Heliyon 6 (2020) e05038

Contents lists available at ScienceDirect

Heliyon
journal homepage: www.cell.com/heliyon

Research article

Implementation of a full-scale constructed wetland to treat greywater from

tourism in Suluban Uluwatu Beach, Bali, Indonesia
Mayang Christy Perdana a, *, Suwarno Hadisusanto b, Ignasius L.Setyawan Purnama b
a
Graduate School of Environmental Science, Universitas Gadjah Mada, Yogyakarta, 55281, Indonesia
b
Universitas Gadjah Mada, Yogyakarta, 55281, Indonesia

A R T I C L E I N F O A B S T R A C T

Keywords: This original research examines a full-scale subsurface Constructed Wetland (CW) system in Indonesia, where
Environmental science most CW research has been limited to laboratory scale experiments. The CW system was located in Bali and built
Plant biology in 2015 in a single series formation. This study aims to demonstrate the performance of the system in treating
Biotechnology
greywater and examine the nutrient content plants’ above-ground biomass. The CW was arranged in linear
Bioengineering
Hydrology
sequence composed of one unplanted (CW1) and five planted treatments of Iris pseudacorus (CW2), Caladium
Subsurface-flow constructed wetland bicolor (CW3), Rhoe discolor (CW4), Sansevieria trifasciata (CW5) and Heliconia psittacorum (CW6). There has been
Wastewater garden little research on Caladium bicolor, Rhoe discolor and Sansevieria trifasciata in a full-scale CW application. Our
Greywater results showed fluctuating efficiency (%) in the reduction of Biological Oxygen Demand (BOD), Chemical Oxygen
Removal efficiency Demand (COD), Total Suspended Solid (TSS), Oil and Grease (O&G), Nitrate and Phosphate. The highest removal
Plant uptake efficiency for CW1, CW2, CW3, CW4, CW5, CW6 were O&G (63.63%), BOD (90.66%), Nitrate (83.55%), BOD
(80%), BOD (82.88%) and Phosphate (89.93%) respectively. After the experimental period, S. trifasciata and
H. psittacorum experienced a decrease in Total N concentration, while H. psittacorum experienced a decrease in
phosphate in above-ground biomass. Species of R. discolor, C. bicolor and I. pseudacorus showed good performance
in terms of their growth and development. Although high removal efficiency was observed at certain times, this
study showed the negative removal efficiencies at times among parameters as a consequence of the low Hydraulic
Retention Time (HRT) and high Hydraulic Loading Rate (HLR).

1. Introduction environmental concern, to build a series of horizontal subsurface flow

constructed wetlands system (HSSFCW) in 2015. This system is known as
As one of the most famous and exotic locations in the world, Bali a WWG (Wastewater Garden) by the local people. Constructed wetland
encounters rapid development of the tourism industry. The tourism in- (CW) denotes an effective and low-cost system which can reduce pol-
dustry uses 65% of local water consumption (Cole, 2012). It has been lutants through physical, chemical and biological processes (Machado
suggested that the disproportionate utilization of water between the et al., 2017). As one of the types of CW, horizontal subsurface constructed
needs for tourism and other sectors diminishes its environmental quality wetlands, have been applied in a wide range of wastewater treatments,
(Sunarta et al., 2014). Moreover, a severe water crisis predicted by 2025 especially for domestic wastewater (Vymazal, 2009; Hua et al., 2017).
if water management is not prioritized (Cole, 2012). Suluban Uluwatu Ornamental plants have also been used in some CWs due to their aes-
Beach is one of the attractive coasts hidden by steep cliffs made of natural thetics and good removal efficiency (Konnerup et al., 2009; Calheiros
limestone formations. Regardless of its apparent beauty, environmental et al., 2015; Mendez-Mendoza et al., 2015). Plants provide the need for
problems related to wastewater are already significant. Before 2015, degrading microbes such as carbon, and habitat for microbes (zones of
wastewater produced by cafes was not treated properly since it was roots, stems and leaves) that are submerged in wastewater (Brix, 2003).
directly discharged to the ravine and infiltrated the limestone or flowed CWs have been studied at a number of different scales. Higher
to the sea. removal efficiency has been shown in laboratory-scale CW compared to
To overcome the problem, a local organization named Project Clean full-scale CW since plants often show better removal efficiency at a
Uluwatu (PCU) collaborated with IDEP, an Indonesian foundation with smaller scale and in controlled environments (Mendez-Mendoza et al.,

* Corresponding author.
E-mail address: [email protected] (M.C. Perdana).

https://doi.org/10.1016/j.heliyon.2020.e05038
Received 24 January 2019; Received in revised form 3 May 2020; Accepted 21 September 2020
2405-8440/© 2020 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
M.C. Perdana et al. Heliyon 6 (2020) e05038

