See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/24309568

Development of a Low-Cost Wastewater Treatment System for Small-Scale

Wineries

Article  in  Water Environment Research · April 2009

DOI: 10.2175/106143008X274356 · Source: PubMed

CITATIONS READS
13 415

4 authors, including:

Michele Arienzo Evan Christen

University of Naples Federico II Penevy Services Pty Ltd
72 PUBLICATIONS   2,648 CITATIONS    116 PUBLICATIONS   1,677 CITATIONS   

SEE PROFILE SEE PROFILE

Wendy C Quayle
The Commonwealth Scientific and Industrial Research Organisation
27 PUBLICATIONS   820 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Postdoc View project

anthropic pollution in the environment View project

All content following this page was uploaded by Evan Christen on 01 December 2016.

The user has requested enhancement of the downloaded file.

Queries for waer-81-02-10

This manuscript/text has been typeset from the submitted material. Please check this proof carefully to make sure
there have been no font conversion errors or inadvertent formatting errors. Allen Press.
 Development of a Low-Cost Wastewater
Treatment System for Small-Scale Wineries
Michele Arienzo1*, Evan W. Christen1,2, Wendy Quayle1, Nicola Di Stefano1
ABSTRACT: A pilot-scale winery wastewater treatment system was
developed to treat wastewater produced by a small winery (approximately
1200 metric tons of grapes crushed). The pilot system consisted of
a sedimentation/aerobic process combined with a bioremediation wastewater
cell planted with Juncus ingens. The main design specifications, detailed
descriptions of the plant, and analysis of the influent and effluent character-
istics (pH, electrical conductivity, total suspended solids, chemical oxygen
demand [COD], etc.) are reported for each segment of the proposed system.
Over the study period, the mean winery wastewater flowrate was 3.5 m3/d
at organic loads of 5000 to 14 000 mg-COD/L. The study measured average
removal rates of 72% for COD and 65% for total organic carbon and
dissolved carbon. The application of wastewater to the soil increased the soil
salinity in the top 30 cm, but remained stable below this. The system also
seemed to be effective at neutralizing the pH of the acidic winery wastewater
and at removing the phosphorus pool (65%) in the wastewater, whereas the
levels of nitrogen and most of the cations increased in the treated effluent.
The absorbing/adsorbing and degradation capacity of the soil of the
wastewater bioremediation cell did not appear to be exhausted after one
vintage. Juncus ingens appeared to grow moderately well, until the end of
the vintage, when dieback began to occur. An infilling with organic matter of
the surface soil layer under the root zone was observed, which reduced water
infiltration and hence the system treatment capacity. The data provide
evidence that this is a potentially effective wastewater treatment approach
for small wineries located in rural areas. Water Environ. Res., 81 (2009).
KEYWORDS: winery, wastewater, treatment, chemical oxygen demand,
bioremediation, soil, irrigation.
doi:10.2175/106143008X274356
Introduction
The winemaking industry generates large volumes of wastewater
that mainly originates from various washing operations during the
crushing and pressing of grapes and from rinsing of fermentation
tanks, barrels, and other equipment or surfaces (Petruccioli et al.,
2002). Winery wastewater contains the remains of grape pulp, skin,
and seeds and different compounds used in filtration, precipitation,
and cleaning processes.
During the year, volumes and pollution loads vary in relation to
the processes being undertaken. Although it is high in organic load
(i.e., as much as 120 000 mg/L chemical oxygen demand [COD]/L;
Chapman et al., 2001), most wastewater is readily biodegradable.
The most readily biodegradable contaminants (i.e., sugars and
alcohols) tend to degrade first, leaving behind wastewater contain-
ing less easily degradable compounds (i.e., phenols and tannins).
1
CSIRO Land and Water, Griffith, NSW, Australia.
2
University of New England, Armidale, NSW, Australia.
* CSIRO Land and Water, PMB 3, Hanwood, 2680, Griffith, NSW,
Australia; e-mail: Michele.Arienzo@csiro.au.
Water Environment Research, Volume 81—Copyright Ó 2009 Water Environment Federation
Uncorrected Page Proofs
In the past, in Australia, winery wastewater was disposed of
directly into ponds, where wastewaters were held to allow solids
to settle; then, either the water was allowed to evaporate or it was
applied onto soil or into water courses. However, current
environmental controls prohibit the return of such waters to
watercourses. In small wineries, simplified systems with low energy
consumption—lagoons, wetlands, and land spreading/irrigation—
are important techniques for effluent treatment and disposal. This
form of treatment and disposal requires integration with the general
landscape capacity to receive the wastewater and obviously requires
that land be available (Bustamante et al., 2005). The Australian
wine industry has sought to establish new protocols for the usage of
its wastewater to maximize its reuse for irrigation and for improving
the fertility and structure of soils. The most prevalent form of
winery wastewater treatment in Australia is biological treatment,
followed by irrigating woodlots, vines, or pasture. Approximately
10% is being used to irrigate vines (Quayle et al., 2006). The large
biodegradable fraction of the organic load associated with winery
wastewater makes it amenable to treatment through aerobic and
anaerobic methods. Most of the biological treatment methods do
not provide a comprehensive solution because of the need to
dispose of sludge or other byproducts derived from the process.
Operating costs of wastewater treatment systems are between $500
and .$500,000 per year. Capital costs ranged from $14,000 to .$2
million (Quayle et al., 2006). The costs of advanced treatment
technologies can be prohibitive for small wineries of a few hundred
to a few thousand metric tons per year.
Small wineries in Australia are often located in rural areas and
thus have land available for treatment and disposal of winery
wastewater, making land-based treatment and disposal attractive.
As such, this research implemented, as a part of our pilot winery
wastewater treatment system, a wastewater bioremediation cell
(WBC), which is an innovative land-based disposal process
combining the soil Filtration and Irrigated Cropping for Land
Treatment and Effluent Reuse (FILTER) technique developed by
the Commonwealth Scientific Industrial Research Organisation of
Australia (Jayawardane et al., 1997, 1999) and the vertical flow
wetland approach to wastewater treatment.
The WBC relies on the soil to act as both a physical filter and as
a medium for chemical exchange and degradation processes. Our
study describes an approach to developing a wastewater treatment
system targeted at smaller scale wineries. Such a wastewater system
needed to be straightforward in design and management, in
acknowledgement of the limited resources available to wineries of
this scale. Key factors in the design were the minimization of capital
investment costs and operating costs (electricity, chemical, and
manpower) and making the system as automated and ‘‘robust’’ as
possible. A prototype system combining a sedimentation phase

