Chemosphere 66 (2007) 916923

www.elsevier.com/locate/chemosphere

Recycled vertical ow constructed wetland (RVFCW)a novel

method of recycling greywater for irrigation in small communities
and households
A. Gross
a

a,*

, O. Shmueli a, Z. Ronen a, E. Raveh

Department of Environmental Hydrology and Microbiology, Zuckerberg Institute for Water Research, Jacob Blaustein Institutes for Desert Research,
Ben-Gurion University of the Negev, Sede Boqer Campus, 84990 Midreshet Ben-Gurion, Israel
b
Department of Plant Science, Gilat Research Center, Mobile Post Negev 85280, Israel
Received 19 February 2006; received in revised form 4 June 2006; accepted 6 June 2006
Available online 17 July 2006

Abstract
The use of greywater for irrigation is becoming increasingly common. However, raw greywater is often contaminated and can cause
environmental harm and pose health risks. Nevertheless, it is often used without any signicant pretreatment, a practice mistakenly considered safe. The aim of this study was to develop an economically sound, low-tech and easily maintainable treatment system that would
allow safe and sustainable use of greywater for landscape irrigation in small communities and households. The system is based on a combination of vertical ow constructed wetland with water recycling and trickling lter, and is termed recycled vertical ow constructed
wetland (RVFCW). The RVFCWs properties, removal eciency, hydraulic parameters and feasibility were studied, as well as the environmental eects of the treated greywater, as reected by soil and plant parameters over time. The RVFCW was ecient at removing
virtually all of the suspended solids and biological oxygen demand, and about 80% of the chemical oxygen demand after 8 h. Fecal coliforms dropped by three to four orders of magnitude from their initial concentration after 8 h, but this was not always enough to meet
current regulations for unlimited irrigation. The treated greywater had no signicant negative impact on plants or soil during the study
period. The feasibility analysis indicated a return over investment after approximately three years. We concluded that the RVFCW is a
sustainable and promising treatment system for greywater use that can be run and maintained by unskilled operators.
 2006 Elsevier Ltd. All rights reserved.
Keywords: Vertical ow constructed wetland; Wastewater reuse; Greywater treatment; Environmental pollution

1. Introduction
The quantity of freshwater available worldwide is
declining, raising the pressing need for its more ecient
use. One method of conserving water is by recycling greywater (GW) for irrigation. GW is domestic wastewater that
includes only wash water (i.e., bath, dish, and laundry
water), whereas blackwater consists of toilet water. Due
to the substantial dierence in their qualities, separating
GW and blackwater would provide for more eective
wastewater treatment, allowing a large volume of water
*

Corresponding author. Tel.: +972 8 6596896; fax: +972 8 6596909.

E-mail address: amgross@bgumail.bgu.ac.il (A. Gross).

0045-6535/$ - see front matter  2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chemosphere.2006.06.006

to be eciently recycled (Lindstrom, 2000). This is particularly important in arid zones, were water is scarce and
recycling GW for private and public landscape irrigation
could reduce potable water use by up to 50% (DHWA,
2002). The use of GW for private garden irrigation is
becoming increasingly common. In most countries, regulations or specic guidelines for GW reuse are not available,
and it is therefore often used without any signicant pretreatment, a practice mistakenly considered safe. In countries such as the USA and Australia, where regulations
for the use of GW have been established, they concentrate
on issues associated with public health but do not consider
potential harmful environmental impacts (Dixon et al.,
1999; DHWA, 2002; ADEQ, 2003). The separation of the

