-Cyp.
effluent concentrations of 54.8 ± 26.7 mg/L and 64.1 ± 28.5 mg/L,
respectively. In the case of COD, the results were similar with removal
efficiencies of 68.0 ± 11.7% and 60.7 ± 15.0% and effluent con-
centrations of 102.8 ± 32.8 mg/L and 123.8 ± 43.2 mg/L for HSSF-
Cyp/Zant and HSSF-Cyp, respectively. These performances observed in
both HSSF systems agreed with the range of results found in the lit-
erature for similar systems with OLRs close to 4.7 gBOD5/m2 d, which
achieved BOD5 and COD removal efficiencies above 60% (Carballeira
et al., 2016). On the contrary, HSSF systems planted with ornamental
plants (Zantedeschia aethiopica, Strelitzia reginae, Anturium andreanum
and Agapanthus africanus) and with similar OLR (4.6 gBOD5/m2 d) re-
ported removal efficiencies 10% and 16% higher than efficiencies found
in this study (Zurita et al., 2009). In this previous research, the average
temperature condition of HSSFs systems was close to 22 °C. The ideal
temperatures required for Zantedeschia aethiopica and Cyperus papyrus
were in the ranges of 18–25 °C and 18–21 °C, respectively (Zurita et al.,
2009; Morales et al., 2013). Truu et al. (2009) determined that the
efficiency of removal contaminants depended on the presence of plants.
This difference in removal efficiencies between research studies can be
related to the average temperature of 14 °C in HSSF-Cyp/Zant and
HSSF-Cyp. In both HSSF systems, Fig. 3a shows a decrease from 71.8%
to 57.8% of BOD5 removal efficiency between S/Sm of 2015 to F/W of
2016. This behavior can be related to the increase in OLR (21%) from
4.1 ± 0.5 g BOD5/m2 d in S/Sm of 2015 to 5.2 ± 2.0 g BOD5/m2 d in
F/W of 2015. However, statistical analyses demonstrated that different
seasons did not influence organic matter removal efficiencies in HSSF-
Cyp/Zant and HSSF-Cyp (p > 0.05).
Regarding suspended solids removal efficiencies and effluent con-
centrations in HSSF-Cyp/Zant and HSSF-Cyp, the performances for TSS
and VSS were 94.9 ± 6.1% and 96.0 ± 4.0%, respectively, with ef-
fluent concentrations of 9.7 ± 8.1 mg/L and 9.6 ± 8.1 mg/L,
120
Ecological Engineering 120 (2018) 116–125
respectively, during F/W and S/Sm of two years of operation
(2015–2016). In this case, non-significant differences between seasons
were observed (p > 0.05). This result shows that these systems were
stable along the seasons due to TSS loading rates (3.2–11.0 gTSS/m2 d),
which were in the range of 2.6–10 gTSS/m2 d (Vymazal, 2005; Caselles-
Osorio et al., 2007) used by other studies to provide removal effi-
ciencies above 78%. In addition, TSS and VSS removal efficiencies de-
termined in this study agree with other authors, who reported perfor-
mances above 90% using TSS loading rates between 2.5 and 15 gTSS/
m2 d (Konnerup et al., 2009; López et al., 2015; Sepúlveda-Mardones
et al., 2017).
Comparing the organic matter and suspended solid removal effi-
ciencies between the polyculture system (HSSF-Cyp/Zant) and mono-
culture system (HSSF-Cyp), there were differences of 5.6%, 10.7%,
1.8% and 0.1% for BOD5, COD, TSS and VSS, respectively, which were
non-significant (p > 0.05). These results differ with those found by
Liang et al. (2011), where a monoculture wetland (planted with Canna
indica) had a significantly higher COD (difference of 12%) removal
rates than the polyculture wetland (Canna indica, Cyperus flabelliformis,
Phragmites australis, Pennistum purpureum and Hymenocallis littoralis)
during the first year of operation (p < 0.05). In this previous study,
during the first year, the polyculture wetland was dominated by Canna
indica and Cyperus flabelliformis but during the latter years was domi-
nated by Phragmites australis. This last species showed a higher growth
(propagation rate: 10 m/yr) than the other two species. However,
Zurita et al. (2009) reported that the polyculture wetlands (81%) were
more effective in the COD removal in comparison to monoculture
wetlands (78%). Moreover, other researchers suggest that the poly-
culture system (average removal loading rate of 58.1 g TSS/m2 d)
seemed to provide the best and most consistent treatment for TSS than
monoculture systems (26.3–33.9 gTSS/m2 d) (Karathanasis et al.,
A.M. Leiva et al.
Fig. 4. Removal efficiencies (box plot) for a) NH4+-N; b) TN; c) PO4−3-P and d) TP during monitored seasons in HSSF-Cyp/Zant and HSSF-Cyp.
2003). This behavior can be due to the rooting biomass of the poly-