2015). Full-scale CWs have the advantage of providing real world results single-series formation on a slope of around 30
(Figure 1). Day time
in comparison to implementation at micro or meso scales (Brisson and temperature ranged from 27 to 46.6
C. The air humidity ranged from 68
Chazarenc, 2009) and several studies have demonstrated full-scale CWs to 70% in the morning, 30–35% during the day and 66–69.2% in the
can achieve more than 60% removal efficiency (Sim et al., 2008; Vys- evening. Light intensity in the morning and evening ranged from 2.920 to
tavna et al., 2017; Ali et al., 2018). However, full-scale CW experiments 12.000 lux, while midday was more than 20.000 lux.
infrequently use replication due to its cost (Brisson and Chazarenc, 2009;
Stefanakis et al., 2019).
Due to technical reasons, the HSSFCW in Suluban Uluwatu was not 2.2. System design and operational condition
managed for a long time leading to the growth of unwanted wild plants in
the system and decreased aesthetics. In many settings, aesthetics should Before reaching the HSSFCW system, greywater was collected in a
be an important aspect of wastewater treatment (Calheiros et al., 2015). storage tank with a capacity 12,990 L then pumped to the HSSFCW
This study recreates HSSFCW at Suluban Uluwatu by planting five through a pipe with a length of 150 m (Figure 2). In the HSSFCW
ornamental plants (Iris pseudacorus, Caladium bicolor, Rhoeo discolor, wastewater flowed through a series of treatments by gravity (Figure 1). In
Sansevieria trifasciata and Heliconia psittacorum). Iris pseudacorus has been this study, existing materials were utilized such as the concrete tanks and
commonly used in many CW's studies (Wu et al., 2011; Barco and Borin, gravels of Ø 4–6 cm as media. Concrete tanks for each treatment had
2017; Burgos et al., 2017; Sere s et al., 2017) as has H. psittacorum different dimensions (see Table 1) with a minimum of (2.4  1.1  0.6)
(Konnerup et al., 2009; Mendez-Mendoza et al., 2015). Only one labo- m3 to a maximum of (5.6  1.1  0.6) m3. The entire system consisted of
ratory scale study was found for Caladium bicolor (Cuong and Loan, 12 tanks from top to bottom, however, this study only used the first 6
2017). No studies were found for Rhoe discolor or Sansevieria trifasciata. tanks. The sequence of treatment was unplanted (CW1), Iris peudacorus
This study examines the implementation of the HSSFCW series and (CW2), Caladium bicolor (CW3), Rhoe discolor (CW4), Sansevieria tri-
does not compare individual planted treatments or the effect of their fasciata (CW5) and lastly Heliconia psittacorum (CW6). All vegetation and
order because of the constraints of the existing construction. The study litters were removed from the media, then the ornamentals were planted
had three main objectives: (1) to evaluate the performance of the and allowed to acclimatize for two months before testing. The waste-
HSSFCW on wastewater produced by cafes and toilets in Suluban Ulu- water flowed for 2.5 h in a day with HRT for each treatment (tank) was
watu Beach; (2) to examine the initial (the time when plants were firstly 30 min and the discharge rate ranged from 0,242 L/s to 0,561 L/s.
acclimatized)and final (the last water sampling conducted) concentration
of Total N and phosphate in above-ground biomass; and (3) to inspect the
2.3. Wastewater sampling and analyses
effect on the plants (growth and development).
The wastewater parameters measured were BOD, COD, Nitrate,
2. Materials and methods
Phosphate, Oil and Grease, TSS (Total Suspended Solid), pH and tem-
perature. Inlet characteristic can bee seen in Table 2. Six samples were
2.1. Duration of study and site description
taken weekly from the outlet hole of each tanks, except for CW6
(H.psittacorum) which had five samples obtained weekly due to technical
The study was conducted from May to August 2018 at dry season. It
reason. BOD and COD were analyzed by titration, Nitrate was measured
was carried out at the entrance to Suluban Uluwatu Beach (8
480 57.500 S
by Brucine, Phosphate by Ammonium Molybdate, Oil and Grease using
115
050 20.100 E) at the altitude of 197 MASL. The reactor was built in a
gravimetry, and TSS using spectrophotometry with HACH DR/2010. pH

Figure 1. (a) The system outlook (b) CW1: unplanted (c) CW2: I.pseudacorus (d) CW3: Caladium bicolor(e) CW4: Rhoeo discolor(f) CW5: Sansevieriatrifasciata (g) CW 6:
Heliconia psittacorum.

2
M.C. Perdana et al. Heliyon 6 (2020) e05038

where, Cin (mg/kg) is the initial concentration of nutrients and Cfin (mg/
kg) is the final concentration of nutrients. The removal efficiency of all
the treatments can be seen in Table 3.

2.6. Statistical analyses

Plant's growth and development was analyzed with SPSS (IBM SPPS
Statistics 22) before and after greywater treatment in order to examine
the plants' capacity to deal with greywater. Data normality was checked
by the Shapiro-Wilk test. As the data had normal distribution, the Paired-
Samples T-test was used to identify significant differences (p < 0.05) in
plant height, shortest leaf length, and number of shoots.

3. Results and discussion

3.1. Hydraulic Loading Rate (HLR)

Table 4 presents the value of HLR for every treatment. The highest
value of HLR was represented by CW1 (1.53m3 day1 m2) while the
lowest one was represented by CW6 (0.65 m3 day1 m2). The HLR tends
to be in descending order from CW1 to CW 6. The plummeted efficiency
could also be triggered by the high value of HLR. A study by Çakir et al.
(2015) testified that there was decreasing removal rate in line with the
increasing of HLR namely 80.4,74.5 and 62.1% were attained as a result
of 0.050; 0.075 and 0.125 m3 day1 m2 HLR, respectively. While a
research conducted by Abou-Elela et al. (2017) revealed that a decreasing
HLR from 0.18 to 0.10m3 day1 m2 improved the removal percentage
on municipal wastewater in pilot scale horizontal subsurface CW.