A. Gross et al. / Chemosphere 66 (2007) 916923

917

toilet stream from domestic wastewater generates euents

which have reduced levels of nitrogen, solids, and organic
matter (especially the barely degradable fraction), but often
contain elevated levels of surfactants, oils, boron and salt.
The components in GW may alter soil properties, damage
plants and contaminate groundwater (Garland et al., 2000;
Gross et al., 2005; Wiel-Shafran et al., 2006). A study
aimed at applying commercial systems to GW reuse in
households demonstrated, in ve dierent commercial systems, failure to treat the GW suciently for unlimited use
(Gross et al., 2003). That study also suggested that this was
so, either because the treatment was too supercial (as the
water is considered safe by many), or because it was a
downscaled wastewater treatment system rather than a
GW treatment system aimed at treating small volumes in
private houses.
The aim of the current study was to develop an
economically sound, low-tech and easy-maintenance
treatment system that would allow safe and sustainable
use of GW for landscape irrigation in small communities
and households.
2. Materials and methods
2.1. Recycled vertical ow constructed wetland
The proposed treatment method is a modication of the
vertical ow constructed wetland (VFCW) described
by IWA (2000), with a novel set-up. Initially, the system
was composed of two containers (0.95 m W 0.95 ml
0.55 mH; about 500 l each) placed one above the other:
(1) the upper one was a VFCW composed of a three-layer
bed consisting of 15 cm planted organic soil over a 30 cm
layer of tu or plastic media and a 5-cm lower layer of
limestone pebbles. The bottom of the beds compartment
was perforated; (2) the lower container was used as a water
reservoir, located directly beneath the VFCW (Fig. 1).
The raw GW owed through a sedimentation tank
which accounted for about 10% of the total system volume
and where only coarse material settled. From this tank it
was pumped or overowed into the root zone of the
VFCW plants, and then it trickled down through the
three-layer lter bed to the reservoir. A centrifuge pump
continuously recycled the GW at a known rate from the
reservoir back to the VFCW. The treated water was then
used for irrigation directly or following a secondary sedimentation. An overow pipe was set from the upper wetland container to the reservoir to prevent overow in
case of the wetland clogging.
2.2. Recycled vertical ow constructed wetland performance
The Recycled vertical ow constructed wetland
(RVFCW) properties, removal eciency and hydraulic
parameters were studied in a short-term study, a 3-month
batch greenhouse study and a long-term case study. All
the studies were conducted after three months of a continu-

Fig. 1. Recycled vertical ow constructed wetland (RVFCW). (A) vertical

ow constructed wetland, (B) reservoir, (C) recycling pump, and (D)
demonstration of lter media layers (peat, tu or high surface area plastic
media, and lime pebbles in top middle and bottom layers, respectively).

ous working period. This procedure ensured the development of bio-lm in the wetland and stabilization of the
system performance in terms of removal eciency and ow.
2.2.1. Short-term study
In the beginning of the study, the pore volume of the lter section and the treated water reservoir were emptied
and 300 l of fresh GW was introduced into the VFCW
(at the root zone). A subsample of the raw GW was collected for analysis (time zero). The GW was continuously
recycled between the reservoir and the VFCW at a known
rate of 390 l h1, determined by a water meter that was
attached to the system. Samples of the treated GW were
taken immediately after it initially passed the bed and then
after 2, 4, 8, 12, 24, and 48 h. Samples were analyzed for:
total suspended solids (TSS), total phosphorus (TP), total
nitrogen (TN), anionic surfactants measured as methylene
blue-active substances (MBAS), dissolved oxygen (DO),
electrical conductivity (EC), pH, 5-day biological oxygen
demand (BOD), chemical oxygen demand (COD), and
total boron (TB). Water-quality analysis followed standard
procedures (APHA, 1998); fecal coliforms (FC) and total
coliforms were measured by the plate count technique
(Merck, 2000).
2.2.2. Batch greenhouse study
The RVFCW was used to evaluate the environmental
eects of treated GW on plants and soils in comparison