culture system, which provided more effective filtration than the
monoculture system. This result was supported by Zurita et al. (2009),
who showed a difference of 15% in TSS removal between polyculture
and monoculture systems. The result also suggested that the matrix
formed by the roots of polyculture species with the substrate probably
trapped the solids more efficiently and lowered the water velocity,
improving the retention of TSS.
3.4. Nutrients and pathogens removal efficiencies of HSSF-Cyp/Zant and
HSSF-Cyp
Figs. 4 and 5 show the removal efficiencies of nutrients (nitrogen
and phosphorus) and pathogens for HSSF-Cyp/Zant and HSSF-Cyp,
respectively, during different seasons. In both HSSF systems, NH4+-N
and TN removal efficiencies fluctuated between
31.7 ± 23.7–35.2 ± 19.1% and 38.8 ± 24.0–42.2 ± 24.4%, re-
spectively (Fig. 4a and b). The average removals achieved in this study
have been reported in shallow HSSF systems (0.3 m of depth) using
Juncus effusus, Iris pseudacorus, Typha latifolia L. and Phragmites australis
as a monoculture (21–36% and 26–38% for NH4+-N and TN)
(Carballeira et al., 2016). Despite this similarity, Kyambadde et al.
(2004) reached NH4+-N removal efficiencies in the range of
61.5–75.3% for HSSF CWs planted with Cyperus papyrus and Mis-
canthidium violaceum. Moreover, an HSSF system planted with Cyperus
articulates achieved performances above 90% for NH4+-N removal
(Caselles-Osorio et al., 2017). This highest performance could be at-
tributed to the carbon-nitrogen ratio (C/N) used in this research (< 2)
compared to C/N used in HSSF-Cyp/Zant and HSSF-Cyp; this latter C/N
was above 4. Generally, the C/N influences important nitrogen removal
processes, such as nitrification and denitrification (Reddy and DeLaune,
2008). Moreover, the HSSF systems previously mentioned were oper-
ated for 4 months, compared to the operational time of HSSF-Cyp/Zant
and HSSF-Cyp of 2 years. Samsó and García (2013) reported that con-
structed wetland stability is achieved between 400 days and 700 days.
121
Ecological Engineering 120 (2018) 116–125
Additionally, the anaerobic conditions that promote HSSF-Cyp/Zant
and HSSF-Cyp (OPR: from −120 to −331 mV; DO concentration: < 2
mg/L), do not allow nitrification mechanism (ORP and OD values
above +100 mV and 1.5 mg/L) (Faulwetter et al., 2009).
Fig. 4c and d show PO4−3-P and TP removal efficiencies between
12.4 ± 21.6–19.8 ± 21.1% and 25.9 ± 24.2%–33.0 ± 24%, re-
spectively, in both HSSF systems. Burgos et al. (2017) and Zurita et al.
(2009) reported removal efficiencies of PO4−3-P and TP in the range of
30–40%. This last study used tezontle gravel, a volcanic red-orange
extrusive rock, for filling their HSSF systems. This material has a high
porosity (0.53) and richness in iron that makes it a good substrate for
phosphorus removal (Zurita et al., 2006). In contrast, gravel, the sup-
port medium used in this work, has a low phosphorus adsorption ca-
pacity between 0.03 g P/kg and 0.05 g P/kg (Vera et al., 2014; Vohla
et al., 2011). Another reason that can influence these lower perfor-
mances is the HRTs used in both HSSF systems (5–6 d). Akratos and
Tsihrintzis (2007) demonstrated that for PO4−3-P and TP, the removal
efficiencies varied from 56.3% to 88.1% using HRTs between 8 and 20 d
in pilot-scale HSSF CWs.
In the case of pathogens removal (Fig. 5), both HSSF systems (HSSF-
Cyp/Zant and HSSF-Cyp) showed FC and TC removal efficiencies of 2.0
Log MPN/100 mL and 1.7–2.2 Log MPN/100 mL, respectively, during
the operational period. These results agree with the removal of 1.8–2.7
Log MPN/100 mL, which was previously reported in HSSF CWs with
ornamental plants species (Burgos et al., 2017; Calheiros et al., 2015;
Carballeira et al., 2016).
For all removal efficiencies (NH4+-N, TN, PO4−3-P, TP, TC and FC),
the effect of seasons and plant species were not significant (p > 0.05).
Zhang et al. (2007) described that nutrient removal efficiencies (NH4+-
N and PO4−3-P) had differences of 4.1% and 1.2%, respectively, be-
tween a polyculture wetland (Canna indica and Schoenoplectus validus)
and monoculture wetland. The same behavior was observed in Liang
et al. (2011), who reported non-significant differences in removal rates
of TN and TP between the polyculture wetland and the monoculture
wetland (p > 0.05). Only the NH4+-N removal efficiency was
A.M. Leiva et al. Ecological Engineering 120 (2018) 116–125

