9.1 Perkenalan
Lahan basah buatan (CWs) adalah sistem pengolahan yang direkayasa untuk
mengoptimalkan proses pengolahan yang ditemukan di lingkungan alami dan oleh
karena itu dianggap sebagai solusi yang berkelanjutan dan ramah lingkungan untuk
permasalahan teknik. Proses yang terjadi di CW sangat kompleks, dan mencakup
sejumlah besar proses fisik, kimia, dan biologis aktif secara simultan yang saling
mempengaruhi satu sama lain. CW adalah sistem yang secara efisien mengolah
berbagai jenis air yang tercemar (misalnya Kadlec dan Wallace, 2009; García et al.,
2010).
Sejak lama, CW dianggap sebagai ‘kotak hitam’. Sedikit upaya yang dilakukan untuk
memahami proses utama yang mengarah pada penghilangan kontaminan. Saat ini,
sebagian besar model lahan basah masih menggunakan pendekatan ‘kotak hitam’ – model
tersebut tidak mempertimbangkan proses di lahan basah secara rinci. Untuk model ‘kotak
hitam’, data dari eksperimen diperlukan untuk memperoleh persamaan model. Dalam
model berbasis proses, persamaan matematika didasarkan pada proses di lahan basah dan
mencakup persamaan keseimbangan energi, massa, muatan, dan sebagainya. Data dari
eksperimen digunakan untuk kalibrasi dan validasi model. Prediksi yang lebih baik dapat
dilakukan dengan menggunakan model ini (Langergraber, 2011).
Saat mengembangkan model lahan basah, sejumlah proses berbeda harus
dipertimbangkan (Langergraber et al., 2009b):
● model aliran (menggambarkan aliran air pada media berpori);
● model transpor (menggambarkan transpor unsur serta proses adsorpsi dan
desorpsi);
● model biokinetik (menggambarkan transformasi biokimia dan proses degradasi
unsur-unsurnya);
Pemodelan Transportasi Reaktif: Aplikasi dalam Masalah Energi Bawah Permukaan dan Lingkungan,
Edisi pertama. Diedit oleh Yitian Xiao, Fiona Whitaker, Tianfu Xu dan Carl Steefel.
© 2018 John Wiley & Sons Ltd. Diterbitkan 2018 oleh John Wiley & Sons Ltd.
454 Pemodelan Transportasi
Tabel 9.1 Perbandingan alat simulasi terbaru untuk lahan basah yang dibangun.
Referensi Langergraber dan Šimůnek Samsó dan Garcia (2013) Mburu dll. (2012)
(2012)
Model Tidak ada aliran
Richards equation Tabel aliran variabel
Aliran
(variably saturated flow) (saturated flow)
Model Advection, dispersion, Advection, dispersion, Tidak ada model
trasnpor
adsorption adsorption
model pada umumnya dan modul HYDRUS Wetland pada khususnya. Kami
selanjutnya menjelaskan tantangan-tantangan yang tersisa dalam menggunakan alat-
alat ini dan merangkum kebutuhan untuk pengembangan dan penelitian lebih lanjut
Proses Aerobic dan anoxic (9) Aerobic, anoxic, dan anaerobic (17)
Komponen Oxigen, organic matter, nitrogen,
Oxygen, organic matter, nitrogen,
dan phosphorus (12) dan sulfur (16)
Jenis CW VF CWs VF dan HF CWs
Low‐loaded HF beds
aplikasi yang melibatkan pengolahan air limbah kota di CWs, modul Lahan Basah
HYDRUS juga telah digunakan sebagai model:
● CWs menangani luapan saluran pembuangan gabungan (Dittmer et al., 2005;
Henrichs et al. 2007, 2009; Meyer, 2011)
● CWs mengolah limbah instalasi pengolahan air limbah untuk tujuan irigasi
(Toscano et al., 2009; Pálfy et al., 2015a)
● CWs mengolah air limbah buatan (e.g. Pálfy and Langergraber, 2014; Rizzo et al., 2014)
● CWs mengolah air abu-abu (Karlsson et al., 2015)
● Limpasan dari lokasi pertanian dan dampak dari zona pengelolaan tepi sungai
(Smethurst et al., 2014).
Selain modul Wetland, perangkat lunak HYDRUS juga mencakup modul transpor zat
terlarut multi-komponen HP1/2/3, modul transpor zat terlarut yang difasilitasi koloid
C-Ride, modul untuk fumigan, dan modul ion utama UnsatChem. Semua modul ini
mensimulasikan aliran air dan proses transpor zat terlarut/panas dalam domain
transpor dua dimensi dan didukung oleh antarmuka pengguna grafis HYDRUS
(2D/3D). Banyak proses modul khusus ini juga tersedia sebagai bagian dari
perangkat lunak HYDRUS‐1D domain publik (Šimůnek et al., 2013).