918

A. Gross et al. / Chemosphere 66 (2007) 916923

with untreated GW and freshwater. GW was prepared articially by mixing 20 g of laundry detergent (Ariel Ultra,
Procter and Gambel, USA), 0.86 g of boric acid, and
400 ml of raw kitchen euents into 150 l of tap water to
resemble the GW quality at a nearby farm (Gross et al.,
2005). Every other day at 08:00, 150 l of treated GW was
removed from the reservoir and replaced with the articial
GW that was introduced into the root lter zone. Thirty
lettuce plants (Lactuca sativa), which are sensitive to water
quality, were used as the model plants for each treatment.
Plants were grown for 89 d (15 Feb 0215 May 02) in 5 l
pots lled with autoclaved quartz sand, and were drip-irrigated (300 ml plant1 d1) twice a day (at 07:00 and
17:00) by computerized irrigation system (Netam, Kibutz
Hazerim, Israel). The investigation was concluded once the
lettuce started to bud. Water samples of untreated GW,
treated GW and freshwater were collected three times a
month, 810 h after GW replacement, and analyzed for:
TSS, TP, TN, total ammonia nitrogen (TAN), nitrite
(NO2N), nitrate (NO3N), EC, pH, anionic surfactants
as MBAS, BOD, COD, TB, total coliforms, and FC. Samples were also analyzed for minerals and metals (Ca, K,
Na, Mg, B, Fe, Al, Cu, Mn, Zn) by inductively coupled
plasma (ICPAES) and atomic adsorption (AA) once a
month. Analysis followed standard procedures (APHA,
1998; Merck, 2000). The water-quality dierences among
the treatments were statistically analyzed by ANOVA
(p < 0.05).
Undisturbed 5 cm soil cores were taken monthly from
seven pots of each treatment and analyzed for: pH, organic
carbon (OC), total Kjeldahl nitrogen (TKN), minerals and
metals (Ca, K, Na, Mg, B Fe, Al, Cu, Mn, Zn). Analysis
followed standard procedures (Soil and Plants Analysis
Council, 1999). For the FC count, ve undisturbed cores
(6 g wet weight from depths of 5 cm) from each treatment
were put into sterile tubes. Pyrophosphate buer (6 ml) was
added and samples were shaken for 1 h. The supernatant
was used for FC count on Chromocult TBX (Tryptone Bile
X-glucuronide) agar plates by the pour plate method
(Merck, 2000). Soil sub samples were analyzed for moisture
content and the FC count was calculated on dry weight
base.
The plants from these pots were tested for wet and dry
weight, leaf weight and surface area, number of leaves,
and minerals and metals as already described.
2.2.3. Long-term case study
Based on the results, sizing and operational procedures
were established. A RVFCW was installed in a private garden of a ve-member household in Midreshet Ben Gurion
(30 510 0400 N and 34 460 5700 E; the Negev desert, Israel) to
water a 150 m2 garden. The familys expected GW ow
was about 450 l d1. The system reused GW originating
from laundry, showers and sinks. The GW owed through
a 40 l sedimentation tank to the root zone in the VFCW
unit and trickled into a 500 l reservoir. The surface area
of the VFCW was determined to be 1 m2 following the

model of Kadlec and Knight (1996) assuming a conservative average retention time of 8 h. The bed depth was
0.5 m and the recycling rate was 390 l h1 or, in other
words, the water passed through the RVFCW bed between
7 and 21 times (824 h retention time per 450 l daily ow at
a recycling rate of 390 l h1). The treated water overowed
to a secondary 40 l settling tank with submerged pump that
was controlled by a oat. Whenever this tank was full,
water was pumped for irrigation.
After establishment for 3 months, samples from the sedimentation tank and at the outlet were taken periodically
and analyzed for the aforedescribed parameters for over
12 months.
Data on the study areas climatic conditions were supplied by the Israeli Meteorological Service. The mean daily
maximum and minimum temperatures SD were 29.2
4 C and 16.5 3 C, respectively; the mean daily maximum and minimum relative humidities were 93 6% and
39 11%, respectively. Daily maximum global radiation
varied from 600 to 1054 (Vat m2), and direct radiation
from 370 to 1012 (Vat m2).
2.3. Feasibility analysis
A feasibility analysis of the RVFCW was conducted for
the long-term case study. Values for capital costs, maintenance, operation and energy were based on real costs.
The average irrigation costs (assuming irrigation with
freshwater) or in other words, the potential savings, was
estimated according to average water use and local water
costs, and other indirect costs such as infrastructure taxes
that are based on water consumption. Based on these costs
and savings, the return over investment (ROI) was calculated. The analysis was conducted with the assistance of
Ad-Wiser economic consulting oce (Ramat Gan, Israel).
3. Results and discussion
3.1. RVFCW design criteria
Typically, the design parameters for the VFCW are:
required area, depth, media and retention times, and the
main one, BOD of the euent (Vymazal et al., 1998). Interestingly, although there are a few models for sizing VFCW,
experience-based rules of thumb are often applied for
wetland design (Vymazal et al., 1998). The following model
(Eq. (1)) suggested by Kadlec and Knight (1996) was used
to determine the area of the RVFCW, taking into consideration that the euent passes the wetland several times:
A

Q
C i  C 
xLn
k
C e  C 

where A denotes the required wetland area (m2), Q denotes

the water ow rate (m3 d1), Ce denotes the outlet target concentration (mg l1), Ci denotes the inlet concentration
(mg l1), C* denotes the background concentration (mg l1),
and k denotes the rst-order areal rate constant (m d1).