6 10

Removal Efficiency (Log MPN/100 mL)

a)
5

(Log MPN/100 mL)

8

TC Influent
4
6
3
4
2

1 2

0 0
6 10
b)

(Log MPN/100 mL)

5 8

FC Influent
4
6
3
4
2

1 2

0 0
S/Sm 2015 F/W 2016 S/Sm 2016

HSSF-Cyp/Zant HSSF-Cyp
Fig. 5. Pathogens removal efficiencies (bar chart) for a) Total Coliform (TC) and b) Fecal Coliform (FC) and influent pathogen concentration (scatter plot) in HSSF-
Cyp/Zant and HSSF-Cyp during monitored seasons.

significantly higher in the monoculture wetland than in the polyculture
(p < 0.05). Likewise, Rodríguez and Brisson (2016) determined that
the monoculture and polyculture of Phragmites australis and Phalaris
arundinacea would not improve the removal efficiencies of NH4+-N.
This study showed that the morphology of plants influenced the redox
conditions of the rhizosphere and, thus, determined the microbiological
mechanisms for the nitrogen removal, such as nitrification and deni-
trification. For this reason, they recommended combining plant species,
because it is the best tradeoff between high removal efficiencies of
pollutants and low nitrate level in the influent. For pathogens removal,
some authors reported that polyculture wetlands appeared to provide
the most consistent FC removal (97%) (Karathanasis et al., 2003). With
these results, it is not clear if it is desirable to maintain a single plant or
a mixed plant species in CWs.
3.5. Ornamental plants development in HSSF-Cyp/Zant and HSSF-Cyp
Table 3 shows the growth characteristics and total biomass pro-
duction (above and belowground) evaluated by zones (A, B, and C) of
each HSSF system during the two operational years (2015–2016).
Comparing HSSF-Cyp/Zant (polyculture system) and HSSF-Cyp
(monoculture system), the biomass production was significantly dif-
ferent in Zone B (the middle zone: 1.4 m from the inlet) and Zone C (the
outlet zone: 2.25 m from the inlet) (p < 0.05). In Zone B, HSSF-Cyp/
Zant (3.7 kg DW/m2) had a larger total biomass compared to HSSF-Cyp
(0.2 kg DW/m2). However, in Zone C, the behavior was completely
different. The total biomass of HSSF-Cyp was 2 times higher than that
found in HSSF-Cyp/Zant (p < 0.05). HSSF-Cyp had a similar coverage
percentage to HSSF-Cyp/Zant in Zones B and C, but higher percentages
in Zone A (> 50% for HSSF-Cyp and < 35% for HSSF-Cyp/Zant)
(p < 0.05). For the density (individuals/m2), the monoculture system
displayed a total density 1.2–1.7 times higher than the polyculture
system, but only in Zone C was it significant (p < 0.05). These results
suggest that, in general, HSSF-Cyp was more stable in growth char-
acteristics (biomass production and density) than HSSF-Cyp/Zant. The
same behavior was observed in Liang et al. (2011), where the mono-
culture wetland had a total biomass of 2.0 kg DW/m2 during the first
year, and the polyculture wetland achieved a total biomass 1.2 times
lower. However, this condition changed during the latter three years,
where the monoculture system had smaller total biomass than the
polyculture system (p < 0.05). Therefore, this study proposes that the
plant growth and the community structure varied annually both in the
monoculture and polyculture wetlands. It has been reported that
polyculture systems were less susceptible to seasonal variations and was
more effective in temporal and spatial compensation in plant growth
(Karathanasis et al., 2003). Nonetheless, the polyculture systems are
unstable due to competition and/or other interactions. Competitions
may cause dominance in certain species and suppression or extinction

Table 3
Growth characteristics and biomass production in HSSF-Cyp/Zant and HSSF-Cyp during the operational time.
Zone Constructed Wetlands

HSSF-Cyp/Zant HSSF-Cyp

A B C A B C

Total biomass (kg DW/m2) 7.4 ± 7.1 3.7 ± 1.0 8.6 ± 2.0 5.0 ± 0.9 0.2 ± 0.0 16.34 ± 1.0
Coverage (%) 35.6 ± 1.9 8.7 ± 1.5 61.6 ± 1.8 57.5 ± 4.7 15.5 ± 2.8 64.4 ± 4.7
Density (individuals/m2) 165 ± 30 32 ± 10 257 ± 38 195 ± 50 45 ± 27 444 ± 51