3.2. Reduction in BOD (Biological Oxygen Demand)

Reduction in BOD by CW1 reached a maximum of 54.4% and a

Figure 2. Schematic design of the system.
minimum of -52.8% (calculated using Eq. (1)). The negative value
appeared only in the first sampling. In CW2, negative values were
and temperature were measured directly on-site using a digital pH meter observed for the first four weeks of sampling, before it reached the
ATC and glassware thermometer. maximum value (90.66%) in the last week of sampling. This value is the
highest reduction efficiency among all parameters measured from
treatment CW2 (Figure 3). BOD in CW3 showed the highest reduction of
2.4. Plant nutrient sampling and analyses
65.7% (week 6) and had negative values in the second and third week of
sampling. CW4 had the highest value of 80% (week 2) but had negative
Above ground biomass was obtained at the time of planting and after
values in the first and last week of sampling. Reduction of BOD had
14 weeks, by sampling leaves and stems picking from all the plants. Total
positive values from the first to fourth of sampling in CW5. The highest
N and phosphate in above-ground biomass were measured with the
reduction of BOD was 82.88% for CW5. CW6 experienced only one
methods SNI-4721-1998 and Isric 6th 2002 respectively. For Total N, the
negative value (-83.05%) in the fourth week.
sample was digested by 1 gr selenium mixture and 3,5 ml H2SO4 for 4 h
In the system, the removal of polluting organic matter is degraded
then distilled to a receiving flask containing 10 ml boric acid and Conway
aerobically and anaerobically by microorganisms (Vasudevan et al.,
indicator. After distillation finished, it was titrated with H2SO4. Phos-
2011). In addition to biochemical reactions, a decrease in BOD can also
phate was measured by diluting sample with 10 ml of HNO3 in a vessel
occur through a physical process, namely through the precipitation and
then heated inside a Microwave Digester at 200
C for 15 min. After-
attachment of particulates to gravels (Abou-Elela, 2017). However, the
wards, the filtrate was added to a solution of concentration Ammonium
negative removal efficiency can occur due to decay of plant organs inside
hepta molybdate and SnCl2. Phosphate concentration was measured with  s et al., 2017) and HRT did not allow for adequate
the substrate (Sere
an absorption spectrophotometer (610 nm wave length).
contact time and establishment of anaerobic conditions in the beds.

2.5. Data calculations 3.3. Reduction in COD (Chemical Oxygen Demand)

Treatment efficiency for each of the parameters was calculated as the CW1 showed only one positive value in the fourth week of sampling
removal percentage: (17.86%). CW2 reached 82% of reduction efficiency in the last week of
sampling and had three negative values in the first three weeks. CW3
Removal efficiency (%) ¼ (Cin-Cout)/Cin x 100 (1) revealed only negative value in the third week and 44.4% was the highest
reduction efficiency observed in week 6. CW4 showed negative values
where, Cin (mg/l) is the inlet concentration and Cout (mg/l) is the outlet
concentration. until the last week of sampling (-36.75%) with 27.78% of the highest
reduction. CW5 reached the highest removal (76.47%) in the fifth week
Difference (%) of concentration for nutrient contained in above-
ground biomass: of sampling and had -4.54 and -160 % as two negative values in the third
and fourth week. CW6 reached 80.77% of reduction in the third week of
Difference (%)¼ (Cin-Cfin)/Cin x 100 (2) sampling with two negative values at the first two weeks.

3
M.C. Perdana et al. Heliyon 6 (2020) e05038

Table 1. Specific dimension of six tanks.

Tank Length (m) Width (m) Surface Area (m2) Depth (m)
CW1 2.4 1.1 2.64 0.6
CW2 2.4 1.1 2.64 0.6
CW3 3 1.1 3.3 0.6
CW4 4 1.1 4.4 0.6
CW5 4.4 1.1 4.84 0.6
CW6 5.6 1.1 6.16 0.6

Table 2. Inlet characteristic.

Water Quality Parameter Average SD Max Min

TSS 280.17  83.71 433 153
Nitrate 1.49  0.52 2.195 0.639
BOD 197.67  94.98 391.5 100.4
COD 285.33  90.14 423.12 199.2
Phosphate 1.76  1.54 4.9 0.19
Oil and Grease 0.177  0.085 0.32 0.08
pH 6.6  0.28 7.2 6.3
Temperature 26.5  0.35 27 26.1