A. Gross et al. / Chemosphere 66 (2007) 916923

The k value for BOD was previously estimated for lowquality GW as 0.16 m d1 (data not shown). The target
BOD concentration (Ce) was chosen as 10 mg l1 according
to the Israeli guidelines for euent quality for use in cities
(Halperin and Aloni, 2003). Based on that, for a water ow
of 500 l d1 (family of ve) and a recirculation rate of
about 400 l h1, the area was calculated to be approximately 1 m2. The depth of the suggested RVFCW was
0.5 m, within the range typically recommended (0.5
0.8 m) for VFCW (Cooper et al., 1996; Vymazal et al.,
1998).
The contaminated GW is introduced into the root
zone in the VFCW unit, which prevents bad odors and
mosquitoes, and reduces the possibility of human contact
and spread of disease. From the root zone the water falls
through the porous media to the collection reservoir,
enhancing aeration (similar to trickling lter principles),
which further prevents the development of odors and
enhances organic matter degradation and nitrication.
The lime pebbles buer the euents natural acidity, as well
as that produced by nitrication and biodegradation.
Recycling the water from the reservoir back to the upper
lter serves to dilute new raw GW and reduces the risk
of organic overload or other damage to the lter, such as
excess chlorination. Moreover, since the water passes
through the lter more than once, the area required to
attain a specic water quality is reduced. In small commu-

919

nities and households, the volume of wastewater may

change considerably over time, from virtually none to several m3 d1. The proposed RVFCW is exible as it recirculates the water and keeps the wetland and bed constantly
wet and operating, and as it is modular, several units can
be adjacently attached. The recycling rate is determined
by the required water quality, the bed dimensions, and
the wastewater ow rate. In practice, the retention time
in the bed was dicult to measure due to the recycling.
Therefore, here, retention time refers to the average time
the water recycles in the system, which is a function of
the target water quality and the size of the reservoir located
underneath the system. This parameter was determined
experimentally as described in the short-term study below.
In most VFCW, graded gravel, usually with a top layer of
washed sharp sand, is used as the bed medium. In the
suggested RVFCW, dierently from conventional VFCW,
the medium was designed to maintain relatively fast ow
and high surface area, enhancing aeration and nitrication
in the VFCW.
3.2. RVFCW performance
3.2.1. Short-term study
The initial GW quality was within the range presented
for GW in Table 1. The RVFCW eciently removed virtually all of the suspended solids, BOD and about 80% of the

Table 1
Average concentrations standard errors (SE) and the ranges of the greywater (GW), treated GW, and freshwater used to irrigate lettuce plants
Analysis

GW

Treated GW

TSS

158 30
85285
466 66
280688
839 47
702984
22.8 1.8
17.227
34.3 2.6
25.045.2
0.3 0.1
0.10.5
0.3 0.2
0.01.0
3.0 1.3
0.05.8
1.2 0.1
1.01.3
6.37.0
7.9 1.7
4.715.6
1.6 0.1
1.41.7
5 107 2 107
9 1041 108

31
06
0.7 0.3
01.5
157 62
60220
6.6 1.1
3.510.2
10.8 3.4
1.421.0
0.9 0.7
0.04.5
0.2 0.2
0.00.9
8.6 4.3
0.023.5
1.3 0.1
1.11.4
7.08.0
0.6 0.1
0.41.3
0.6 0.1
0.40.8
2 105 1 105
3 1027 105

BOD
COD
TP
TN
Total ammonia
nitrogen (TAN)
NO2N
NO3N
EC (dS m1)
PH
Anionic
Surfactants
Boron
FC (CFU 100 ml1)