122
A.M. Leiva et al.
in other species because of the same growth forms, similar individual
size, and similar light demand (Liu et al., 2012). In the case of HSSF-
Cyp/Zant (polyculture system), Cyperus papyrus dominated, and it
tended to develop into a new monoculture. During the two operational
years, the total density (average of Zones A, B and C) of Cyperus papyrus
was 150 individuals/m2 compared to Zantedeschia aethiopica, which
achieved 3 individuals/m2 (data no shown). The competition between
Cyperus papyrus and Zantedeschia aethiopica can be related with the
difference in the root distribution of these species. According to Lai
et al. (2011), wetland plants can be divided into two groups: fibrous-
root plants and thick-roots plants. In this case, Cyperus papyrus can be
classified such a fibrous-root plant and Zantedeschia aethiopica had the
same characteristics of thick-root plants. The previous study showed
that fibrous-root plants had significantly higher growth rates (16.07 g/
m2) than thick-root plants (7.56 g/m2). In HSSF-Cyp/Zant, it is possible
that the growth of Cyperus papyrus were higher and excluded Zante-
deschia aethiopica when they grow together. For this reason, a suitable
selection of plant species is recommended for decreasing the competi-
tion in polyculture system and avoiding unstable plant community in
CWs (Katharina et al., 2002).
On the contrary, in HSSF-Cyp (monoculture system), the total bio-
mass and the density in Zone C were higher than Zones A and B
(p < 0.05). Zone C corresponds to the outlet zone of CWs. This phe-
nomenon can be associated with higher TN concentrations of the in-
fluent (87.5–102.2 mg/L, see Table 1) in inlet zones (Zones A and B)
that can affect a plant’s development. Clarke and Baldwin (2002) in-
dicate that concentrations of NH4+-N close to 100 mg/L limited the
growth of plants commonly used in CWs, such as Schoenoplectus ta-
bernaemontani.
3.6. Nutrients uptake in HSSF-Cyp/Zant and HSSF-Cyp
Fig. 6 shows the TN and TP contents in different plant tissues for
both HSSF systems. Zantedeschia aethiopica planted in HSSF-Cyp/Zant
had the highest TN and TP contents in stems (18.7 g N/kg·DW and
3.08 g P/kg·DW, respectively), roots (4.2 g N/kg·DW and 0.52 g P/kg
DW, respectively) and flowers (36.7 g N/kg·DW and 4.6 g P/kg·DW,
respectively) (p < 0.05). These results agree with those found by
Carneiro et al. (2015), who reported TN and TP contents in flowers of
Zantedeschia spp. between 25.5–34.7 g N/kg·DW and 4.1–4.9 g P/
kg·DW, respectively. This plant species was commonly used in CWs in
tropical regions, and it presents more vigor and unstressed signs in the
HSSF system than vertical subsurface flow (VSSF) CWs. This char-
acteristic is probably due to its mechanisms of adaptation to water
clogging (Zurita et al., 2008, 2009). Comparing Cyperus papyrus planted
Fig. 6. Distribution of a) TN and b) TP content in plant tissues (Flower; Root; Stem) for two species plants (Cyp: Cyperus papyrus; Zant: Zantedeschia aethiopica) in both
HSSF systems.
123
Ecological Engineering 120 (2018) 116–125
in HSSF-Cyp/Zant with those planted in HSSF-Cyp, non-significant
differences in TN and TP contents were observed (p > 0.05). Abou-
Elela and Hellal (2012) used Cyperus papyrus in VSSF systems and re-
ported TN and TP contents of 19.3 g N/kg·DW and 2.2 g P/kg·DW, re-
spectively, which were similar to those found in this study, which
fluctuated between 21.8–25.1 g N/kg·DW and 3.3–3.8 g P/kg·DW, re-
spectively, in both HSSF systems. The TN and TP distribution rank in
the three plant tissues was as flower > stem > root for the three
plants evaluated (p > 0.05). This distribution is explained by the
translocation of nutrients from the roots to the aerated zones of plants.
Specifically, flowers are an important structure, because they are
charged of reproduction (Stottmeister et al., 2003). However, the nu-
trient contents of different tissues varied significantly depending on soil
conditions, seasonal variations and the growth rhythm of plants
(Baldwin et al., 2006; Liu et al., 2012).
Table 4 shows the mass balance of nutrients (nitrogen and phos-
phorus) in HSSF-Cyp/Zant and HSSF-Cyp during 2015 and 2016. For
the nitrogen mass balance, the plant uptake varied between 8 and 12%
in both HSSF systems during 2015 without a significant difference be-
tween CWs (p > 0.05). Regarding the literature, the contribution of
plants to nitrogen removal varied in the range of 0.5–40% (Vymazal,
2007), similar to those values found in this study. Additionally, there
exists limited information about nitrogen uptake in Cyperus spp. in
CWs. Leto et al. (2013) noted a nitrogen uptake of 40.6 g N/m2 for
aboveground and 62.7 g N/m2 for belowground parts (a total of
103.3 g N/m2 yr). In the present study, during the second year (2016),
the nitrogen uptake was above 80 g N/m2 (Table 4). For Zantedeschia
aethiopica, there is no information about the nitrogen uptake of this
species in CWs.
The same behavior was observed for the phosphorus mass balance.
During 2015, the plant uptake was 12% and 41% for HSSF-Cyp/Zant
and HSSF-Cyp, respectively. For Cyperus papyrus, Kyambadde et al.
(2004) reported that the plant uptake of phosphorus was 88.8%. This
value is similar to those found in HSSF-Cyp (78%) and HSSF-Cyp/Zant
(95%) during 2016 (Table 4). This similarity in the percentage of
phosphorus plant uptake between HSSF-Cyp/Zant and HSSF-Cyp during
2016 can be related to the competition between Cyperus papyrus and
Zantedeschia aethiopica in HSSF-Cyp/Zant that develops a new mono-
culture of Cyperus papyrus species. In addition, Vymazal (2007) ob-
served that phosphorus uptake for emergent species were in the range
of 1.4–37.5 g P/m2, similar to those found in the present study, which
varied between 3 and 14 g P/m2 in HSSF-Cyp/Zant and HSSF-Cyp.
Compared to common species used in CWs, such as Phragmites australis
and Typha orientalis Presl. (9–27 g P/m2), the phosphorus uptake was
also similar (Liu et al., 2012).
A.M. Leiva et al. Ecological Engineering 120 (2018) 116–125