Tabel 9.3 Perbandingan komponen CW2D dan CWM1 pada implementasi HYDRUS.
Pemodelan Transportasi Reaktif Multikomponen di Lahan Basah Buatan
CW2D komponen
1) SO: Oksigen terlarut, O2 Komponen yang larut
SO: Dissolved oxygen, O2
2) CR: Mudah terurai secara hayati, larut
COD 1) SF: COD larut yang dapat difermentasi
dan mudah terbiodegradasi
3) CS: Larut secara perlahan dan dapat terurai
secara hayati COD 2) SA: Produk fermentasi sebagai acetate
4) CI: Larut dalam larutan COD 3) SI: Inert soluble COD
5) XH: Bakteri heterotrofik 4) SNH: Ammonium dan ammonia nitrogen
6) XANs: Bakteri pengoksidasi amonia 5) SNO: Nitrate dan nitrite nitrogen
autotrofik (Nitrosomonas spp.) 6) SSO4: Sulfate sulfur
7) XANb: Bakteri pengoksidasi nitrit 7) SH2S: Dihydrogensulfide sulfur
autotrofik (Nitrobacter spp.) Particulate components
8) NH4N: Ammonium dan ammonia 8) XS: Slowly biodegradable particulate COD
nitrogen 9) XI: Inert particulate COD
9) NO2N: Nitrite nitrogen 10) XH: Heterotrophic bacteria
10) NO3N: Nitrate nitrogen 11) XA: Autotrophic nitrifying bacteria
11) N2: Elemental nitrogen 12) XFB: Fermenting bacteria
12) PO4P: Phosphate phosphorus 13) XAMB: Acetotrophic methanogenic
Komponen tambahan (dalam bacteria
implementasi HYDRUS) 14) XASRB: Acetotrophic sulfate reducing
13) Pelacak non-reaktif bacteria
Nitrifikasi dimodelkan sebagai proses dua 15) XSOB: Sulfide oxidizing bacteria
langkah Komponen tambahan (dalam
implementasi HYDRUS)
16) Pelacak non-reaktif
458 Pemodelan Transportasi
Tabel 9.4 Definisi komponen CW2D dan CWM1 pada fasa cair dan padat (Langergraber dan
Šimůnek, 2011).
Komponen 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
L, didefinisikan dalam fase cair saja; S, didefinisikan hanya dalam fase padat; L + S, didefinisikan dalam fase
cair dan padat.
Tabel 9.5 Perbandingan proses CW2D dan CWM1 (Langergraber dan Šimůnek, 2011).
Tabel 9.8 Matriks stoikiometri CW2D (Langergraber dan Šimůnek, 2005; lihat Tabel 9.11 untuk definisi koefisien stoikiometri ).
i→ 1 2 3 4 5 6 7 8 9 10 11 12
Tabel 9.9 Laju reaksi dalam CW2D (Langergraber dan Šimůnek, 2005; lihat Tabel 9.10
untuk definisi koefisien laju).
Organisme heterotrofik
1) Hydrolysis
cCS /cXH
Kh K cCS /cXH cXH
X
2) Pertumbuhan aerobik heterotrof pada COD yang mudah terurai secara hayati
cO2 cCR
H fN,Het c
XH
K Het,O2 cO2 K Het,CR cCR
5) Lysis of heterotrophs
bH cXH
7) Lysis of Nitrosomonas
bHANs cXANs
9) Lysis of
Nitrobacter bHANb
cXANb
cXY
sXY , H, ANs, ANb
whereY
Langergraber (2007).
Pemodelan Transportasi Reaktif Multikomponen di Lahan Basah Buatan 463
Table 9.11 Ketergantungan suhu, parameter stoikiometri, parameter komposisi dan parameter
yang menggambarkan transfer oksigen dalam model biokinetik CW2D (Langergraber dan Šimůnek,
2005).