Figure 1—Design of the pilot wastewater treatment system.
(SS), low-level aeration (AS), and the WBC was installed at a small
rural winery in southeastern Australia for the 2007 vintage.
This paper describes the principal design specifications of the
pilot-scale system and analysis of the influent and effluent character-
istics (pH, total suspended solids [TSS], COD, etc.) for each segment
of the pilot treatment system. The paper also reports the physical,
chemical, and hydraulic characteristics of the WBC before and after
winery wastewater disposal and the water quality parameters after the
winery wastewater had filtered through the WBC.
Materials and Methods
Site Selection. The trial of the experimental treatment system
was undertaken at a small winery crushing approximately 1200 metric
tons/y of grapes in southern New South Wales, Australia (latitude
234.339297, longitude 146.059659). The region is semi-arid with
approximately 1800 mm reference evapotranspiration and approxi-
mately 400 mm/y of rain. Before implementing the experimental pilot
system, the wastewater treatment consisted of coarse solid screening,
holding in a large tank, and then sprinkling onto a piece of land
covered by grass at the rear of the winery. This system led to soil
degradation, odors, and did not allow for reuse of the wastewater.
Construction of the pilot wastewater treatment system began in
December 2006, with the construction of the WBC, and then the
other components were added in readiness for the beginning of the
vintage period in February 2007. It was considered that this was
a pilot plant, and, as such, significant monitoring would be required
and adjustments to design and management made as required.
System Design and Construction. Coarse Solids Screen-
ing. The primary treatment was a commonly used wedge wire
rotary screen (Figure 1) designed to remove solids .1 mm, such as
grape skins, leaves, stems, and seeds. This screen was part of the
previous wastewater treatment system and was intended to remove
the particulate portion (suspended and settleable solids) of the
wastewater and reduce the COD of the raw wastewater.
Hydraulic Design. To scale the treatment system, the volume of
wastewater for treatment needed to be determined. As there were
no wastewater records, the volume was derived by assessing the
amount of grapes crushed. The overall crush of approximately 1200
metric tons/y of grapes was estimated to generate approximately
900 m3 of wine. Based on an average usage of 2 L of water/L of
wine, approximately 1800 m3/y of wastewater was assumed to be
produced. Assuming that 50% of this wastewater is created in the
approximately 3 to 4 months of vintage, this indicated mean
wastewater volumes of approximately 8 m3/d during vintage and
approximately 3.5 m3/d during the nonvintage period. The WBC
area and drainage design and control were designed using concepts
of maintaining an appropriate water balance and leaching fraction to
control salinity (Fouss et al., 2007).
Water Environment Research, Volume 81—Copyright Ó 2009 Water Environment Federation
Uncorrected Page Proofs
Arienzo et al.
Flow Equalization. A 10-m3 equilibration tank was installed
after the coarse solids screening. The outflow was regulated by an
automatic control valve, with opening cycles to avoid blockage by
sedimentation and compensate for changing levels in the tank, to give
a consistent outflow to the next treatment segment. Any overflow of
the equalization tank that occurred, as a result of unforeseen cir-
cumstances, was managed by a pipe to a wastewater storage pond.
Sedimentation Segment. A set of three polyvinyl chloride tanks
designed for domestic wastewater, each with a volume of 3 m3, was
installed below ground (Figure 1). The tanks were initially used to
test a simple sedimentation approach to remove suspended and
settable solids and then try to achieve anaerobic digestion later in
the trial. No pH adjustment of the wastewater was implemented, as
the aim was to minimize costs, and some literature reported
reasonable anaerobic degradation under a varying pH range (Perez
et al., 2005). The wastewater from this sedimentation segment was
pumped into the following aerobic segment.
Aerobic Segment. A low-energy aeration system, in the form of
a trickling filter, was built to take the wastewater after the
sedimentation segment (Figure 1). The trickling filter system
consisted of a bed of stones created in a concrete ring tank 1.9 m
high and 1.5 m diameter, through which the water trickled from top
to bottom. This treatment is a low-rate aeration process, because the
flow passes over the stones, absorbing oxygen from the air in the
voids of the filter bed.
The outflow from the aerobic segment was transferred by gravity
to a 13.5-m3 tank, where the treated wastewater was held before
release to the WBC.
Wastewater Bioremediation Cell. Figure 2 shows the design of
the WBC. This controlled drainage system enabled manipulation of
the water table and hence controls of the depth of the aerated and
anoxic soil layers above the drains, to maximize the pollutant
removal and provide adequate root zone conditions for plant growth.
It was approximately 10.0 m long, 10.0 m wide, and 2.0 m deep at
the center; the hole was lined with plastic, and 100-mm diameter
agricultural drainage pipes, with a geomembrane filter sock around
them, were placed at the deepest point running the full width. These
were then covered with 100 to 150 mm of washed 10-mm river
gravel. On top of the gravel, a soil mix was applied. The soil mix was
made up of the clay loam soil excavated from the hole and some
sandy loam soil plus grape marc. On top of the soil, 450 kg of
gypsum powder was spread, to improve the infiltration character-
istics of the soil, and, finally, a 100-mm layer of 10-mm river gravel
was added. The WBC plot was transplanted with a native rush
(Juncus ingens) living predominately along the Murray River in
New South Wales and Victoria. The plants were established for
6 weeks with freshwater before winery wastewater was applied. The
winery wastewater was applied between February and June 2007,
Arienzo et al.