Table 4
Nitrogen and phosphorus mass balance for HSSF-Cyp/Zant and HSSF-Cyp.
Parameters Constructed Wetlands Year Influent Effluent Plant uptake Retention

2 2 2
(g/m ) (g/m ) (%) (g/m ) (%) (g/m2) (%)

TN HSSF-Cyp/Zant 2015 508 ± 80 343 ± 43 68 ± 2 15 ± 1 8 ± 0.1 165 ± 36 32 ± 2

2016 366 ± 7 202 ± 4 55 ± 1 116 ± 9 71 ± 4 164 ± 3 45 ± 0
HSSF-Cyp 2015 508 ± 80 366 ± 42 72 ± 3 20 ± 1 24 ± 0.1 185 ± 38 28 ± 3
2016 366 ± 7 219 ± 9 60 ± 4 87 ± 35 61 ± 31 53 ± 17 40 ± 4

TP HSSF-Cyp/Zant 2015 64 ± 14 45 ± 12 70 ± 3.1 3 ± 1.0 12 ± 0.1 19 ± 2 30 ± 3

2016 58 ± 4 44 ± 4 75 ± 2.1 14 ± 1.0 95 ± 5 15 ± 0 25 ± 2
HSSF-Cyp 2015 64 ± 14 51 ± 8 80 ± 5.7 7 ± 1.0 41 ± 1 13 ± 7 20 ± 6
2016 58 ± 4 45 ± 4 77 ± 1.2 10 ± 4 78 ± 31 13 ± 0 23 ± 1