Langergraber (2007).
i→ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
j Proses Komponen SO SF SA SI SNH SNO SSO4 SH2S XS XI XH XA XFB XAMB XASRB XSOB
↓ expressed O2 COD COD COD N N S S COD COD COD COD COD COD COD COD
as →
3 Pertumbuhan 1 YH
−1/YH ν3,5 1
aerobik XH on SF
2.86 YH
4 Pertumbuhan
1 −1/YH ν4,5 1
aerobik XH on SA 1 YH
5 Pertumbuhan 1 YH
−1/YH ν5,5 1
aerobik XH on SA
2.86 YH
6 Lysis of XH fBM,SF ν6,5 νLysis fBM,XI −1
7
Pertumbuhan 4.57 YA 1 1/YA 1
iN,BM
aerobik XA on YA YA
SNH
8 Lysis of XA fBM,SF ν8,5 νLysis fBM,XI −1
9 Growth of XFB −1/YFB 1 YFB
ν9,5 1
YFB
10 Lysis of XFB fBM,SF ν10,5 νLysis fBM,XI −1
11 Pertumbuhan XAMB −1/YAMB ν11,5 1
12 Lysis of XAMB fBM,SF ν12,5 νLysis fBM,XI −1
13
Pertumbuhan XASRB −1/YASRB ν13,5 1 YASRB 1 YASRB 1
2 YASRB 2 YASRB
14 Lysis of XASRB fBM,SF ν14,5 νLysis fBM,XI −1
15
Aerobic growth 2 YSOB ν15,5 1/YSOB −1/YSOB 1
of XSOB on SH2S
YSOB
16
Pertumbuhan ν16,5 1 YSOB 1/YSOB −1/YSOB 1
anoksik dari 0.875 YSOB
XSOB on SH2S
17 Lysis of XSOB fBM,SF ν17,5 νLysis fBM,XI −1
Table 9.13 Koefisien stoikiometri untuk nitrogen amonia (lihat Tabel 9.17 untuk definisi
komposisi dan parameter stoikiometri).
Tabel 9.14 Laju reaksi di CWM1 bagian 1 (Langergraber et al., 2009b ; lihat Tabel 9.16 dan 9.17
untuk definisi koefisien laju).
Organisme heterotrofik
1) Hidrolisis
1k h∗
[ X S ( X H + X FB )
K X + ( X S / ( X H + X FB ) ) ]
∗( X H +n h∗X FB )
1 µH ∗
( SF
K SF + S F )(
∗
SF
SF+ S A
∗
SO
)(
K OH + S0
∗
S NH
)(
K NHH +S NH
∗¿
)
3) Pertumbuhan aerobic pada XH on SA (mineralisasi)
n g ¿ µ H∗
( SF
K SF + S F )(
∗
SF
SF+ SA
∗
K OH
)(
K OH + S 0
∗
)(
S NO
K NOH + S NO
∗
S NH
K NHH + S NH)( )
¿¿
4) Pertumbuhan anoksik XH on SF (denitrifikasi)
µH ∗
( SA
K SA + S A)(
∗
SA
SF+ SA
∗
SO
)(
K OH + S 0
∗
)(
S NH
K NHH + S NH
∗¿
)
5) Pertumbuhan anoksik XH pada SA (denitrifikasi
n g ¿ µ H∗
( SA
K SA +S A
∗
)( SA
SF+ S A
∗
K OH
)(
K OH +S 0
∗
S NO
)(
K NOH + S NO
∗
S NH
K NHH + S NH )( )
¿¿
6) Lisis of XH
b H∗X H
Bakteri autotrofik
7) Pertumbuhan aerobic pada XA on SNH (nitrifikasi)
μ A∗
( S NH
K NHA + S NH
∗
)(S NO
K OA +S O
∗¿
)
8) Lisis of XA
b A∗X A
Fermenting bacteria
9) Growth of XFB (fermentasi)
468 Pemodelan Transportasi
µFB∗
( SF
)
K S F+ SF
∗¿
(Countinued)
Modeling Multicomponent Reactive Transport in Constructed Wetlands
µ AMB∗
( SA
K SA MB+ S A
∗¿
)
12)
¿
( S NH
K NHAMB + S NH
∗X AMB
)
Lysis of XAMB
b AMB∗X AMB
Bakteri pereduksi sulfat asetotrofik
13) Growth of XASRB
µ ASRB∗
( SA
K S ASRB + S A )
∗¿
14)
¿
( S NH
K NHASRB +S NH
∗ X ASRB
)
Lysis of XASRB
b A SRB∗X ASRB
Bakteri pengoksidasi sulfida
15) Aerobic growth of XSOB on SH2S
μSOB∗
( SH 2S
K SSOB + S H 2 S
∗
SO
)(
K OSOB + S O
∗
S NH
K NHSOB +S NH )(
∗ X SOB
)
16) Anoxic growth of XSOB on SH2S
( S H 2 17)
)( )( )( )
Ly XS S K O SOB SN H
S sis O
μSOB ¿ n SOB∗ ∗ OB ∗ ∗ ∗X SOB
K SSOB + S H 2ofS SOSOB +S O K O SOB+ S 0 K N HSOB +S N H
Tabel 9.15 Parameter kinetik dalam model biokinetik CWM1 (Langergraber et al., 2009b).
bahan filter berpasir berbeda yang digunakan untuk lapisan utama. Selain pasir dengan
ukuran butir 0,06–4 mm, juga diuji pasir dengan ukuran butir 1–4 mm. Perbedaan tingkat
konsentrasi limbah NO3‐N untuk kedua bahan filter dapat disimulasikan. Konsentrasi
limbah NO3‐N untuk substrat berukuran 1‐4 mm lebih tinggi, karena bahan organik yang
kurang mudah terdegradasi dihasilkan melalui hidrolisis yang dapat dimanfaatkan untuk
denitrifikasi di lapisan yang lebih dalam (Langergraber, 2003).