Figure 2—Schematic diagram of the WBC.

from the holding tank by gravity to the center of the WBC, where it from the sump out to a holding pond. The total materials cost of
flooded out across the gravel. The outflow water from the aerobic the system was approximately U.S.$20,000. For more detail on the
segment was applied to the WBC, with irrigation volumes ranging technical design and setup of the complete system, see Di Stefano
between 2 and 12 m3, with a mean of 7 m3. The time span between et al. (2008).
each irrigation event was approximately 3 days between February 5 Wastewater Sampling. The wastewater volumes and chemical
and March 15, during which, the irrigation volume varied from 2 to composition at the inlet and outlet of each segment were monitored
7 m3. These volumes were frequently adjusted based on the actual from January to June 2007. Composite 1-L wastewater samples
evapotranspiration rate (ETo) measured at the site. The irrigation rate directly from the winery were taken over 24 hours by an automatic
was maintained at 110 to 130% of the ETo. The cumulative volumes sampler and returned to the laboratory immediately for analysis.
of effluent disposed on the WBC were approximately 100 m3. These Samples were collected on a daily basis during the vintage and
conditions were aimed at maintaining plant growth, treating the weekly in the nonvintage season.
winery wastewater, and allowing recovery of the physico, chemical, Physico-Chemical Analyses. The inlet and outlet water of the
and hydraulic soil properties of the WBC after each irrigation event. different segments of the system was tested for pH, electrical
The treated wastewater was drained from the WBC by gravity conductivity (EC), TSS, COD concentrations (mg/L), organic
through the agricultural drain pipe to a 3-m-deep sump and pumped loading rates (kg-COD/kg), total organic carbon (TOC), and
Water Environment Research, Volume 81—Copyright Ó 2009 Water Environment Federation
Uncorrected Page Proofs

dissolved organic carbon (DOC). All assays were carried out in
duplicate. Hydraulic retention times (HRTs) (days) and organic load
removal efficiency (COD removed/COD applied) were calculated.
In addition, total levels of nitrogen (N), phosphorus (P), sulfur (S),
calcium (Ca), magnesium (Mg), sodium (Na), potassium (K), boron
(Bo), copper (Cu), iron (Fe), silicon (Si), zinc (Zn), chlorine (Cl),
and sulfate (SO4) were analyzed following Standard Methods
(APHA et al., 1992). The COD was determined using a commercially
available high-range kit (Chemetrics Inc., Calverton, Virginia);
mercury-containing, 0 to 15 000 mg/L, supplied by Watertest
Systems (Minto, Australia) and a single-set wavelength photometer
(640 nm). An automated logging prototype was developed, using
a programmable logic control system for online monitoring of
volume, pH, electrical conductivity, and suspended solids of the
wastewater. The location of the sensors is shown in Figure 1. The
volumes of wastewater flowing in and out of each segment were
measured by magnetic propeller-type flow meters (Figure 1).
Soil-Plant Sampling and Analysis. Before wastewater was
applied to the WBC and after the study period, triplicate soil cores
were randomly taken from the WBC and divided into 0 to 30, 30 to
60, and 60 to 90 cm depths. The soils were air-dried, crushed, and
sieved through a 2-mm sieve. A Guelph permeameter (Eikelkamp,
Giesbeek, Netherlands) was used for in situ measurement of soil-
saturated hydraulic conductivity. Three sets of steady-state measure-
ments were made at depths of 15, 40, and 70 cm. The hydraulic
conductivity measurement system and procedure were described by
Chen and Banin (1975). Soil properties were characterized using
standard methods of soil analysis (Rayment and Higginson, 1992).
After 150 days of wastewater application, 10 Juncus ingens
plants were randomly sampled from the WBC by an auger
removing the entire plant up to 20 cm deep. Stem length was
measured, and leaf chlorophyll content was determined; two discs
(0.8-cm diameter) were collected at the mid-length of the leaf, from
both mid-rib sides. Discs were immersed in acetone and ground in
an agar mortar. After centrifugation at 4000 rpm for 30 minutes,
chlorophyll a (Chla) and b (Chlb) contents were determined by UV-
Vis spectrophotometry (Perkin Elmer Lambda 3B, Melbourne,
Australia). The absorbance of the solution was measured at 470,
647, and 664.5 nm. Formulas and extinction coefficients used for
determination of chlorophyllous pigments were published by
Lichtenthaler (1987). Roots were carefully removed from the soil,
gently shaken, and washed, to eliminate any attached soil. The dry
weight of plant parts (roots and shoots) was determined after oven-
drying at 708C. The results from these plants were compared with
nursery grown plants using fresh water.
Statistical Analysis. To determine the efficiency of the system,
the significance of the differences of water and soil quality
parameters of the inlet and outlet water and of plant growth on
control and treated soil was determined by an unpaired t-test
(p50.05). Relationships between variables were assessed using
linear regression and correlation analyses.
Results and Discussion
Influent Volume. Figure 3 shows the variation of the influent
flowrates from the beginning of February to the end of May 2007.
Several sharp peaks of up to 25 m3/d were observed, especially
during the vintage period; then, influent flowrates, with the
exception of a few peaks close to 10 m3 during May, tended to
level off to values lower than 5 m3. An accidental spillage occurred
in the winery at the end May, creating a peak of 10 m3/d of winery
wastewater to be treated. Overall, the winery generated approxi-
Water Environment Research, Volume 81—Copyright Ó 2009 Water Environment Federation
Uncorrected Page Proofs
Arienzo et al.
Figure 3—Variation of the influent and effluent rates of the
winery wastewater.
mately 380 m3 of wastewater during the study period, of which
approximately 100 m3 were processed by the WBC.
Wastewater Composition at Different Treatment Stages.
The water quality parameters monitored at the inlet and outlet of
each of the three segments of the treatment system from January 31
to June 6, 2007, are summarized in Table 1.
Sedimentation Segment. Both inlet and outlet waters were
acidic, with similar mean pH values of 5.2 and a similar range of
pH, from 3.5 to 7.0. The pH similarity means that no further
fermentation or precipitation of the organic acids occurred in this
section. Peaks in pH values of 3.5 of the inlet water were detected
during the vintage, at the end of February, and during nonvintage, at
the end of April, likely as a result of wine and juice loss, use of
mineral acids for recharging ion-exchange columns, and caustic
wash water (Chapman et al., 2001).
The inlet water electrical conductivity values ranged between
0.2 to 3.7 dS m21, with a mean value of 1.2 dS m21 and a relative
standard deviation (RSD) of 88%. The outlet water electrical
conductivity values had a slightly wider electrical conductivity
range between 0.3 and 4.1 dS m21 and a mean value of 1.5 dS m21.
Similar to pH, the salinity was extremely variable over time. Peaks
of electrical conductivity close to 4 dS m21 for the inlet water
occurred at the beginning of the vintage. The mean concentration of
TSS was similar for both inlet and outlet water, at approximately
280 mg/L. As for electrical conductivity, the highest TSS influent
values were found during the vintage operations, with maximum
values close to 800 mg/L. The inlet water was mostly free of large
suspended matter; this was removed by the rotary screen, so the
wastewater came to the sedimentation segment with a large load of
smaller suspended particles, primary dead yeast cells, cell fragments
(lees), grit, dirt, diatomaceous earth, and bentonite (Chapman et al.,
2001). This solid material was not removed when residence times in
the sedimentation unit were short (see following Aerobic Segment
section). As such, the mean COD values of the outlet water were
somewhat higher than that of the inlet water (7547 versus 5483
mg/L). This could be the result of the sedimentation system not
being in equilibrium, in terms of volumes of water added and
removed, and because the system was in a turbulent rather than
a steady state, so there was insufficient detention of the organic load
in this part of the system. Both inlet and outlet water showed similar
peak values of approximately 14 200 mg/L COD toward the end of
April. Thereafter, the values reduced to a smaller similar value
(approximately 280 mg/L) at the end of the sampling period. The
mean values of TOC and DOC tended to decrease by approximately
10% after the sedimentation segment. These observations, with
Arienzo et al.