Furthermore, significant differences between years (2015 and 2016)
were observed in the plant uptake of nutrients. In 2016, the plant up-
take of nitrogen and phosphorus increased in HSSF-Cyp/Zant and HSSF-
Cyp between 61–71% and 78–95%, respectively (p < 0.05). These
differences can be related to the growing of ornamental plants and the
vegetation establishment during the first year of operation.
In this study, the results of removal efficiencies of different pollu-
tants, biomass production and nutrient uptake showed that HSSF-Cyp/
Zant and HSSF-Cyp have similar performance and non-significant dif-
ference were observed between polyculture and monoculture system.
Despite this behavior, polyculture systems present advantages related
with the aesthetic benefits for the community. Moreover, Rodríguez
and Brisson (2016) reported that the combination of plant species in
CWs improve the resistance to environmental stress and disease and the
system landscape.
4. Conclusions
In the present study, HSSF-Cyp/Zant and HSSF-Cyp reached re-
moval efficiencies for organic matter (COD and BOD5) and suspended
solids (TSS and VSS) above 60% and 90%, respectively, during the two
operational years. Nutrients (nitrogen and phosphorus) and pathogens
(FC and TC) removal efficiencies ranged between 10–40% and 1.7–2.2
Log MPN/100 mL, respectively. For all parameters, the effect of dif-
ferent seasons was not significant (p > 0.05). Additionally, HSSF-Cyp/
Zant (polyculture system) did not show significant differences in the
removal efficiencies when compared to HSSF-Cyp (monoculture
system).
Moreover, differences between HSSF-Cyp/Zant and HSSF-Cyp in the
biomass production and the TN and TP contents were observed
(p < 0.05). In this study, HSSF-Cyp achieved a total biomass of
21.5 kg DW/m2, whereas Zantedeschia aethiopica planted in HSSF-Cyp/
Zant reached the highest TN (59.6 g N/kg·DW) and TP (8.28 g P/kg·DW)
content. However, TN and TP mass balances determined that the effect
of monoculture and polyculture systems was not significant (p > 0.05).
Despite there is no evidence that the polyculture system (HSSF-Cyp/
Zant) improve the treatment performance and the biomass production
when is compared with the monoculture system (HSSF-Cyp), this
system provide others benefits like improving the system landscape and
enhancing the habitat quality to the community and the resistance to
environmental stress.
Acknowledgement
This work was supported by INNNOVA BIO BIO Proyect N° 13.3327-
543 IN.IIP and CONICYT/FONDAP/15130015.
References
Abou-Elela, S.I., Hellal, M.S., 2012. Municipal wastewater treatment using vertical flow
constructed wetlands planted with Canna, Phragmites and Cyprus. Ecol. Eng. 47,
209–213. http://dx.doi.org/10.1016/j.ecoleng.2012.06.044.
Akratos, C.S., Tsihrintzis, V.A., 2007. Effect of temperature, HRT, vegetation and porous
media on removal efficiency of pilot-scale horizontal subsurface flow constructed
wetlands. Ecol. Eng. 29 (2), 173–191. http://dx.doi.org/10.1016/j.ecoleng.2006.06.
013.
American Public Health Association (APHA), 1998. Standard Methods for the
Examination of Wastewater, 20th ed. APHA, Washington (DC).
Andrés, E., Araya, F., Vera, I., Pozo, G., Vidal, G., 2018. Phosphate removal using zeolite
in treatment wetlands under different oxidation-reduction potentials. Ecol. Eng. 117,
18–27. http://dx.doi.org/10.1016/j.ecoleng.2018.03.008.
Baldwin, D.S., Rees, G.N., Mitchell, A.M., Watson, G., Williams, J., 2006. The short term
effects of salinization on anaerobic nutrient cycling and microbial community
structure in sediment from a freshwater wetland. Wetland 26, 455–464. http://dx.
doi.org/10.1672/0277-5212(2006) 26[455:TSEOSO]2.0.CO;2.
Belmont, M.A., Metcalfe, C.D., 2003. Feasibility of using ornamental plants (Zantedeschia
aethiopica) in subsurface flow treatment wetlands to remove nitrogen, chemical
oxygen demand and nonylphenol ethoxylate surfactants-a laboratory-scale study.
Ecol. Eng. 21 (4), 233–247. http://dx.doi.org/10.1016/j.ecoleng.2003.10.003.
Belmont, M.A., Cantellano, E., Thompson, S., Williamson, M., Sánchez, A., Metcalfe, C.D.,
2004. Treatment of domestic wastewater in a pilot-scale natural treatment system in
central Mexico. Ecol. Eng. 23 (4), 299–311. http://dx.doi.org/10.1016/j.ecoleng.
2004.11.003.
Burgos, V., Araya, F., Reyes-Contreras, C., Vera, I., Vidal, G., 2017. Performance of or-
namental plants in mesocosm subsurface constructed wetlands under different or-
ganic sewage loading. Ecol. Eng. 99, 246–255. http://dx.doi.org/10.1016/j.ecoleng.
2016.11.058.
Brix, H., 1997. Do macrophytes play a role in constructed treatment wetlands? Water Sci.
Technol. 35 (5), 11–17. http://dx.doi.org/10.1016/S0273-1223(97)00047-4.
Calheiros, C.S., Bessa, V.S., Mesquita, R.B., Brix, H., Rangel, A.O., Castro, P.M., 2015.
Constructed wetland with a polyculture of ornamental plants for wastewater treat-
ment at a rural tourism facility. Ecol. Eng. 79, 1–7. http://dx.doi.org/10.1016/j.
ecoleng.2015.03.001.
Carballeira, T., Ruiz, I., Soto, M., 2016. Effect of plants and surface loading rate on the
treatment efficiency of shallow subsurface constructed wetlands. Ecol. Eng. 90,
203–214. http://dx.doi.org/10.1016/j.ecoleng.2016.01.038.
Carneiro, D.N., Coelho, L.L., Paiva, P., Almeida, E.F., Carneiro, L.F., 2015. Evaluation of
macronutrient demand in calla lily (Zantedeschia aethiopica). Aust. J. Crop Sci. 9 (8),
761.
Caselles-Osorio, A., Puigagut, J., Segú, E., Vaello, N., Granés, F., García, D., García, J.,
2007. Solids accumulation in six full-scale subsurface flow constructed wetlands.
Water Res. 41 (6), 1388–1398. http://dx.doi.org/10.1016/j.watres.2006.12.019.
Caselles-Osorio, A., Vega, H., Lancheros, J.C., Casierra-Martínez, H.A., Mosquera, J.E.,
2017. Horizontal subsurface-flow constructed wetland removal efficiency using
Cyperus articulatus L. Ecol. Eng. 99, 479–485. http://dx.doi.org/10.1016/j.ecoleng.
2016.11.062.
Clarke, E., Baldwin, A.H., 2002. Responses of wetland plants to ammonia and water level.
Ecol. Eng. 18 (3), 257–264. http://dx.doi.org/10.1016/j.ecoleng.2016.02.047.
Faulwetter, J.L., Gagnon, V., Sundberg, C., Chazarenc, F., Burr, M.D., Brisson, J., Stein,
O.R., 2009. Microbial processes influencing performance of treatment wetlands: a
review. Ecol. Eng. 35 (6), 987–1004. http://dx.doi.org/10.1016/j.ecoleng.2008.12.
030.
García, J., Aguirre, P., Mujeriego, R., Huang, Y., Ortiz, L., Bayona, J., 2004. Initial con-
taminant removal performance factors in horizontal flow reed beds used for treating
urban wastewater. Water Res. 38, 1669–1678. http://dx.doi.org/10.1016/j.watres.
2004.01.011.
Henze, M., Harremoës, P., La Cour Jansen, J., Arvin, E., 2002. Wastewater treatment:
biological and chemical processes, 3rd ed. Springer, Heidelberg.
Kadlec, R., Wallace, S., 2009. Treatment Wetlands, 2nd ed. CRC Press, Boca Raton.
Karathanasis, A.D., Potter, C.L., Coyne, M.S., 2003. Vegetation effects on fecal bacteria,
BOD, and suspended solid removal in constructed wetlands treating domestic was-
tewater. Ecol. Eng. 20 (2), 157–169. http://dx.doi.org/10.1016/S0925-8574(03)
00011-9.
Katharina, A., Engelhardt, M., Ritchie, M.E., 2002. The effect of aquatic plant species
richness on wetland ecosystem processes. Ecology 83, 2911–2924. http://dx.doi.org/
10.1890/0012-9658(2002) 083[2911:TEOAPS]2.0.CO;2.
Konnerup, D., Koottatep, T., Brix, H., 2009. Treatment of domestic wastewater in tropical,
subsurface flow constructed wetlands planted with Canna and Heliconia. Ecol. Eng.
35 (2), 248–257. http://dx.doi.org/10.1016/j.ecoleng.2008.04.018.