Mean % removal
98
100
81
71
69

8

92

Freshwater

Standards

10a

0.5 0.1
0.40.6

10a

0.08 0.00
0.020.13
5.7 1.5
47.3
0.1 0.1
00.2
<0.05

8b
2b

4.9 0.7
4.15.7
1.2 0.1
1.21.3
7.18.8

65

0.30 0.03

0.5b

99

<1

< 1a

The GW was treated with recycled vertical ow constructed wetland (RVFCW) and the mean percent removal demonstrates its performance. Samples
were taken twice a month between march and may 2002 (n = 6), and results are in mg l1 unless otherwise.
a
Standard applies to the Israeli guidelines for euent quality for use in cities (Halperin and Aloni, 2003).
b
Standards from the Israeli Water Law (1975) for unlimited use (except drinking).

A. Gross et al. / Chemosphere 66 (2007) 916923

100

100

75

75

50

25

TSS
COD
BOD

% Removal

% Removal

920

50

25
Total phosphorus
Total nitrogen

0
0

25

50

75

100

25

50

100

108

100

107

Fecal coliforms
Total Coliforms

106

CFU 100 ml-1

75

% Removal

75

50

25
Boron
Anionic surfactants

105
104
103
102
101
100

25

50

75

100

Time (hours)

25

50

75

100

Time (hours)

Fig. 2. Percent removal of: total suspended solids (TSS), 5-day biological oxygen demand (BOD), chemical oxygen demand (COD), total phosphorus,
total nitrogen, boron, anionic surfactant, fecal coliforms and total coliforms over time by a RVFCW. The greywater was recycled through the lter
continuously at a rate of 390 l h1. Initial values (time zero) as in GW Table 1.

COD after 8 h (Fig. 2). TN and TP were also signicantly

reduced after 8 h by 75% and 60%, respectively. FC
dropped by three to four orders of magnitude from their
initial concentration after 8 h, but were still higher than
the Israeli standard for unlimited irrigation (less than 1
CFU 100 ml1; Halperin and Aloni, 2003). Over 90% of
the anionic surfactants were removed after 8 h and about
50% of the TB. After 12 h, most of the parameters reached
a steady state, with 9099% of the TSS, BOD, TN and
anionic surfactants, 7075% of the TP, and 50% of the
TB having been removed. FC and total coliform populations continued to decrease, and after about 88 h, their
counts were about 10 and 100 CFU 100 ml1, respectively.
Signicant accumulation of TB was found on the peat layer
of the VFCW unit but not in the other bed layers, or in the
plants (data not shown). The plants in this study were
dwarf papyrus (Cyperus haspan) and pennywort (Hydrocotyle leucocephala). It might be possible to increase TB
removal by introducing plants that are known to accumulate boron, such as reeds (Phragmites australis), duckweeds
(Lemna spp.) and mares tail (Hippuris bulgaris) (Raskyn
et al., 1997; Meaguer, 2000; Del-Campo and Oron, 2002),
or by increasing the peat volume. The EC and pH values
were similar to their initial values and within acceptable
ranges for irrigation (data not shown). These results suggest that 812 h of GW recycling was sucient to produce

high-quality water for landscape irrigation. If surface irrigation is used, a disinfection method should be considered.
3.2.2. Batch greenhouse study
The experimental set-up was designed to test the performance of the RVFCW over a longer period of time and to
emphasize the dierent eects of treated GW as compared
to freshwater and untreated GW on the soil and plants. In
addition, the experiment was used to compare the fertilization properties of the dierent sources. The average
water quality and ranges of the dierent sources is summarized in Table 1. The average TSS, BOD, and FC were
158 mg l1, 466 mg l1 and 5 107 CFU 100 ml1, respectivelymuch higher than current standards for very high
quality treated wastewater that can be used for irrigation
in cities (Halperin and Aloni, 2003). Two parameters were
used to evaluate the RVFCW treatment system: (a) its ability to remove variables of environmental and health concern, and (b) whether the output water met current
Israeli standards as already described (Table 1). Concentrations of minerals and metals were within acceptable limits
for reuse and were not signicantly dierent (p < 0.05)
between sources (data not shown). The treated GW met
the current standard except with regard to the FC count,
which decreased by an average two orders of magnitude
from 107 to around 105 CFU 100 ml1. Results were simi-