Table 1—Mean (in parenthesis, RSD%), minimum, and maximum values of the chemical parameters and removal
efficiency of sedimentation segment (SS), anaerobic segment (AS), and WBC of the pilot-scale wastewater treatment.
Comparison of in- and out-water mean values by unpaired t-test are reported (key: ***p < 0.001, **p < 0.01, and *p < 0.05
significance, ns 5 not significant).

Parameter In-SS Out-SS % Rem t-test Out-AS % Rem t-test Out-WBC % Rem t-test

pH 5.1 (21) 5.2 (15) — ns 5.2 (16) — ns 7.0 (2) — *

Minimum 3.5 3.8 3.9 6.6
Maximum 7.0 6.8 7.0 7.4
Ec dS m21 1.2 (88) 1.5 (65) 225 ns 1.6 (67) 26.6 ns 3.6 (24) 2125 **
Minimum 0.2 0.3 0.7 0.3
Maximum 3.7 4.1 5.1 4.5
TSS (mg/L) 288 (73) 284 (52) 1.4 ns 110 (93) 61.2 ** 203 (58) 284.5 **
Minimum 26 50 63 19
Maximum 785 585 400 450
COD (mg/L) 5483 (86) 7547 (57) 237.6 * 7292 (54) 3.3 ns 1557 (114) 78.6 **
Minimum 0 560 470 0
Maximum 14040 14240 15400 6840
TOC (mg/L) 1796 (84) 1625 (68) 9.5 * 1700 (67) 24.6 ns 650 (84) 61.7 **
Minimum 130 130 180 120
Maximum 5210 3990 4430 2300
DOC (mg/L) 1502 (81) 1377 (71) 8.3 * 1490 (74) 28.2 ns 636 (84) 57.3 **
Minimum 130 100 170 110
Maximum 4340 3380 3790 2030
Total P (mg/L) 10.5 (44) 11.7 (58) 211.4 ns 11.5 (42) 1.7 ns 3.7 (50) 70.5 **
Minimum 3.3 1.7 4.6 1.6
Maximum 15.4 23.9 22.5 7.8
Total N (mg/L) 39.3 (77) 43.2 (77) 29.9 ns 40.3 (49) 6.7 ns 57.5 (86) 242.6 *
Minimum 3.0 6 4 32
Maximum 97 131 73 86
Ca (mg/L) 34.6 (27) 30.2 (34) 12.7 ns 35.0 (37) 215.8 ns 185 (22) 2360 **
Minimum 17.3 17.7 19.2 75
Maximum 49 70 76 253
Mg (mg/L) 11.0 (24) 10.3 (34) 6.4 ns 10.3 (34) 0.0 ns 132 (28) 21181 **
Minimum 6.0 5.1 5.7 20.7
Maximum 15.5 20.2 17.6 172
Na (mg/L) 202 (97) 174 (76) 14.2 ns 195 (69) 212.1 ns 230 (22) 216.0 *
Minimum 36.6 35.4 41 61
Maximum 720 532 585 303
K (mg/L) 287 (98) 264 (108) 8.0 ns 300 (117) 213.6 ns 450 (20) 250.0 **
Minimum 32.4 35 48.6 255
Maximum 997 1250 1520 651
Cu (mg/L) 0.1 (169) 0.1 (88) 0.0 ns 0.06 (70) 0.0 ns 0.1 (60) 0.0 ns
Minimum 0.008 0.006 0.01 0.007
Maximum 0.65 0.4 0.19 0.3
Fe (mg/L) 3.6 (38) 2.6 (91) 27.8 * 3.0 (63) 215.3 ns 28.7 (66) 2617 **
Minimum 2.1 0.64 0.74 7.0
Maximum 6.1 11.5 12.6 73.6
Si (mg/L) 7.0 (39) 5.5 (50) 21.4 * 5.8 (32) 25.4 ns 12.4 (21) 259.0 **
Minimum 1.8 2.4 4.0 5.0
Maximum 11.7 16 14 17