Untuk CW skala percontohan VF dengan lapisan utama pasir 0,06–4 mm, distribusi
biomassa mikroba dan bakteri pada kedalaman filter yang berbeda juga diukur (Tietz
dkk., 2007). Mengonversi data terukur menjadi biomassa COD menghasilkan nilai rata-
rata 3400 hingga 5100 µg COD/g DW (berat kering substrat, yaitu pasir) untuk
sentimeter pertama lapisan utama, 1100 hingga 2600 µg COD/g DW dari 1– 5 cm, dan
640 hingga 1400 µg COD/g DW masing-masing dari 5–10 cm. Sebagian besar biomassa
dapat ditemukan di 10 cm bagian atas lapisan VF (Langergraber et al., 2007).
COD biomassa mikroba yang disimulasikan pada sentimeter pertama lapisan utama
adalah antara 5600 dan 3400 µg COD/g DW (kisaran nilai yang diukur) bila
menggunakan laju lisis heterotrofik antara 0,25 dan 0,35 hari−1. Gambar 9.3
membandingkan COD biomassa mikroba yang dihitung (dari respirasi yang diinduksi
substrat) dan simulasi pada kedalaman berbeda pada lapisan utama untuk laju lisis
heterotrofik sebesar 0,30 hari−1. Ketika membandingkan COD biomassa yang diukur
dan disimulasikan pada kedalaman yang berbeda pada lapisan utama, simulasi
tampaknya memperkirakan COD biomassa secara berlebihan pada kedalaman 1–5 cm
dan memprediksi COD biomassa pada kedalaman 5–10 cm secara berlebihan. Hal ini
dapat menjadi indikasi bahwa
Modeling Multicomponent Reactive Transport in Constructed Wetlands
Tabel 9.16 Ketergantungan suhu, parameter stoikiometri, parameter komposisi dan parameter yang
menggambarkan transfer oksigen dalam model biokinetik CW2D (Langergraber et al., 2009b).
INLET
Figure 9.1 Representasi skema CW skala pilot
dalam ruangan (nilai dalam cm).
6
0 Lapisan utama
Lapisan drainase
Tabel 9.17 Parameter hidrolik tanah model van Genuchten‐Mualem sebesar 0,06–4 mm lapisan utama
(parameter bentuk α, N dan L: van Genuchten, 1980).
kandung kadar air ukuran parameter kondivitas
an air keruh hidrolik
jenuh
sisa ffjr ffjs α N Ks
L
Parameter (–) (–) (cm−1) (–) (–) (cm/h)
HYDRUS parameter
Untuk ‘pasir’ 0.045 0.43 0.145 2.68 0.5 29.7
Ukuran Ks dan
porositas 0.045 0.30 0.145 2.68 0.5 117
set parameter yang pas 0.013 0.30 0.147 2.42 0.636 117
4
Data Terukur
3.5 HYDRUS parameter untuk ’Pasir’
KS dan porositas diukur
3 Fitted parameter set
Laju aliran limbah (L/h)
2.5
1.5
0.5
0
0 1 2 3 4 5 6
Waktu memuat (h)
Figure 9.2 Laju aliran efluen diukur dan disimulasikan untuk satu kali pemberian 10 liter ('Kumpulan
parameter HYDRUS atau Pasir', 'Ks dan porositas terukur' dan 'Kumpulan parameter pas' mewakili
kumpulan parameter dari Tabel 9.17).
Modeling Multicomponent Reactive Transport in Constructed Wetlands
Table 9.18 Median values of measured and simulated influent and effluent concentrations for NH4‐N
and NO3‐N in mg/l (adapted from Langergraber, 2003).
7'000
Simulated
5'000
4'000
3'000
2'000
1'000
0
0 5 10 15 20 25 30
Filter depth (cm)
Figure 9.3 Calculated and simulated microbial biomass COD in different depths of the main layer
(Source: adapted from Langergraber et al., 2007; Langergraber, 2015).