Decrease in COD can occur through degradation carried out by het- In many studies, denitrification is considered a major mechanism for
erotrophic microorganisms both anaerobically and aerobically (Priya nitrate reduction in the CW system (Matheson and Sukias, 2010; Saeed
et al., 2013) and also through physical processes such as sedimentation, and Sun, 2012; Merino-Solis et al., 2015). Some studies have shown that
filtration (Merino-Solis et al., 2015; Sudarsan et al., 2017) and deposi- plants contribute to the reduction of nitrates not only through direct
tion(Tilak et al., 2017). The physical processes are more important in uptake, but also as a source of organic carbon which can trigger deni-
reducing COD than biological processes (Sudarsan et al., 2017). trification (Lin et al., 2002; Almeida et al., 2017). However, some studies
Increase in COD can be caused by an increase of organic load due to suggest that plants only have a small contribution to reduce nitrate in CW
decay of plant organs that are still left in the substrate. It can also be and their significance is still debated (Shelef et al., 2013; Li et al., 2013).
caused by the existence of a non-biodegradable substances which is Matheson and Sukias (2010) show that plant uptake contributes up to
shown by the BOD/COD ratio value. Values below 0.3 indicate that the about 9% of the total nitrate reduction in the system, while 71% of nitrate
organic load in wastewater is non-biodegradable (Abou-Elela, 2017). The reduction is dominated by denitrification. Plant uptake is just another
wastewater source from cafes/restaurants and potential cleaning chem- mechanism for reducing nitrate but when the plant has withered and
icals is also presumably contributed the low value of BOD/COD ratio. dies, the nutrients contained in the tissue can return to the system. This is
This study found some data with BOD/COD ratios below 0.3, specifically in accordance with the observation of this study that nitrate concentra-
0.24 (CW3: 6th week of sampling); 0.22 (CW4: 6th week of sampling); 0.1 tions at the outlets varied significantly and were not uniformly reduced
and 0.15 (CW5: 3rd and 5th week of sampling, respectively); 0.075, 0.08, by the presence of plants (Figure 3e). The uptake of nitrate can be
and 0.29 (CW6: 1st, 2nd and 3rd week of sampling, respectively). inhibited by various factors, including the presence of competing
The non-biodegradable components are suspected to be lignin and chemicals. Nitrite can inhibit nitrate uptake because it has the same
hemicellulose derived from the wild plant organs left in substrate before transporter and binding site. Ammonium is also considered to be an in-
the study was carried out. Lignin is one of the constituent components of hibitor by disrupting the affinity of the transport system (Bose dan Sri-
plants along with cellulose and other fiber ingredients that make up plant vastava, 2001).
cells. Lignin and hemicellulose are organic materials that are not easily A second possible cause of reduced removal efficiency in nitrate by
degraded by microbes. The solubility of natural lignin in water is less at CW systems is low activity of denitrification or nitrate ammonification. If
low pH (Zhu and Theliander (2011). Other factors which can increase the aerobic degradation process has high activity, the activity of anaer-
COD include the presence of glucose, fructose, ethanol and acetic acid obic degradation processes such as denitrification will be reduced (Hua
(Welz et al., 2011). However, those substances were not measured in this et al., 2017). This statement is reinforced by Stein and Hook (2005) who
study. showed that high nitrification activity and COD, which indicates high
aerobic activity connected with low anaerobic degradation.
3.4. Removal of nitrate
3.5. Removal of phosphate
Removal of nitrate in CW1 showed negative values in weeks 1,2,3, 6
Phosphate removal for CW1 showed negative values in the first four
of sampling with only 13.23% of the highest reduction (the removal was
weeks of sampling. The maximum value was 18.79% in the last week.
calculated by Eq. (1)). CW2 experienced -323% removal efficiency only
CW2 reached the highest removal efficiency of 90.62%, while CW3
in the last week but had 50.63% as the highest removal at the fifth week.
showed a negative value only in the first week and reached removal of
CW3 reached 83.55% removal in the sixth weeks and had negative value
49.27%. CW4 still experienced a negative value in the last week of
only in the fifth week of sampling. CW4 reached 28.82% as the highest
sampling and reached 71.87% removal in the second week. CW5 showed
value of removal in at the second week. A negative value also occurred in
negative value only in third week (-16.17%), while CW6 had negative
CW5 (-96.59%) for the last week of sampling. CW6 had 78.1% of removal
at the fifth week of sampling.

4
M.C. Perdana et al. Heliyon 6 (2020) e05038

-56.54
-93.16
-114.8

-83.05
-100

-400
Min
Table 4. Value of HLR for each tank.

Tank HLR (m3 day1 m2)

78.1
89.9
74.3
66.7
62.6
80.8
Max
CW1 1.53
CW2 1.53

-64.7160  194.03
-11.4440  68.14
CW3 1.22

16.0140  57.47
-3.3320  60.55
6.9540  77.54
13.1180  61.5
CW4 0.92
Average SD CW5 0.83
CW6 0.65
CW 6

-16.2
-11.5

-20.6
-859

-100

-160
values for weeks 4,5 and 6 with 89.93% as the highest removal at the
Min

second week of sampling.

Phosphate can be reduced by physical-chemical reactions related to
46.63
62.10
68.18
62.50
82.88
76.47
Max

adsorption, precipitation (Burgos et al., 2017) and ion exchange (Mojiri

et al., 2017) on CW media. Phosphate reduction can also be caused by
10.1033  88.48

uptake by plants (Vera et al., 2014; Haritash et al., 2015) but the plant's
30.9817  30.6
36.0917  37.2

38.6083  46.2
-343.8  384.8

4.8600  56.5
Average SD

uptake capacity is considered limited (Vymazal, 2004).

Increases in dissolved of phosphate may have occurred due to
CW 5

changes in chemistry causing dissociation. If the media is anaerobic, Fe

(III) will be reduced to Fe (II) which will result in a weak bond with
phosphate (Verhoeven and Meuleman, 1999). As a result, phosphate
-143.7
-307.7
-100.5

-124.7
-114.6

concentration will increase in the outlet. Increase of phosphate value

Min

-60

could also be caused by the release from dead plant organs in the sub-
strate present before the study was carried out. Dead plant material is
28.8
34.4
63.8
72.2

71.9
Max

80

thought to be a reason since the system was densely populated with wild
plants prior to this experiment. Nutrients can return to the system due to
-85.28  154.7
-15.65  66.4

wilting plants and can affect the effectiveness of CW reduction (Menon

Average SD

15.9  61.27

6.58  74.14
-13.6  6.11
21.9  51.6

and Holland, 2014). Low phosphorus removal could be also attributed to

the fact that the units were not new, hence the gravel media was probably
CW 4

already saturated from previous operation, providing limiting adsorption

sites (Tatoulis et al., 2017).
-42.7
-23.9

-19.1
-900
Min

-26

-25

3.6. Removal of TSS (Total Suspended Solid)

83.6
49.3
63.6
64.3
65.7
44.4
Max

Removal efficiency of TSS in CW1 showed positive values in weeks

-160  367.5
22.52  40.2
21.55  27.6

19.29  23.5
Average SD

2,3,4 and 5 with 33% as the highest value. CW2 showed a negative value
26.9  38.1

8.8  32.7

of removal only in the third week with 86.85% as the highest removal for
CW 3

the last week. CW3 reached 63.6% of removal as the highest value of
removal in the second week with two negative values at the first and
fourth week. CW4 had 63.79% as the highest value for removing TSS
-26.8
-31.5
-323

-150
-132
Min

with only one negative value at the first week. CW5 and CW6 had 68.18%
-59

and 74.34% respectively as the highest removal. Both CW5 and CW6 had
50.6
90.6
86.9
83.3
90.7
Max

negative value for the last week of sampling.