those relative to the constancy of the mean electrical conductivity Hydrous iron and aluminum oxides alone were reported to adsorb
and TSS values, led to the conclusion that sedimentation and 33% of the DOC (Dahm, 1981). The data appear to be consistent
precipitation processes were very limited. Only very limited with the removal of aluminum (approximately 25%), iron
removal of the organic carbon occurred, likely as a result of (approximately 28%), and silicon (approximately 21%), which are
adsorption on fine detritus, mainly consisting of uncrystallized the main constituents of the clay mineral lattice. Negative
aluminosilicate clays and hydrous oxides of iron and aluminum. significant correlation (p , 0.001) was found between DOC values
Water Environment Research, Volume 81—Copyright Ó 2009 Water Environment Federation
Uncorrected Page Proofs
 Arienzo et al.

and total concentration of aluminum and iron in the outlet water. Table 2—Mean, minimum, and maximum organic loading
Winery wastewater has a unique organic composition, with the (OL, kg COD d21) and removal efficiency as function
organic load mainly consisting of ethanol and sugars, organic acids, of the HRT in the sedimentation and aeration phases
aldehydes, other microbial fermentation products, and soaps and (SS and AS, respectively).
detergents from cleanup operations (Chapman et al., 2001).
Aerobic Segment. No significant degradation of the organic OL OL
matter was provided by this aerobic segment; the COD value of the
In-SS Out-SS Removal In-AS Out-AS Removal
outlet water ranged between 470 and 15 400 mg/L. Both the TSS
kg d21 kg d21 % kg d21 kg d21 %
and COD values of the outlet water were still largely in excess of
the Australian Environmental Protection Agency (Melbourne, 1-day HRT
Australia) regulatory limits for surface water of 50 and 15 mg/L, Mean 1.2 2.4 2203 2.4 1.9 22
respectively (ANZECC/NHMRC, 1992; Chapman et al., 2001). Minimum 0.2 0.7 21015 0.7 0.5 24
After the initial setup until approximately March 8, the system Maximum 1.9 4.3 17 4.3 4.5 33
operated under a highly variable flow regime, as a result of RSD(%) 56 63 2178 63 75 61
management difficulties. This greatly affected the effectiveness of 2-day HRT
Mean 3.2 3.5 225 3.5 2.5 24
the sedimentation and aeration units. The HRT of the sedimenta-
Minimum 0.6 1.2 285 1.2 0.8 25
tion/aerobic segments varied markedly, and the system was only
Maximum 5.2 5.9 34 5.9 4.6 75
able to be run effectively at constant HRTs during the post-vintage RSD(%) 54 53 2213 53 64 130
period. 5-day HRT
The mean organic loading (kg-COD/d) of the inlet and outlet Mean 4.4 1.8 67 2.2 1.5 39
water of the sedimentation and aeration phases relative to the HRTs Minimum 0.5 0.1 46 1.0 0.5 20
of 1, 2, and 5 days and the removal efficiency are shown in Table 2. Maximum 8.0 4.3 80 4.3 3.5 56
Clearly, the efficiency of substrate removal is a function of the HRT. RSD(%) 75 103 22 69 100 46
In the case of the sedimentation phase, only an HRT of 5 days was
effective, with a mean organic load removal of 67%, to a maximum
of 80%. This seems to be consistent with the fact that the unit is
essentially a sedimentation system, where only long HRTs are of the wastewater in the sedimentation and aeration segments of the
effective to allow settling of the smaller suspended particles. A long pilot system. Figure 4 shows the differences in the cumulative COD
HRT creates a high buffering capacity to cope with the highly load of the inlet and outlet water of the WBC. At the end of the
variable nature of the incoming wastewater. This is in agreement sampling period, the cumulative COD load of the effluent was
with findings of Kadlek and Knight (1996). In the case of the low- approximately 50% lower than that of the influent (60 versus 30 kg
level aeration phase, the organic load removal ranged from 22%, for COD). It is interesting to note that, whereas, during the vintage, the
an HRT of 1 day, to 39.5%, for an HRT of 5 days. Thus, it is likely maximum organic load removal did not exceed 30%, after vintage,
that, as the biological layer of the aeration phase thickens through the removal efficiency of the soil of the WBC appeared to increase
microbial growth with time, the aerobic degradation increases. with time, up to 50%. The COD cumulative load of the effluent, in
Wastewater Bioremediation Cell Segment. The wastewater fact, leveled off after vintage, to approximately 30 kg COD. The
underwent important changes in composition after passing through mean daily organic load of the effluent was less than half of the
the soil of the WBC (Table 1). The pH of the outlet water increased influent (0.9 versus 2.2 kg/d).
by approximately 2 units, which can be the result of the fixation The mean values of TOC and DOC show that most of the organic
onto clay particles of the organic acids, which are mainly carbon in the outlet water was in a soluble form (approximately
responsible for the low pH of winery wastewater (Perez et al., 90%, Table 1), and both significantly decreased, by approximately
2005). The pH values of the outlet water complied with the 60%. Removal of soluble organic carbon from the wastewater
regulatory rules, pH 5 to 8.5, for use as irrigation water (ANZECC/ by application to soil can be attributed to microbial uptake and
NHMRC, 1992). adsorption. Chapman et al. (1995a), in a study to determine the
The average value of the salinity of the outlet water was more than removal of DOC from synthetic winery wastewater by soil,
2 times that of the inlet water (3.6 versus 1.6 dS m21), as a result of determined that adsorption to soil particles reduced the DOC
evapoconcentration and the leaching of salts from the soil and the concentration by 16 to 49% within the first hour after application to
dissolution of soil carbonates by the acidic water. Whereas the inlet soil. The rate of organic carbon mineralization is faster for sugars
water electrical conductivity values varied from 0.7 to 5.1 dS m21, and organic acids and slower for ethanol, the major organic
the outlet water electrical conductivity value was more constant and component, and polyphenols (Saadi et al., 2006). Oxidative
close to 3.6 dS m21—a value which was only slightly higher than microbial decay was responsible for the removal of the remaining
the maximum limit for irrigation water of 3.3 dS m21. The TSS carbon in solution. Tare and Bokil (1982) reported that soil
concentration of the outlet water increased by approximately 85%. adsorption capacity appeared also to be correlated with the content
The higher TSS of the outlet water was probably the result of fine of silt and clay and organic carbon, which provide extended
materials, clay, and silt being leached out of the soil profile. surfaces for adsorption of the organic substrates and for support of
The COD values of the outlet water decreased by approximately microbial populations. Thus, the potential for rapid removal of
79%, with a mean value of 1500 mg/L, which was 5 times lower soluble organic carbon from winery wastewater may be partly
than that of the inlet water (mean 7300 mg/L). The COD data determined by soil texture. This in environmentally significant,
displayed several peaks throughout the sampling time. Most of because removal of the soluble carbon will prevent anoxic
these peaks appeared to be correlated with those recorded in the conditions from forming in soils being irrigated with the treated
previous segments and delayed, as a result of the hydraulic retention water from the WBC.
Water Environment Research, Volume 81—Copyright Ó 2009 Water Environment Federation
Uncorrected Page Proofs
Arienzo et al.