2
The organic loads for bed 1 to 3 were 20, 27 and 40 g COD/m /d, which correspond
to hydraulic loading rates of 32.2, 43.0, and 64.7 mm/d. The main layer of the filter con-
sisted of 50 cm sandy substrate (gravel size 0.06–4 mm, d10 = 0.2 mm; d60 = 0.8 mm).
The beds are planted with common reed (Phragmites australis). The data gained
from this experiment have been used to verify the temperature model incorporated
into CW2D (Langergraber, 2007).
The width of the transport domain in the numerical simulations was 4 m, and its
depth 0.8 m, while the transport domain itself was discretized into 31 columns and 33
rows. This resulted in a two‐dimensional finite element mesh consisting of 1023
nodes and 1920 triangular finite elements. An atmospheric boundary condition was
assigned to the top of the system representing the influent distribution system, and a
constant pressure head boundary condition (constant head of −4 cm) was assigned to
one side of the drainage layer (Langergraber, 2007).
Results for water flow comparing measured and simulated effluent flow rates and the
cumulative effluent of the experimental CW are shown in Langergraber (2007).
Simulated effluent flow rate and cumulative effluent flow match the measured data well.
476 Pemodelan Transportasi
70
NH4-N simulated
60 COD simulated
NH4-N measured
50
COD measured
COD, NH4-N (mg/l)
40
30
20
10
0
0 2 4 6 8 10 12 14 16 18 20
Temperature (°C)
Figure 9.4 Measured and simulated COD and NH4‐N effluent concentrations using the modified
parameter set (adapted from Langergraber, 2007).
Using the calibrated flow model, the effluent concentrations during summer could be
simulated using the standard CW2D parameter set (Langergraber and Šimůnek,
2005). However, measured COD and NH 4‐N effluent concentrations at low
temperatures could not be simulated because hydrolysis and nitrification at low
temperatures were overpredicted. The standard CW2D parameter set considers
temperature dependen- cies only for maximum growth, decay and hydrolysis rates.
Figure 9.4 shows measured and simulated COD and NH4‐N effluent concentrations
using a modified parameter set that includes temperature dependencies for the half‐
saturation constants of hydrol- ysis and nitrification. Using this modified parameter set,
it was possible to simulate the COD and NH4‐N effluent concentrations at low
temperatures (Langergraber, 2007).
parameters describing the hydraulic properties of the filter material using the inverse
simulation implemented in HYDRUS (Šimůnek et al., 2011).
Morvannou et al. (2012) showed that macropores/cracks in some layers of a VF
CW might serve as preferential flow paths through which water can bypass most of
the soil porous matrix in a largely unpredictable way. This is especially true for the
sludge layer in French‐type VF CWs (Molle et al., 2005; Troesch and Esser, 2012).
Water flow in such systems cannot be modeled with uniform flow models (such as
using the Richards equation combined with the van Genuchten‐Mualem function in
HYDRUS; Šimůnek et al., 2011). The comparison between measured and simulated
tracer breakthrough curves indicates that the nonequilibrium approach (i.e. using a
model that separately describes flow and transport in preferred flow paths and slow
or stagnant pore regions) seems to be more appropriate for simulating preferential
flow. Such a dual‐porosity model therefore also needs to be incorporated into the
software tools to describe water flow and solute transport accurately in French VF
CWs.
For HF CWs, in which the filter material is usually gravel and the saturated
hydraulic conductivity is much larger than the actual water flow in the filter, tracer
experiments are usually used to calibrate the water flow and single solute transport
models.
Table 9.19 Comparison of measured and calculated volumetric nitrification rates (adapted
from Morvannou et al., 2011).
Calculated from
Method Measured with solid respirometry simulation results*
150
NH4 inflow
Half NH4 inflow
No NH4 inflow
100
NH4out [mg/I]
50
0
0 20 40 60 80 100 120
t [d]
Figure 9.5 Influence of inflow NH4‐N concentrations on effluent NH4‐N concentrations (adapted from
Langergraber and Morvannou, 2014).
Rizzo et al. (2014) described the setup of a model to simulate experimental data
from a HF CW fed with artificial wastewater. During the experiments, the only
nitrogen parameter measured in influent was TKN (Total Kjehldal Nitrogen). As the
model requires influent concentrations of ammonia nitrogen, the organic N content of
differ- ent COD fractions had to be adapted from standard values for the type of used
artificial wastewater. This results in different ratios of ammonia nitrogen to TKN.
Figure 9.5 shows the impact of different influent concentrations of ammonia nitrogen
on simu- lated effluent concentrations. When all influent TKN is assumed to be
ammonia (i.e. no organic nitrogen is present), simulated effluent concentrations of
ammonia nitrogen are higher than measured effluent concentrations. In contrast,
when all influent TKN is assumed to be organic nitrogen (i.e. no ammonia is
present), simulated effluent concen- trations are too low.