82

Removal efficiency of TSS was generally low (Figure 3c). Removal

efficiency of TSS. The low removal efficiency of TSS could be caused by
-35.19  142.7

-12.5  77.77
Average SD

17.27  38.4

-16.7  75.7

low HRT (Hydraulic Retention Time). Low HRT causes short duration of
22.26  49

7.2  49

contact of wastewater with the substrate, so that the physical process

CW 2

does not occur optimally. Panrare et al. (2015) stated that HRT extension
Table 3. Removal effectiveness (%) of all treatments.

would provide a better reduction in TSS parameters, even though it was

applied to non-plant treatments. Merino-Solís et al. (2015) revealed that
-33.3
-52.8
-19.5
-265
-591
Min

33

HRT that is good maximize effectiveness of HSSFCW varies from 3 to 15

days depending on the type of pollutant in the waste.
13.23
18.79

63.63
54.44
17.86
-39.5
Max

3.7. Removal of Oil and Grease (O&G)

54.44  39.01
-49.4  106.9

-39.45  30.7
-120  236.8
Average SD

29.7  35.17

-5.59  13.7

CW1 removed 63.63% O&G in the fourth week as the highest removal
value. CW2 reached 83.33% as its highest removal in the last week of
CW 1

sampling. CW3 showed -900% as the lowest removal efficiency in the last
week and had 64.28% as the highest value of removal at the third week.
Oil and Grease
Water Quality

CW4 showed 72.22% as its highest value of removal in the second week,
Phosphate
Parameter

while CW5 showed 33.33% of removal both in the first and the last week.
Nitrate

However, CW6 still showed negative value until the last week of sam-
COD
BOD
TSS

pling (-100%).

5
M.C. Perdana et al. Heliyon 6 (2020) e05038

The removal of Oil and Grease (O&G) can be conducted by plants via 3.8. Nutrients in above-ground biomass
translocation and accumulation in the plant's body. A study conducted by
Mustapha et al. (2018) provided evidence that T. latifolia had the capa- Nutrients which are measured in above-ground biomass was Total N
bility to accumulate O&Gs in its tissue and decrease 70–80% of O&G in and Phosphate concentration (mg/g). It represents the initial and final
the system. Plants' root can also physically trap O&G (Winanti et al., concentration (before and after the wastewater flowed to the plants). The
2018). difference of the concentrations was determined using Eq. (2).
The drop of O&G efficiency at CW4 and CW5 in the present study I. pseudacorus showed an increase 71.6% of Total N and 27.5% of Phos-
could be related to tank leakage. Leakage can lead to greater O&G phate. C. bicolor had a slight increase of Total N (15.7%), but there was no
retention on the surface of the substrate. After leaks in the system were any phosphate detected in the biomass. R, discolor also showed a slight
repaired at CW4 and CW5, a higher level of O&G was measured at the increase of Total N (4.5%) but it had more increase in Phosphate
outlet. It is suggested that when liquid waste could flow constantly, the (20.9%). S. trifasciata experienced a decrease of Total N (-24.8%) and
remnants of oil and fat left in the system then rose and experienced a even a slight increase of Phosphate (5.9%). H. psittacorum showed a
wash-out, increasing the concentration of oil and fat at the outlet. decrease of Total N and Phosphate namely -33.8% and -45.4% respec-
tively. Further detail of concentration can be seen in Table 5.

BOD COD
150 200
100 100
Effectiveness (%)

Effectiveness (%)
CW 1 CW 1
0
50 CW 2 CW 2
-100 1 2 3 4 5 6
0 CW 3 CW 3
1 2 3 4 5 6 -200
CW 4 CW 4
-50
-300
CW 5 CW 5
-100 -400
CW 6 CW 6
-150 -500
Week Week

a b

TSS Oil and Grease

200
100 0
Effectiveness (%)

CW 1 CW 1
1 2 3 4 5 6
50 -200 CW 2
Effectiveness (%)

CW 2
CW 3 -400 CW 3
0
CW 4 CW 4
1 2 3 4 5 6 -600
-50 CW 5 CW 5
-800
CW 6 CW 6
-100 -1000
Week
-150 Week

c d

Nitrate
200
0 CW 1
Effec veness (%)

1 2 3 4 5 6
-200 CW 2
-400 CW 3

-600 CW 4
CW 5
-800
CW 6
-1000
Week

Phopshate
200
CW 1
Effec veness (%)

0
1 2 3 4 5 6 CW 2
-200
CW 3
-400
CW 4
-600 CW 5

-800 CW 6
Week

Figure 3. Removal Efficiency in each of treatments for (a) BOD (b) COD (c) TSS (d) Oil and Grease (e) Nitrate (f) Phosphate.

6
M.C. Perdana et al. Heliyon 6 (2020) e05038

Table 5. Concentration of Total N and Phosphate before and after system worked in plants’ above-ground biomass.