Figure 4—Cumulative values of COD concentrations of the influent and effluent and of volumes of the irrigation water
disposed on the WBC.

Total phosphorus concentration (Table 1) significantly decreased
in the outlet water (by approximately 70%), probably becoming
biologically incorporated, rather than precipitated by calcium. The
calcium concentration, with those of magnesium, sodium, potas-
sium, iron, and silicon, markedly increased, as a result of hydrolysis
of the clay minerals (Chapman et al., 1995b). Dissolution of soil
carbonate by the acidic winery wastewater will exhaust the soil’s
capacity to absorb/adsorb anions with time (Conte et al., 2001).
Chapman et al. (1995b) found that timing between successive
applications of winery wastewater to soils is fundamental for
adequate removal of soluble organic carbon.
Total mean nitrogen concentrations were typically quite low
(,60 mg/L), suggesting that most of the little nitrogen available
was used or converted by the Juncus ingens plants. Its value
increased by approximately 40% in the outlet water, possibly as
a result of nitrate concentrations from proteins in the grapes and
yeasts used in winemaking. Di-ammonium phosphate used in
winemaking is likely to be a causal factor for the presence of
nitrogen, which is typically low in such effluent.
Compared with the effluent guidelines for surface water bodies
and irrigation outlined by the Australian legislation (ANZECC/
NHMRC, 1992; Chapman et al., 2001), generally the quality of the
outlet water from the WBC complied with the rules for pH,
electrical conductivity, chlorine, copper, zinc, and sodium absorp-
tion ratio and exceed the regulations for the levels of COD, total
nitrogen, phosphorus, and sodium.
Effect of Wastewater on Soil. Table 3 shows that the soil in
the WBC was moderately alkaline and contained appreciable
amounts of organic matter through the profile. The soil was sandy
clay in the top 0 to 30 cm and clayey sand at 60 to 90 cm depth and
hence may be considered conductive for water movement, limiting
the likelihood of water standing on the surface that may lead to
odors or unsightliness. The cation exchange capacity (CEC) was
high and had consistent percentages of both calcium and potassium.
These characteristics are beneficial for sustained capacity to capture
the soluble components of the wastewater.
Table 4 shows that the chemical properties of the soil were
changed by application of the winery wastewater. There were
changes in the soil pH in the deeper layers, whereas the top layer
remained unchanged. The soil pH increased by 0.5 and 0.8 pH units
at depths of 30 to 60 and 60 to 90 cm, respectively. This can be
explained by the high soil buffering capacity (Tam and Wong,
1996) and the high levels of base cations in the influent. Also,
hydrogen ion (H1) would be forced off soil-exchange sites and
bound to nitrate (NO2 3 ), which then leaches from the system.
Several studies report increases of soil pH values following

Table 3—Soil properties in the WBC before disposal of winery wastewater (data are means of nine replicates).

EC Organic NO3-N NH4-N Tot-P Av-P. CEC Ca Mg Na K Cu Zn

Depth (dS carbon N CaCO3 (Cmol(+)/
(cm) pH m21) Texture (%) (mg/kg) (%) (mg/kg) (%) kg) (Cmol(+)/kg) (mg/kg)

0 to 30 7.8 0.73 Sandy 2.0 14 30 0.2 131 97 0.89 27 17.5 4.75 1.6 8.4 20 50
clay
RSD 3.5 52.1 18.2 54.1 88.2 10.2 15.5 24.2 18.5 30.2 95.2 25.1 10.2 20.1 41.2
30 to 60 8.0 0.43 Clayey 2.6 14 29 ,0.1 340 106 0.75 21 14.0 3.5 1.0 7.3 19 46
sand
RSD 4.1 54.3 19.2 58.2 87.5 12.2 14.5 25.5 20.2 32.5 94.2 24.5 12.2 18.5 42.2
60 to 90 7.8 0.56 Clayey 2.4 19 20 ,0.1 337 113 0.72 22 2900 3.6 1.1 7.9 19 42
sand
RSD 2.7 57.0 20.2 62 110 12.5 11.4 22.3 27.2 14.5 93.0 23.5 8.9 14.5 41.2

Water Environment Research, Volume 81—Copyright Ó 2009 Water Environment Federation

Uncorrected Page Proofs
 Arienzo et al.