124
A.M. Leiva et al.
Kyambadde, J., Kansiime, F., Gumaelius, L., Dalhammar, G., 2004. A comparative study
of Cyperus papyrus and Miscanthidium violaceum based constructed wetlands for
wastewater treatment in a tropical climate. Water Res. 38 (2), 475–485. http://dx.
doi.org/10.1016/j.watres.2003.10.008.
Lai, W.L., Wang, S.Q., Peng, C.L., Chen, Z.H., 2011. Root features related to plant growth
and nutrient removal of 35 wetland plants. Water Res. 45 (13), 3941–3950. http://
dx.doi.org/10.1016/j.watres.2011.05.002.
Leto, C., Tuttolomondo, T., La Bella, S., Leone, R., Licata, M., 2013. Effects of plant
species in a horizontal subsurface flow constructed wetland–phytoremediation of
treated urban wastewater with Cyperus alternifolius L. and Typha latifolia L. in the
West of Sicily (Italy). Ecol. Eng. 61, 282–291. http://dx.doi.org/10.1016/j.ecoleng.
2013.09.014.
Liang, M.Q., Zhang, C.F., Peng, C.L., Lai, Z.L., Chen, D.F., Chen, Z.H., 2011. Plant growth,
community structure, and nutrient removal in monoculture and mixed constructed
wetlands. Ecol. Eng. 37 (2), 309–316. http://dx.doi.org/10.1016/j.ecoleng.2010.11.
018.
Liu, X., Huang, S., Tang, T., Liu, X., Scholz, M., 2012. Growth characteristics and nutrient
removal capability of plants in subsurface vertical flow constructed wetlands. Ecol.
Eng. 44, 189–198. http://dx.doi.org/10.1016/j.ecoleng.2012.03.011.
López, D., Fuenzalida, D., Vera, I., Rojas, K., Vidal, G., 2015. Relationship between the
removal of organic matter and the production of methane in subsurface flow con-
structed wetlands designed for wastewater treatment. Ecol. Eng. 83, 296–304. http://
dx.doi.org/10.1016/j.ecoleng.2015.06.037.
López, D., Sepúlveda, M., Vidal, G., 2016. Phragmites australis and Schoenoplectus cali-
fornicus in constructed wetlands: development and nutrient uptake. J. Soil. Sci. Plant.
Nutr. 16, 763–777.
Merino-Solís, M., Villegas, E., de Anda, J., López-López, A., 2015. The effect of the hy-
draulic retention time on the performance of an ecological wastewater treatment
system: an anaerobic filter with a constructed wetland. Water 7, 1149–1163. http://
dx.doi.org/10.3390/w7031149.
Morales, G., López, D., Vera, I., Vidal, G., 2013. Constructed wetlands with ornamental
plants for removal of organic matter and nutrients contained in sewage. Theoria 22
(1), 33–46 (In Spanish).
Neubauer, M., Plaza de los Reyes, C., Pozo, G., Villamar, C.A., Vidal, G., 2012. Growth
and nutrient uptake by Schoenoplectus californicus (CA Méyer) Sójak in a constructed
wetland fed with swine slurry. J. Soil. Sci. Plant. Nutr. 12, 421–430. http://dx.doi.
org/10.4067/S0718-95162012005000004.
Reddy, K.R., DeLaune, R.D., 2008. Biogeochemistry of Wetlands Science and
Applications. CRC Press, Boca Ratón.
Rodríguez, M., Brisson, J., 2016. Does the combination of two plant species improve
removal efficiency in treatment wetlands? Ecol. Eng. 91, 302–309. http://dx.doi.org/
10.1016/j.ecoleng.2016.02.047.
Sadzawka, R., Carrasco, M., Demanet, R., Flores, H., Grez, R., Mora, M., Neaman, A.,
2007. Methods for Analysis of Plant Tissue, 2nd ed. Instituto de Investigaciones
Agropecuarias de Chile (INIA), Santiago de Chile (In Spanish).
Samsó, R., García, J., 2013. Bacteria distribution and dynamics in constructed wetlands
based on modelling results. Sci. Total Environ. 461, 430–440. http://dx.doi.org/10.
1016/j.scitotenv.2013.04.073.
Sepúlveda-Mardones, M., López, D., Vidal, G., 2017. Methanogenic activity in the bio-
mass from horizontal subsurface flow constructed wetlands treating domestic was-
tewater. Ecol. Eng. 105, 66–77. http://dx.doi.org/10.1016/j.ecoleng.2017.04.039.