A. Gross et al. / Chemosphere 66 (2007) 916923

related to their concentrations in the irrigation waters

(Table 1).
3.2.3. Long-term case study
A RVFCW was operated in a private household for over
a year and its performance was monitored. Overall, the
average performance of the RVFCW in terms of removal
eciencies was similar to the greenhouse study, as
described in Table 1 and already discussed. The Israeli
guidelines for euent quality for use in cities (Halperin
and Aloni, 2003) dene the major water quality criteria
to be 10 mg l1 for both TSS and BOD, and less than 1
CFU 100 ml1 for FC. Results of the current study demonstrated that the TSS and BOD values meet these standards
criteria, except on a few occasions where a slight deviation
from these criteria was observed (Fig. 3). Similar to the

-1

TSS (mg l )

lar in pattern to the short-term study (Fig. 2). These results

suggest that if the FC concentration in the raw GW is high,
the water must be disinfected prior to surface irrigation.
Other means of using the water for subsurface irrigation
might also be considered in order to avoid disinfection.
As expected, no FC were found in the freshwater
treatment.
In the soils that were irrigated with treated GW, an average 2 CFU FC g1 soil were found, and about 104 CFU
g1 soil in soils that were irrigated with raw GW (Table
2). The reduction in FC during their transit through the
unsaturated soil prole is due to three primary processes:
(a) adsorption to soil particles, (b) ltering of aggregate
lumps, and (c) inactivation (die-o) due to chemical reactions and microbial antagonism within the soil (Feachem
et al., 1983). In most soils at 2030 C, the reduction is several-fold within the rst few days and does not usually continue beyond 20 days (Feachem et al., 1983; Spackman
et al., 2003). OC, TN and EC accumulation in the soils
was correlated to their concentrations in the irrigation
water (Tables 1 and 2).
Limited nutrient sources (phosphorus and nitrogen) in
the freshwater and treated GW as compared to the raw
GW (Table 1) resulted in retarded growth in these treatments, as reected by the leaves surface area, weight and
number per lettuce head (Table 3). Nevertheless, in the
GW-irrigated plants, brown patches (chlorosis) developed
on the leaf tips. This was caused by the elevated salinity
and TB levels in the plants leaves (Table 3), which was cor-

921

450
375
300
225
150
75

Guidelines for
unlimitted irrigation

Inlet

20

Outlet

10
0

Inlet

300

1

OC (%
dry soil)

TN (mg kg
dry soil)

pH

FC (CFU g
dry soil)

GW
Treated GW
Freshwater

0.29 0.02
0.26 0.01
0.22 0.01

385 35
176 18
70 7

8.2
8.5
8.6

3.1 104 6.0 104

4.0 103 2.0 103
<1

-1

100
20

Guidelines for
unlimitted irrigation
Outlet

10
0

106

-1

Source

BOD5 (mg l )

1

200

Fecal coliforms (CFU 100 ml )

Table 2
Average ( standard error) concentrations of organic carbon (OC), total
nitrogen (TN), pH, and fecal coliform count (FC) in sandy soil following 3
months irrigation with freshwater, greywater (GW) and treated GW.
Source water quality as in Table 1

Table 3
Average growth parameters per lettuce head following the dierent
treatments (weight, number of leaves, and surface area) and average
concentration of boron (B) and chloride (Cl) in the lettuce leaves from
the dierent treatments, after 3 months of irrigation (n = 6)
Irrigation
source

Greywater
Treated GW
Freshwater
Initial

Leaves per lettuce head

Weight
(g)
a

13.3
3.5b
2.2c
0.2c

Number

Surface
area (cm2)

35.7
17.5b
12.4c
6c

1200
260b
166c
55c

Cl in
leaves
(mg kg1
dry leaf)
127
84b
65b
39c

B in leaves
(mg kg1
dry leaf)

105
104

0.5
0.2b
0.2b
0.2b

a,b,c
Values in a column followed by dierent letters dier signicantly at
p < 0.05.