Treatment Total N (mg/kg) Phosphate (mg/kg)

Initial Final Initial Final

CW 2 1200.454 2059.986 115.35 147.09
CW 3 3307.562 3828.63 247.87 Not detected
CW 4 2686.7 2808.637 206.7 249.99
CW 5 2069.344 1556.133 123.58 130.96
CW 6 2473.785 1635.287 196.03 106.75

All plants have a higher tendency to uptake N (Figure 4) than Phos- had showed good performance in dealing with greywater in a previous
phate (Figure 5). Minimum N uptake can reach 10 times higher than study (Perdana et al., 2018). Similar results were also found for Caladium
phosphate. Barco and Borin (2017) in their study stated that plants tend bicolor and Rhoe discolor which encountered good growth but no flow-
to store a smaller amount of P compounds than N compounds in their ering occurred. In the first and second weeks of acclimatization, both of
roots, especially I.pseudacorus plants. Our research supports this obser- these plants seemed withered. The R. discolor showed its wilting marked
vation, demonstrating that above-ground biomass, N levels are up to 10 by brownish leaves on the underside. However, when entering the third
times higher than Phosphate (Table 5). Total N was higher than Phos- week of acclimatization until the end of the sampling, both showed
phate in all test species. Wu et al. (2011) also revealed that I.pseudacorus remarkable growth and development. C.bicolor and R.discolor produced a
has an N uptake capacity greater than P because P was bound tighter by lot of shoots and clumps. The growth of the leaves reaches almost twice of
the substrate. In this study, the increase in total N uptake in I. pseudacorus the initial length. Both plants experienced significant growth (p < 0.05).
before and after the system was 71.6% while the phosphate uptake was Unsatisfactory growth and development were found in S. trifasciata
27.5%. and H.psittacorum. A number of leaves on each plant of S.trifasciata were
A decrease in total N concentration in above-ground biomass was withered. It is suspected that this plant could survive in the CW system
found in S. trifasciata (CW5) and H.psittacorum (CW6) (Figure 4) while a but may not reproduce. In another study, Weedon (2016) also found poor
decrease in phosphate was found in H. psittacorum (CW6) and possibly development presented by P.australis in CW system, showing that certain
C.bicolor (CW3) (Figure 5). This could be caused by a number of factors. plants cannot always adapt successfully. However, in this study some of
Given that the reduction of N and Phosphate was observed in the final the difficulties in growth experienced by S.trifasciata (CW5) and H.psit-
tanks of the system, it is possible that there was a reduction in available N tacorum (CW6) may be attributed to the technical problem (leakage) at
and Phosphate causing deficiency. Because of the previously mentioned CW5 at the beginning of acclimatization. This could have caused both
reasons for temporal fluctuations in greywater concentrations of Nitrate reduced water and nutrient availability to the plants in CW5 and CW6. In
and Phosphate, it is suggested that the point measurements of Nitrate and addition, the T-test showed that there was no significant difference in
Phosphate may not indicate the average concentrations over time. In terms of their growth and development before and after the experimental
contrast the total N and Phosphate of plants indicates the accumulation period. Some of the roots of the wild plants that had grown before the
or loss during the experimental period. Another reason for variation study were suspected to have decayed in the substrate and lead to the
could be related to interspecies variation in transpiration. increase of organic loads.
Sudarsan et al. (2017) pointed out that the uptake of nitrate and
phosphate carried out by leaves is the smallest among other parts of the 4. Conclusion
plant because of the process of stomata transpiration in the leaves.
Another possibility is that N and P have been used by metabolic processes This study denotes fluctuating removal efficiency among parameters
in the plant. This assumption is supported by Collins et al. (2006) which measured and the negative removal efficiency was promoted mainly due
revealed that substances stored in the body of the plant can be metabo- to lack of HRT and high value of HLR. As less known ornamental plants,
lized resulting the reduction of nutrient concentration in plant tissues. C.bicolor and R.discolor are recommended to be applied in CW for their
endurance. Both of them are also promising to be implemented toward
3.9. Plant growth and development other types of wastewater for future research.

During the study, I.pseudacorus showed significant growth (plant

height, paired T-Test, p < 0.05) and also produced flowers). I.pseudacorus

4500 300

4000
250
3500
Concentra on (mg/g)

Concentra on (mg/g)

3000 200

2500
150
2000

1500 100

1000
50
500

0 0
I.pseudacorus C.bicolor R.discolor S.trifasciata H.psi acorum I.pseudacorus C.bicolor R.discolor S.trifasciata H.psi acorum

Figure 4. Total N concentration in above-ground biomass. Figure 5. Phosphate concentration in above-ground biomass.