Pucher and Langergraber (2015) simulated vertical flow filters treating domestic
wastewater using experimental data from Dal Santo et al. (2010). The main layer of
the filters consisted of sand with a grain size of 1–4 mm or zeolite with a grain
size of 2–5 mm. Both filters had an impounded drainage layer. After calibrating the flow
model for the sand and zeolite filters (Table 9.20), the procedure for calibrating the
biokinetic model was as follows: the standard parameter set of the CW2D biokinetic
model (Langergraber and Šimůnek, 2012) was used in the first simulation of the sand
filter (‘simulated sand 1’ in Table 9.21). This resulted in overprediction of nitrification,
that is, simulated ammonia nitrogen concentrations were much lower than measured
−1
data. After reducing the maximum nitrification rate from 0.9 to 0.175 d , the
simulated effluent concentrations of ammonia nitrogen matched measured data well
(‘simulated sand 2’ in Table 9.21). After adding the measured adsorption isotherms
according to Dal Santo et al. (2010) and a first‐order rate coefficient for
nonequilibrium adsorption
Modeling Multicomponent Reactive Transport in Constructed Wetlands
Table 9.20 Estimated soil hydraulic parameters of the van Genuchten‐Mualem model (shape
parameters α, N and L: van Genuchten, 1980).
Table 9.21 Measured and simulated influent and effluent concentrations (in mg/l, measured data: average values
and standard deviations in brackets, simulated data: average values over 1 day).
−1
of 0.01 d , a good match between simulated and measured data could be obtained for
the zeolite filter (‘simulated zeolite’ in Table 9.21).
Pálfy and Langergraber (2014) simulated experimental results from batch ‐fed col-
umn experiments using the CWM1 biokinetic model. They described the need to
adjust some parameters of the biokinetic model to be able to simulate anaerobic,
anoxic, and aerobic processes to occur in parallel. These phenomena occurred in
practice and can be explained by the local effect of root zone re ‐aeration. Figure 9.6
shows measured and simulated sulfate concentrations before and after the adjust-
ment of parameters of the biokinetic model. Batch experiments can be used to cali-
brate the biokinetic model parameters as there is no impact of water flow on the
treatment performance.
SO4-S
20
Measured concentrations
Model results
15
Concentration
[mg L–1]
10
0
0 5 10 15 20
time [days]
SO4-S
20
Measured concentrations
Model results
15
Concentration
[mg L–1]
10
0
0 5 10 15 20
time [days]
Figure 9.6 Measured and simulated sulfate concentrations for a batch‐fed column at 24 °C planted
with Carex (left: with the standard parameter set of the biokinetic model; right: after adjusting
inhibition and half‐saturation coefficients to allow anaerobic, anoxic, and aerobic processes to occur
in parallel). Adapted from Pálfy (2013).
particle sizes (<1, 1–3, 3–10, 10–30, and 30–100 µm, respectively). It was further
assumed that particles are transported in the liquid phase until they are attached.
Once attached the particles undergo hydrolysis and cCR (readily biodegradable
organic matter; Langergraber and Šimůnek, 2005) in the liquid phase is produced. An
addi- tional assumption was that particles with different sizes have the same density
and degradability.
The model was tested using measurements of particle concentrations in different
depths within the 50‐cm main layer of the VF pilot ‐scale CW, as described earlier.
Particle measurements at the inlet were used to calculate fractionation of COD
influent. Measured and simulated soluble particulate COD concentrations for the 10–
30 µm par- ticle class before and after loading of the VF filter along the filter depth
are shown in Figure 9.7. ‘Measured’ COD was derived from particle measurements.
In general, the
Modeling Multicomponent Reactive Transport in Constructed Wetlands
70
10–30 µm
60
50
Particulate COD (mg/l)
40
30 Measured
Simulated - before loading
20
Simulated - after loading
10
0
0 10 20 30 40 50
Filter depth (cm)
Figure 9.7 Comparison of measured and simulated particulate COD for the 10–30 µm particle class
(adapted from Langergraber and Šimůnek, 2009).
first test showed a good agreement between measured and simulated particulate
COD. However, there is still a great need for further testing of the particle transport
model and for finding an optimal parameter set for various conditions present in
constructed wetlands.