Table 4—Soil properties in the WBC after disposal of winery wastewater (data are means of nine replicates).

EC Organic NO3-N NH4-N Tot-P Av-P. CEC Ca Mg Na K Cu Zn

Depth (dS carbon N CaCO3 (Cmol(+)/
(cm) pH m21) Texture (%) (mg/kg) (%) (mg/kg) (%) kg) (Cmol(+)/kg) (mg/kg)

0 to 30 7.7 1.2 Sandy 2.7 5.7 35 0.2 424 113 1.32 33 26.9 1.3 2.44 9.5 19.6 47.7
clay
RSD 5.4 50.1 17.5 65.0 97.2 10.2 22.2 26.2 19.2 31.2 95.2 27.5 10.1 22.2 45.2
30 to 60 8.5 0.50 Sandy 3.0 2.7 20 ,0.1 410 130 1.0 28 18.2 4.2 1.5 9.7 18.6 39.6
clay
RSD 5.8 52.5 18.8 63.2 99.2 11.2 27.4 24.5 21.2 30.2 94.2 25.5 8.5 23.1 48.2
60 to 90 8.6 0.40 Sandy 2.1 3.5 21.5 ,0.1 412 119 0.85 25 16.5 3.5 1.1 9.3 19.3 42.3
clay
RSD 5.7 51.4 20.8 66.6 104 12.8 24.4 27.4 20.1 28.0 96.5 29.4 9.8 21.9 44.8