7
M.C. Perdana et al. Heliyon 6 (2020) e05038

Declarations Cole, S., 2012. A political ecology of water equity and tourism: a case study from Bali.
Ann. Tourism Res. 39, 1221–1241.
Collins, C., Fryer, M., Grosso, A., 2006. Plant uptake of non-ionic organic chemicals.
Author contribution statement Environ. Sci. Technol. 40, 45–52.
Cuong, Nguyen Xuan., Loan, Nguyen Thi., 2017. Performance evaluation of three native
M.C. Perdana: Conceived and designed the experiments; Performed plants for sewage wastewater treatment in constructed wetland. International
Journal of Environment and Pollution Research 5, 1–7.
the experiments; Analyzed and interpreted the data; Contributed re- Haritash, A.K., Sharma, A., Bahel, K., 2015. The potential of Canna lily for wastewater
agents, materials, analysis tools or data; Wrote the paper. treatment under Indian conditions. Int. J. Phytoremediation 17, 999–1004.
S. Hadisusanto: Conceived and designed the experiments; Analyzed Hua, Y., Peng, L., Zhang, S., Heal, K.V., Zhao, J., Zhu, D., 2017. Effects of plants and
temperature on nitrogen removal and microbiology in pilot-scale horizontal
and interpreted the data; Contributed reagents, materials, analysis tools subsurface flow constructed wetlands treating domestic wastewater. Ecol. Eng. 108,
or data. 70–77.
I. L. Setyawan Purnama: Conceived and designed the experiments; Konnerup, D., Koottatep, T., Brix, H., 2009. Treatment of domestic wastewater in tropical,
subsurface flow constructed wetlands planted with Canna and Heliconia. Ecol. Eng.
Analyzed and interpreted the data; Contributed reagents, materials, 35, 248–257.
analysis tools or data. Li, L., Yang, Y., Tam, N.F., Yang, L., Mei, X.Q., Yang, F.J., 2013. Growth characteristics of
six wetland plants and their influences on domestic wastewater treatment efficiency.
Ecol. Eng. 60, 382–392.
Funding statement Lin, Y.F., Jing, S.R., Wang, T.W., Lee, D.Y., 2002. Effects of macrophytes and external
carbon sources on nitrate removal from groundwater in constructed wetlands.
This work was supported by Indonesia Endowment Fund for Educa- Environ. Pollut. 119, 413–420.
Machado, A.I., Beretta, M., Fragoso, R., Duarte, E., 2017. Overview of the state of the art
tion (LPDP). of constructed wetlands for decentralized wastewater management in Brazil.
J. Environ. Manag. 187, 560–570.
Matheson, F.E., Sukias, J.P., 2010. Nitrate removal processes in a constructed wetland
Competing interest statement
treating drainage from dairy pasture. Ecol. Eng. 36, 1260–1265.
Mendez-Mendoza, A.S., Bello-Mendoza, R., Herrera-L opez, D., Mejía-Gonzalez, G.,
The authors declare no conflict of interest. Calixto-Romo, A., 2015. Performance of constructed wetlands with ornamental plants
in the treatment of domestic wastewater under the tropical climate of South Mexico.
Water Pract. Technol. 10, 110–123.
Additional information Menon, R., Holland, M.M., 2014. Phosphorus release due to decomposition of wetland
plants. Wetlands 34, 1191–1196.
Merino-Solís, M.L., Villegas, E., de Anda, J., L opez-Lopez, A., 2015. The effect of the
Data associated with this study has been deposited at Mendeley (Title: hydraulic retention time on the performance of an ecological wastewater treatment
Inlet and outlet measurement from WWG in Suluban) under the https:// system: an anaerobic filter with a constructed wetland. Water 7, 1149–1163.
doi.org/10.17632/btj35rr277.1. Mojiri, A., Ahmad, Z., Tajuddin, R.M., Arshad, M.F., Gholami, A., 2017. Ammonia,
phosphate, phenol, and copper (II) removal from aqueous solution by subsurface and
surface flow constructed wetland. Environ. Monit. Assess. 189, 337.
Acknowledgements Mustapha, H.I., van Bruggen, H.J.J.A., Lens, P.N., 2018. Vertical subsurface flow
constructed wetlands for the removal of petroleum contaminants from secondary
refinery effluent at the Kaduna refining plant (Kaduna, Nigeria). Environ. Sci. Pollut.
We are thankful to Bapak Gede Sugiarta from IDEP Foundation for the
Control Ser. 1–12.
guidance and information concerning the WWG system in Suluban Ulu- Panrare, A., Sohsalam, P., Tondee, T., 2015. Constructed wetland for sewage treatment
watu and to Buya from PCU (Project Clean Uluwatu) for permitting and and thermal transfer reduction. Energy Procedia 79, 567–575.
Perdana, M.C., Sutanto, H.B., Prihatmo, G., 2018. Vertical Subsurface Flow (VSSF)
providing the existing reactor and study site.
constructed wetland for domestic wastewater treatment. IOP Conf. Ser. Earth
Environ. Sci. 148, 012025.
References Priya, Sharma, G., Brighu, U., 2013. Comparison of different types of media for nutrient
removal efficiency in vertical upflow constructed wetland. Int. J. Environ. Eng.
Abou-Elela, S.I., 2017. Constructed Wetlands: the Green Technology for Municipal Manag. 4, 2231–2319.
Wastewater Treatment and Reuse in Agriculture. Saeed, T., Sun, G., 2012. A review on nitrogen and organics removal mechanisms in
Abou-Elela, S.I., Elekhnawy, M.A., Khalil, M.T., Hellal, M.S., 2017. Factors affecting the subsurface flow constructed wetlands: dependency on environmental parameters,
performance of horizontal flow constructed treatment wetland vegetated with operating conditions and supporting media. J. Environ. Manag. 112, 429–448.
 s, M., Hnatkova, T., Vymazal, J., Vanek, T., 2017. Removal efficiency of constructed
Sere
Cyperus papyrus for municipal wastewater treatment. Int. J. Phytoremediation 19
(11), 1023–1028. wetland for treatment of agricultural wastewaters. Chemistry Journal of Moldova:
Ali, M., Rousseau, D.P., Ahmed, S., 2018. A full-scale comparison of two hybrid Gene (Amst.) 12 (1), 45–52.
constructed wetlands treating domestic wastewater in Pakistan. J. Environ. Manag. Shelef, O., Gross, A., Rachmilevitch, S., 2013. Role of plants in a constructed wetland:
210, 349–358. current and new perspectives. Water 5, 405–419.
Almeida, A., Carvalho, F., Imaginario, M.J., Castanheira, I., Prazeres, A.R., Ribeiro, C., Sim, C.H., Yusoff, M.K., Shutes, B., Ho, S.C., Mansor, M., 2008. Nutrient removal in a pilot
2017. Nitrate removal in vertical flow constructed wetland planted with and full scale constructed wetland, Putrajaya city, Malaysia. J. Environ. Manag. 88,
Vetiveriazizanioides: effect of hydraulic load. Ecol. Eng. 99, 535–542. 307–317.
Barco, A., Borin, M., 2017. Treatment performance and macrophytes growth in a restored Sudarsan, J.S., Annadurai, R., Mukhopadhyay, M., Chakraborty, P., 2017. Domestic
hybrid constructed wetland for municipal wastewater treatment. Ecol. Eng. 107, wastewater treatment using constructed wetland: an efficient and alternative way.
160–171. Sustain. Water Resour. Manag.
Burgos, V., Araya, F., Reyes-Contreras, C., Vera, I., Vidal, G., 2017. Performance of Sunarta, I.N., Mahendra, M.S., Wiranatha, A.S., Paturusi, S.A., 2014. Tourism
ornamental plants in mesocosm subsurface constructed wetlands under different development impacts on water resources in northern kuta district of badung Bali. E-
organic sewage loading. Ecol. Eng. 99, 246–255. Journal of Tourism 1, 160–175.
Bose, B., Srivastava, H.S., 2001. Absorption and accumulation of nitrate in plants: Stefanakis, A.I., Bardiau, M., Silva, D., Taylor, H., 2019. Presence of bacteria and
influence of environmental factors. Indian J. Exp. Biol. 101–110. bacteriophages in full-scale trickling filters and an aerated constructed wetland. Sci.
Brisson, J., Chazarenc, F., 2009. Maximizing pollutant removal in constructed wetlands: Total Environ. 659, 1135–1145.
should we pay more attention to macrophyte species selection? Sci. Total Environ. Stein, O.R., Hook, P.B., 2005. Temperature, plants, and oxygen: how does season affect
407 (13), 3923–3930. constructed wetland performance? J. Environ. Sci. Health 40, 1331–1342.
Brix, H., 2003. Plants Used in Constructed Wetland and Their Functions,1st International Tatoulis, T., Akratos, C.S., Tekerlekopoulou, A.G., Vayenas, D.V., Stefanakis, A.I., 2017.
Seminar on the Use of Aquatic Macrophytes for Wastewater Treatment in Constructed A novel horizontal subsurface flow Constructed Wetland: reducing area requirements
Wetland, pp. 1–30. and clogging risk. Chemosphere 186, 257–268.
Çakir, R., Gidirislioglu, A., Çebi, U., 2015. A study on the effects of different hydraulic Tilak, A.S., Wani, S.P., Datta, A., Patil, M.D., Kaushal, M., Reddy, K.R., 2017. Evaluation
loading rates (HLR) on pollutant removal efficiency of subsurface horizontal-flow of Ageratum conyzoides in field scale constructed wetlands (CWs) for domestic
constructed wetlands used for treatment of domestic wastewaters. J. Environ. Manag. wastewater treatment. Water Sci. Technol. 75, 2268–2280.
164, 121–128. Vera, I., Araya, F., Andres, E., Saez, K., Vidal, G., 2014. Enhanced phosphorus removal
Calheiros, C.S., Bessa, V.S., Mesquita, R.B., Brix, H., Rangel, A.O., Castro, P.M., 2015. from sewage in mesocosm-scale constructed wetland using zeolite as medium and
Constructed wetland with a polyculture of ornamental plants for wastewater artificial aeration. Environ. Technol. 35, 1639–1649.
treatment at a rural tourism facility. Ecol. Eng. 79, 1–7.