Several clogging models have been developed for BIO_PORE (Samsó and García,
2013). Samsó and García (2014) described the relationship between bacterial communi-
ties and accumulated solids, leading to clogging in a HF CW. The so‐called
‘Cartridge Theory’ describes how the active bacterial zone moves in the filter. In this
model, the effect of the inert organic matter accumulated in pores upon the hydraulic
conductivity is not considered. Description of limited growth of bacteria was
implemented by adding two parameters. The first parameter provides a negative
feedback term to the growth of all bacteria groups to prevent their unlimited growth
in areas where substrate concentrations are high (which is especially relevant for the
influent zones of HF CWs). The second parameter provides a negative feedback term
to the growth equa- tions, but in this case, it decreases the growth rate of bacteria due
to progressive accu- mulation of inert solids in the pore space of the granular media.
Samsó et al. (2015a) described in their simulation study the effect of bacteria density
and accumulated inert solids on effluent pollutant concentrations, but again only
without considering the effect of organic matter accumulation on water flow. Samsó
et al. (2016) additionally described the effects of a pore size reduction on the
hydraulic conductivity using the biological clogging model proposed by Mostafa and
Van Geel (2007). Using this approach, Samsó et al. (2016) were able to simulate
realistic behaviour of HF CWs, including overland flow and re‐infiltration of water
in the unused filter material.
References
Bradford, S.A., Šimůnek, J., Bettahar, M., et al. (2003). Modeling colloid attachment,
straining, and exclusion in saturated porous media. Environmental Science and
Technology 37 (10): 2242–2250.
Dal Santo, S., Canga, E., Pressl, A., et al. (2010). Investigation of nitrogen removal in a two‐
stage subsurface vertical flow constructed wetland system using natural zeolite. In:
Proceedings of the 12th IWA Specialized Group Conference on Wetland Systems for
Water Pollution Control (ed. F. Masi and J, Nivala), 263–270. 3–8 October 2010, San
Servolo, Venice, Italy.
Dittmer, U., Meyer, D., and Langergraber, G. (2005). Simulation of a subsurface vertical
flow constructed wetland for CSO treatment. Water Science and Technology 51 (9):
225–232.
García, J., Rousseau, D.P.L., Morató, J., et al. (2010). Contaminant removal processes in
subsurface‐flow constructed wetlands: a review. Critical Reviews in Environmental
Science and Technology 40 (7): 561–661.
Henrichs, M., Langergraber, G., and Uhl, M. (2007). Modelling of organic matter
degradation in constructed wetlands for treatment of combined sewer overflow.
Science of the Total Environment 380 (1–3): 196–209.
Henrichs, M., Welker, A., Uhl, M. (2009). Modelling of biofilters for ammonium reduction
in combined sewer overflow. Water Science and Technology 60 (3), 825–831.
Kadlec, R.H. and Wallace, S. (2009). Treatment Wetlands, 2nd ed. Boca Raton, FL:
CRC Press.
Karlsson, S.C., Langergraber, G., Pell, M., et al. (2015). Simulation and verification of
hydraulic properties and organic matter degradation in sand filters for greywater
treatment. Water Science and Technology 71 (3): 426–433.
Kumar, J.L.G. and Zhao, Y.Q. (2011). A review on numerous modeling approaches for
effective, economical and ecological treatment wetlands. Journal of Environmental
Management 92: 400–406.
Langergraber, G. (2001). Development of a simulation tool for subsurface flow constructed
wetlands. PhD dissertation, BOKU University, Vienna, Austria.
484 Pemodelan Transportasi
Meyer, D., Chazarenc, F., Claveau‐Mallet, D., et al. (2015). Modelling constructed wetlands:
Scopes and aims – A comparative review. Ecological Engineering 80: 205–213.
Molle, P., Liénard, A., Boutin, C., et al. (2005). How to treat raw sewage with constructed
wetlands: an overview of the French systems. Water Science and Technology 51 (9): 11–
21.
Morvannou, A., Choubert, J.‐M., Vanclooster, M., and Molle, P. (2011). Solid respirometry
to characterize nitrification kinetics: A better insight for modelling nitrogen
conversion in vertical flow constructed wetlands. Water Research 45 (16): 4995–
5004.
Morvannou, A., Forquet, N., Vanclooster, M., and Molle, P. (2012). Which hydraulic model
to use in vertical flow constructed wetlands? Proceedings of the 13th IWA Specialized
Group Conference on Wetland Systems for Water Pollution Control – Conference Papers
Volume 2, 145–153. 25–29 November 2012, Perth.
Morvannou, A., Froquet, N., Troesch, S., and Molle, P. (2015). Impact of design and
operating conditions on vertical flow constructed wetland performances: modelling
and global assessment confrontation. 6th International Symposium on Wetland
Pollution Dynamics and Annual Conference of the Constructed Wetland Association –
Book of Abstracts, 196–197. 13–18 September 2015, York, UK.
Mostafa, M. and Van Geel, P.J. (2007). Conceptual models and simulations for biological
clogging in unsaturated soils. Vadose Zone Journal 6: 175–185.