Guidelines for
unlimitted irrigation

103
102
101

Outlet

100
Mar-03

Inlet

Jun-03

Sep-03

Dec-03

Mar-04

Date
Fig. 3. The total suspended solids (TSS), 5-day biochemical oxygen
demand (BOD), and fecal coliforms concentration over time in raw and
treated greywater (GW). Greywater was treated with recycled vertical ow
constructed wetland (RVFCW). The solid line represent the allowed
standard concentration for irrigation in cities (Halperin and Aloni, 2003).

922

A. Gross et al. / Chemosphere 66 (2007) 916923

other experiments, a signicant reduction in FC, of up to

four orders of magnitude, was demonstrated. Nevertheless,
this was not always enough to meet the Israeli standard for
wastewater reuse in cities. Simple disinfection techniques
(i.e., chlorination or UV radiation), subsurface irrigation
or use of other barriers, such as night irrigation, could be
used to mitigate this problem. Further improvement in
the overall performance of the system may be achieved
by ensuring a minimal retention time of the wastewater
in the system. This can be established by reducing the possibility of short circuits, where water that has only been
slightly treated is mixed in the reservoir and pumped for
irrigation. The simple management practice of emptying
the reservoir underneath the bed early in the morning
(before the users begin their activity) would virtually eliminate this possibility. If the volume of the reservoir is larger
than the volume of wastewater produced daily, and assuming that by 22:30 most of the water-related activity has
ended and will not resume before 6:30 the next morning,
a minimal 8 h retention time would exist.

practice mistakenly considered safe. The suggested

RVFCW is a promising, low-cost, low-tech treatment system that can be run and maintained by unskilled operators.
It can treat GW suciently to meet current standards for
unlimited irrigation, except for the complete removal of
FC, which can be achieved by the introduction of a small
disinfection unit. Subsurface irrigation can also be used
to mitigate this problem. In addition to developing the
appropriate technology, we believe that proper standards,
their enforcement and education are necessary to resolve
the potential risks currently associated with GW reuse.
Acknowledgements
This research was funded by the Israeli Water Commission and the Texas Department of Agriculture in the USA.
We gratefully acknowledge Mr. Ramy Mously for his assistance in constructing the RVFCW and Ms. Natasha Bondarenko for her assistance with the chemical analysis.
References

3.3. Feasibility analysis

Water scarcity and increasing environmental awareness
have served to enhance water reuse. Nevertheless, the costs
involved in such reuse, especially on a small scale, as well as
safety and maintenance issues, prevent water reuse from
being widespread. The cost of irrigating a 150 m2 garden
in the Negev desert is approximately US $30 month1
(based on a water cost of US $1.1 m3 and an average
use of 5 l d1 m2 (data not shown) and other sewage taxes
of about US $5 month1). The average consumption of
irrigation water for a 150 m2 garden would be 23 m3
month1 (0.005 m3 d1 m2 150 m2 per 30 d month1).
The cost of the RVFCW was about US $600 and the
annual maintenance and labor costs were about US $100.
In the current study, 210 m3 GW year1 were reused for
irrigation, accounting for 45% of the familys annual consumption, and covering about 83% of the irrigation water,
suggesting a gross savings of US $25 month1 or US $300
year1. Taking into consideration annual maintenance fees
of US $100, the net annual savings amounts to US $200.
Under this set of assumptions and conditions, the ROI
would be approximately 3 years, which is considered
attractive. Note that the analysis did not consider the cost
of GW separation as this would vary greatly (i.e., in new
vs. old houses, for example).
4. Conclusions
The use of raw GW for irrigation is becoming increasingly common and could potentially cause environmental
damage and pose public-health risks. In most countries,
regulations or specic guidelines for GW reuse are not
available or are insucient, as they consider health risks
while underestimating environmental ones. GW is therefore often used without any signicant pretreatment, a