8
M.C. Perdana et al. Heliyon 6 (2020) e05038

Vasudevan, P., Griffin, P., Warren, A., Thapliyal, A., Tandon, M., 2011. Localized Winanti, E.T., Rahmadyanti, E., Fajarwati, I.N., 2018. Ecological approach of campus
domestic wastewater treatment: part I-constructed wetlands (an overview). J. Sci. wastewater treatment using constructed wetland. IOP Conf. Ser. Mater. Sci. Eng. 288,
Ind. Res. 583–594. 1–9.
Verhoeven, J.T., Meuleman, A.F., 1999. Wetlands for wastewater treatment: Weedon, C.M., 2016. Tertiary sewage treatment by a full-scale compact vertical flow
opportunities and limitations. Ecol. Eng. 12, 5–12. constructed wetland. Environ. Technol. 38, 140–153.
Vymazal, J., 2004. Removal of phosphorus in constructed wetlands with horizontal sub- Welz, P.J., Ramond, J.B., Cowan, D.A., Prins, A., Burton, S.G., 2011. Ethanol degradation
surface flow in the Czech Republic. Water Air Soil Pollut. Focus 4, 657–670. and the benefits of incremental priming in pilot-scale constructed wetlands. Ecol.
Vymazal, J., 2009. The use constructed wetlands with horizontal sub-surface flow for Eng. 37, 1453–1459.
various types of wastewater. Ecol. Eng. 35, 1–17. Wu, H., Zhang, J., Li, P., Zhang, J., Xie, H., Zhang, B., 2011. Nutrient removal in
Vystavna, Y., Frkova, Z., Marchand, L., Vergeles, Y., Stolberg, F., 2017. Removal constructed microcosm wetlands for treating polluted river water in northern China.
efficiency of pharmaceuticals in a full scale constructed wetland in East Ukraine. Ecol. Ecol. Eng. 37 (4), 560–568.
Eng. 108, 50–58. Zhu, W., Theliander, H., 2011. Equilibrium of lignin precipitation. 16th international
symposium on wood. Fiber and Pulping Chemistry-Proceedings, ISWFPC, Tianjin 1,
195–199.