Shelef, O., Gross, A., Rachmilevitch, S., 2013. Role of plants in a constructed wetland:
125
Ecological Engineering 120 (2018) 116–125
current and new perspectives. Water. 5 (2), 405–419. http://dx.doi.org/10.3390/
w5020405.
Stottmeister, U., Wießner, A., Kuschk, P., Kappelmeyer, U., Kästner, M., Bederski, O.,
Müller, R.A., Moormann, H., 2003. Effects of plants and microorganisms in con-
structed wetlands for wastewater treatment. Biotechnol. Adv. 22 (1), 93–117. http://
dx.doi.org/10.1016/j.biotechadv.2003.08.010.
Truu, M., Juhanson, J., Truu, J., 2009. Microbial biomass, activity and community
composition in constructed wetlands. Sci. Total Environ. 407 (13), 3958–3971.
http://dx.doi.org/10.1016/j.scitotenv.2008.11.036.
Vera, I., García, J., Sáez, K., Moragas, L., Vidal, G., 2011. Performance evaluation of eight
years experience of constructed wetland systems in Catalonia as alternative treatment
for small communities. Ecol. Eng. 37 (2), 364–371. http://dx.doi.org/10.1016/j.
ecoleng.2010.11.031.
Vera, I., Araya, F., Andrés, E., Sáez, K., Vidal, G., 2014. Enhanced phosphorus removal
from sewage in mesocosm-scale constructed wetland using zeolite as medium and
artificial aeration. Environ Technol. 35 (13), 1639–1649. http://dx.doi.org/10.1080/
09593330.2013.877984.
Vohla, C., Kõiv, M., Bavor, H.J., Chazarenc, F., Mander, Ü., 2011. Filter materials for
phosphorus removal from wastewater in treatment wetlands-a review. Ecol. Eng. 37
(1), 70–89. http://dx.doi.org/10.1016/j.ecoleng.2009.08.003.
Vymazal, J., 2005. Horizontal sub-surface flow and hybrid constructed wetlands systems
for wastewater treatment. Ecol. Eng. 25 (5), 478–490. http://dx.doi.org/10.1016/j.
ecoleng.2005.07.010.
Vymazal, J., 2007. Removal of nutrients in various types of constructed wetlands. Sci.
Total Environ. 380 (1), 48–65. http://dx.doi.org/10.1016/j.scitotenv.2006.09.014.
Vymazal, J., 2011. Plants used in constructed wetlands with horizontal subsurface flow: a
review. Hydrobiology 674 (1), 133–156. http://dx.doi.org/10.1007/s10750-011-
0738-9.
Zhang, Z., Rengel, Z., Meney, K., 2007. Nutrient removal from simulated wastewater
using Canna indica and Schoenoplectus validus in mono-and mixed-culture in wet-
land microcosms. Water Air Soil Pollut. 183 (1–4), 95–105. http://dx.doi.org/10.
1007/s11270-007-9359-3.
Zhou, Q., Zhu, H., Bañuelos, G., Yan, B., Liang, Y., Yu, X., Cheng, X., Chen, L., 2017.
Effects of vegetation and temperature on nutrient removal and microbiology in
horizontal subsurface flow constructed wetlands for treatment of domestic sewage.
Water Air Soil Pollut. 228 (3), 95. http://dx.doi.org/10.1007/s11270-017-3280-1.
Zurita, F., Belmont, M.A., De Anda, J., Cervantes-Martínez, J., 2008. Stress detection by
laser-induced fluorescence in Zantedeschia aethiopica planted in subsurface-flow
treatment wetlands. Ecol. Eng. 33 (2), 110–118. http://dx.doi.org/10.1016/j.
ecoleng.2008.02.004.
Zurita, F., Belmont, M.A., De Anda, J., White, J.R., 2011. Seeking a way to promote the
use of constructed wetlands for domestic wastewater treatment in developing
countries. Water Sci. Technol. 63 (4), 654–659. http://dx.doi.org/10.2166/wst.2011.
229.
Zurita, F., De Anda, J., Belmont, M.A., 2006. Performance of laboratory-scale wetlands
planted with tropical ornamental plants to treat domestic wastewater. Water Qual.
Res. J. Can. 41, 410–417.
Zurita, F., De Anda, J., Belmont, M.A., 2009. Treatment of domestic wastewater and
production of commercial flowers in vertical and horizontal subsurface-flow con-
structed wetlands. Ecol. Eng. 35 (5), 861–869. http://dx.doi.org/10.1016/j.ecoleng.
2008.12.026.