ADEQ (Arizona Department of Environmental Quality), 2003. Using

Greywater at Home. Publication No. C 0106. Arizona Department of
Environmental Quality, Phoenix, AZ.
APHA, 1998. Standard Methods for the Examination of Water and
Wastewater. American Public Health Association, Washington DC,
USA.
Cooper, P.F., Job, G.D., Green, M.B., Shutes, R.B.E., 1996. Reed Beds
and Constructed Wetlands for Wastewater Treatment. WRC Publications, Medmenham, Marlow, UK.
Del-Campo, C.M., Oron, G., 2002. Boron removal from water and
accumulation by duckweed, Lemna gibba. Proceedings of 8th International Conference on Wetland Systems for Water Pollution Control.
Arusha, Tanzania, pp. 627633.
DHWA (Department of Health Western Australia), 2002. Draft Guidelines for the Reuse of Greywater in Western Australia. Department of
Health, Perth, Australia, p. 37.
Dixon, A.M., Butler, D., Fewkes, A., 1999. Guidelines for graywater reuse: health issues. J. Chartered Inst. Water Environ. Manag. 13, 322
326.
Feachem, R.G., Bradley, D.J., Garelick, H., Mara, D.D., 1983. Sanitation
and Disease: Health Aspects of Excreta and Wastewater Management.
John Wiley & Sons, Chichester, England.
Garland, J.L., Levine, L.H., Yorio, N.C., Adams, J.L., Cook, K.L., 2000.
Graywater processing in recirculating hydroponic systems: Phytotoxicity, surfactant degradation, and bacterial dynamics. Water Res. 34,
30753086.
Gross, A., Azulai, N., Oron, G., Ronen, Z., Arnold, M., Nejidat, A., 2005.
Environmental impact and health risks associated with greywater
irrigationa case study. Water Sci. Technol. 52 (8), 161169.
Gross, A., Shmueli, O., Wiel-Shafran, A., Oron, G., Ronen, Z., 2003.
Reuse of greywater for landscape irrigation in small communities and
households. Water and Irrigation 437, 2939 (in Hebrew).
Halperin, R., Aloni, U., 2003. Standards for Treated Wastewater Reuse in
the City, for Recreation and in Industry (in Hebrew). Ministry of
Health, Jerusalem, Israel, p. 15.
Israeli Water Law, 1975. Water Standards (Surface Water Quality) (in
Hebrew). The State of Israel, Jerusalem, Israel.
IWA (specialists group on use of macrophytes in water pollution control),
2000. Constructed Wetlands for Pollution Control. IWA Publishing,
London, UK.
Kadlec, R.H., Knight, R.L., 1996. Treatment Wetlands. CRC Press/Lewis
Publishers, Boca Raton, FL.

A. Gross et al. / Chemosphere 66 (2007) 916923

Lindstrom, C.R., 2000. Graywater: facts about graywaterwhat it is, how
to treat it, when and where to use it. Cambridge, MA.Electronic
manuscript. Available from: <www.graywater.com>.
Meaguer, R.B., 2000. Phytoremediation of toxic elemental and organic
pollutants. Curr. Opin. Plant Biol. 3, 153162.
Merck, 2000. Microbiology Manual. Laboratory Products Division,
Merck, Darmstadt, Germany.
Raskyn, I., Smith, R., Salt, D., 1997. Phytoremediation of metals: Using
plants to remove pollutants from the environment. Curr. Opin.
Biotechnol. 8, 222226.
Soil Plants Analysis Council, 1999. Soil Analysis Handbook of Reference
Methods. CRC Press, Boca Raton, FL.

923

Spackman, R., Entry, J.A., Sojka, R.E., Ellsworth, J.W., 2003. Polyacrylamide for coliform bacteria removal from agricultural wastewater.
J. Soil Water Conserv. 58, 276283.
Vymazal, J., Brix, H., Cooper, P.F., Habrel, R., Perer, R., Laber, J.,
1998. Removal mechanisms and types of constructed wetlands. In:
Vymazal, J., Brix, H., Cooper, P.F., Green, M.B., Habrel, R. (Eds.),
Constructed Wetlands for Wastewater Treatment in Europe. Backhuys
Publishers, Leiden, The Netherlands, p. 366.
Wiel-Shafran, A., Ronen, Z., Weisbrod, N., Adar, E., Gross, A., 2006.
Changes in soil properties following irrigation with surfactant-rich
greywater. Ecol. Eng. (in press).