Pálfy, T. (2013). Verification of the implementation of CWM1 in the HYDRUS
wetland module. MSc thesis, BOKU University, Vienna, Austria.
Pálfy, T.G. and Langergraber, G. (2014). The verification of the Constructed Wetland
Model No. 1 implementation in HYDRUS using column experiment data. Ecological
Engineering 68: 105–115.
Pálfy, T.G., Gribovszki, Z., and Langergraber, G. (2015a). Design ‐support and performance
estimation using HYDRUS/CW2D: A horizontal flow constructed wetland for polishing
SBR effluent. Water Science and Technology 71 (7): 965–970.
Pálfy, T.G., Meyer, D., and Molle, P. (2015b). Orage: Simulation of planted detentive filters
treating urban storm water. 6th International Symposium on Wetland Pollution
Dynamics and Annual Conference of the Constructed Wetland Association – Book of
Abstracts, 216–217. 13–18 September 2015, York, UK.
Pucher, B. and Langergraber, G. (2015). Simulation of two‐stage vertical flow sand and
zeolite filters treating domestic wastewater using the HYDRUS wetland module. 6th
International Symposium on Wetland Pollution Dynamics and Control and Annual
Conference of the Constructed Wetland Association – Book of Abstracts, 198–199. 13–18
September 2015, York, UK.
Rizzo, A., Langergraber, G., Galváo, A., et al. (2014). Modelling the response of horizontal
flow constructed wetlands to unsteady organic loads with HYDRUS ‐CWM1. Ecological
Engineering 68: 209–213.
Samsó, R. and Garcia, J. (2013). BIO PORE, a mathematical model to simulate biofilm
growth and water quality improvement in porous media: Application and calibration for
constructed wetlands. Ecological Engineering 54: 116–127.
Samsó, R. and García, J. (2014). The Cartridge Theory: A description of the functioning of
horizontal subsurface flow constructed wetlands for wastewater treatment, based on
modelling results. Science of The Total Environment 473–474: 651–658.
Samsó, R., Blázquez, J., Agulló, N., et al. (2015a). Effect of bacteria density and accumulated
inert solids on the effluent pollutant concentrations predicted by the constructed
wetlands model BIO_PORE. Ecological Engineering 80: 172–180.
486 Pemodelan Transportasi
Samsó, R., Meyer, D., and García, J. (2015b). Subsurface flow constructed wetland models:
review and prospects. In: The Role of Natural and Constructed Wetlands in Nutrient
Cycling and Retention on the Landscape (ed. J. Vymazal), 149–174. Springer
International Publishing.
Samsó, R., García, J., Molle, P., and Forquet, N. (2016). Modelling bioclogging in variably
saturated porous media and the interactions between surface/subsurface flows:
application to Constructed Wetlands. Journal of Environmental Management 165:
271–279.
Šimůnek, J., van Genuchten, M.Th., and Šejna, M. (2008). Development and applications of
the HYDRUS and STANMOD software packages and related codes. Vadose Zone
Journal 7 (2): 587–600.
Šimůnek J., Šejna, M., and van Genuchten, M.Th. (2011). The HYDRUS Software Package
for Simulating the Two‐ and Three‐Dimensional Movement of Water, Heat, and
Multiple Solutes in Variably‐Saturated Media. Technical Manual, Version 2.0. PC‐
Progress, Prague.
Šimůnek, J., Jacques, D., Langergraber, G., et al. (2013). Numerical modeling of
contaminant transport using HYDRUS and its specialized modules. Journal of the
Indian Institute of Science 93 (2): 265–284.
Smethurst, P.J, Petrone, K., Langergraber, G., et al. (2014). Nitrate dynamics in a rural
headwater catchment: measurements and modeling. Hydrological Processes 28 (4):
1820–1834.
Tietz, A., Langergraber, G., Sleytr, K., et al. (2007). Characterization of microbial
biocoenosis in vertical subsurface flow constructed wetlands. Science of the Total
Environment 380 (1–3): 163–172.
Toscano, A., Langergraber, G., Consoli, S., and Cirelli, G.L. (2009). Modelling pollutant
removal in a pilot‐scale two‐stage subsurface flow constructed wetlands. Ecological
Engineering 35 (2): 281–289.
Troesch, S. and Esser, D. (2012). Constructed wetlands for the treatment of raw
wastewater: The French experience. Sustainable Sanitation Practice 12 (July 2012):
9–15. Available at www.ecosan.at/ssp
van Genuchten, M.Th. (1980). A closed form equation for predicting the hydraulic
conductivity of unsaturated soils. Soil Science Society of America Journal 44: 892–898.