application of effluents (Campbell et al., 1984; Falkiner and
Polglase, 1997). The application of winery wastewater induced an
approximately 2-fold increase in the electrical conductivity of the
surface layer (1.2 versus 0.73 dS m21), which was still well below
the threshold for a saline soil (4 dS m21). Thus, the mixed soil used
seems to have ensured adequate leaching and avoided excessive salt
accumulation.
The increase in soil salinity also involved the alteration of the
CEC (Cabrera et al., 2005). This alteration was more evident in the
surface and middle layers, with an average increase in CEC of
approximately 20%. The increase in CEC could be attributed to
newly generated negative charges on surfaces that possess variable
charge (i.e., pH-dependent) characteristics, such as edge sites of
clay minerals, sesquioxides, and the undissolved organic matter
remaining in the soil. Notwithstanding the high levels of sodium
and potassium in the wastewater applied to the WBC, no significant
increase in salinity or sodicity was observed.
Winery wastewater application induced an increase in organic
matter content in the upper soil layers with a slight migration
towards the lower horizons. The immobilization of the wastewater
organic matter in the upper layers facilitated its mineralization. This
is also shown by the significant increase of ammonium-nitrogen
(NH4-N) in the upper layers. However, the accumulation of organic
matter did result in reduced infiltration rates.
Overall, the soil did not undergo any drastic changes and can be
said to be in good condition, capable of continuing to treat
significant further volumes of wastewater.
Physical Properties. In situ measurements of the saturated
hydraulic conductivity revealed that the hydraulic conductivity of
the soil layer 0 to 70 cm of the WBC was very low—below 1 mm/h.
This indicates that there was a throttle in the soil upper layers. This
throttle is likely the result of the buildup of organic matter in the
surface layers. This infilling was caused by the suspended matter
that was abundant in the wastewater (approximately 200 mg/L),
which can seal the large pores of the soil. These results agree with
the observations of Seki et al. (1998), who studied the effect of
wastewater disposal on soil hydraulic conductivity. The authors
found that, when soil is continuously submerged with nutrient
solution, its saturated hydraulic conductivity decreases with time.
Allison (1947) explained that one of the causes of this decrease
is biological clogging of soil pores with microbial cells and their
synthesized products, slimes, or polysaccharides. Bacterially
synthesized gas can also decrease hydraulic conductivity, by
occluding pore necks as bubbles in sand (Seki et al., 1998). Other
Water Environment Research, Volume 81—Copyright Ó 2009 Water Environment Federation
Uncorrected Page Proofs
authors (Zenjari and Nejmeddine, 2001) also observed fast filling-in
of deep soil layers, where biodegradation is likely to be difficult.
Plant Growth. Growth of Juncus ingens in the WBC was
variable, with most of the plants showing symptoms of chlorosis.
Only a few plants displayed necrosis of the leaf, and a few (,5.0%)
died. Significant biomass differences (t-test [p , 0.05]) were
observed between plants grown with tap water and those grown
with winery wastewater, the former having higher biomass
production (approximately 60%). This might be related to the
reaction of the root system by enlarging the water and ion-absorbing
root surface and hence root biomass (Rozema and Visser, 1981).
The soluble organic carbon of winery wastewater can be converted
into soil organic carbon and thus improve soil fertility. Thus, it is
likely that disposing of the treated effluent on soil will have
a fertilizing effect. Moreover, plants grown with the wastewater
displayed similar shoot length (approximately 60 cm) and
chlorophyll content (0.20 mg/g) to those grown with tap water.
Conclusions
The proposed pilot winery wastewater system showed that the
system has the potential to provide an alternative system of
wastewater management for small wineries located in rural areas.
The preliminary analysis of soil characteristics, after the application
of winery wastewater, did not highlight significant problems of
degradation of the top soil, which is the most relevant to rhizosphere
activity. The consistent removal of the organic carbon through the
soil of the WBC means that this retained carbon can be considered
a soil fertility resource, as a result of its conversion to humic acids
and stimulation of soil microbial activities.
There is a need to investigate the physical, chemical, and
biological effects on soil and the potential toxicity to crops of the
residual organic carbon under long-term loading. This refractory
DOC, which has the potential to interact with hydrophobic organic
contaminants, assists in their co-transport to groundwater and
causes oxygen depletion. Until this subject is thoroughly
investigated, safe spreading of treated winery wastewater should
be restricted to areas that are not directly connected to drinking
water aquifers.
If land irrigation is the final fate of the treated winery wastewater,
the effluent generated by the system needs to be tested in a field
setting to evaluate the fate of nutrients, such as potassium, calcium,
and magnesium, for which, transport was increased in the soil of the
WBC. Vines and other crops (i.e., lucerne and oats) can take up
significant quantities of nutrients, such as potassium, during growth.
The wastewater generated by this winery, approximately 1800 m3/y,
Arienzo et al.
is negligible in terms of irrigation, considering that the annual
amount of water typically used to irrigate is approximately 5000
m3/ha. Thus, the winery wastewater could be also used as an
additional resource for irrigation of the grounds around the winery.
More research is needed to optimize the operation of the different
treatment segments. The aerobic segment displayed very little
potential to remove the organic load, and such a system is probably
ineffective. Some soil blockage in the WBC occurred because the
WBC was constantly loaded, with little opportunity for drying and
reduction of the anoxic conditions, which would have allowed
oxidation of the organic matter to occur. This surface soil throttling
led to the system treating less water than should have been possible,
based on the deeper soil hydraulic conductivity, which was shown
to be high and steady over time. This indicates that an additional
WBC would be required in this system, so that the wastewater could
be cycled. A low-cost, low-energy system, such as a horizontal flow
wetland, also could be used to significantly reduce the organic load
before application to the WBC. This would allow the WBC to
perform a ‘‘polishing’’ function and act as a buffer against low-
quality wastewater.
Credits
The authors acknowledge funding from the Grape and Wine
Research and Development Corporation of Australia (Goodwood,
South Australia) and the owners and staff of Piromit Wines,
Griffith, NSW, Australia, for choosing to be part of an experimental
design and for their invaluable assistance throughout the study.
Submitted for publication November 22, 2007; revised manuscript
submitted June 30, 2008; accepted for publication August 11, 2008.
References
Allison, L. E. (1947) Effect of Microorganisms on Permeability of Soil
Under Prolonged Submergence. Soil Sci., 63, 439–450.
American Public Health Association; American Water Works Association;
Water Environment Federation (1992) Standards Methods for the
Examination of Water and Wastewater, 19th ed.; American Public
Health Association: Washington, D.C.
ANZECC/NHMRC (1992) Australian and New Zealand Guidelines for the
Assessment and Management of Contaminated Sites. Australia and
New Zealand Environment and Conservation Council/National Health
and Medical Research Council: Canberra, Australia.
Bustamante, M. A.; Paredes, C.; Moral, R.; Moreno-Caselles, J.; Pérez-
Espinosa, A; Pérez-Murcia, M. D. (2005) Uses of Winery and Distillery
Effluents in Agriculture: Characterisation of Nutrient and Hazardous
Components. Water Sci. Technol., 51, 145–151.
Cabrera, M. L.; Kissel, D. E.; Vigil, M. F (2005) Nitrogen Mineralization
from Organic Residues. J. Environ. Qual., 34, 75–79.
Campbell, C. A.; Jame, Y. W.; Winkleman, G. E. (1984) Mineralization
Rate Constants and Their Use for Estimating Nitrogen Mineralization
in Some Canadian Soils. Can. J. Soil Sci., 64, 333–343.
Chapman, J.; Baker, A. P.; Wills, S. (2001) Winery Wastewater Handbook:
Production, Impacts and Management; Winetitles: Adelaide, Australia.
Chapman, J. A.; Correll, R. L.; Ladd, J. M. (1995a) Removal of Extractable
Organic Carbon from Synthetic Winery Wastewater by Repeated
Application to Soil. Aust. J. Grape Wine Res., 1, 76–85.
Chapman, J. A.; Correll, R. L.; Ladd, J. M. (1995b) Removal of Extractable
Organic Carbon from Winery and Distillery Wastewaters by Applica-
tion to Soil. Aust. J. Grape Wine Res., 1, 30–47.
Chen, Y.; Banin, A. (1975) Scanning Electron Microscope (SEM)
Observations of Soil Structure Changes Induced by Sodium-Calcium
Water Environment Research, Volume 81—Copyright Ó 2009 Water Environment Federation
Uncorrected Page Proofs
View publication stats
Exchange in Relation to Hydraulic Conductivity. Soil Sci., 120,
428–436.
Conte, G.; Martinuzzi, N.; Giovannelli, L.; Pucci, B.; Masi, F. (2001)
Constructed Wetlands for Wastewater Treatment in Central Italy. Water
Sci. Technol., 44, 339–343.
Dahm, C. N. (1981) Pathways and Mechanisms for Removal of Dissolved
Organic Carbon from Leaf Leachate in Streams. Can. J. Fish. Aq. Sci.,
28, 68–76.
Di Stefano, N.; Quayle, W.; Arienzo, M.; Zandona, R.; Blackwell, J.;
Christen, E. (2008) A Low Cost Land Based Winery Treatment
System: Development and Preliminary Results. CSIRO Land and Water
Science Report, 43/08.
Falkiner, R. A.; Polglase, P. J. (1997) Transport of Phosphorus Through Soil
in an Effluent-Irrigated Tree Plantation. Aust. J. Soil Res., 35, 385–397.
Fouss, J. L.; Evans, R. O.; Ayars, J. E.; Christen, E. W. (2007) Water Table
Control Systems, Chapter 18. In Design and Operation of Farm
Irrigation Systems, 2nd ed., Hoffman, G. J., Evans, R. G., Jensen,
M. E., Elliot, R. L (Eds.); American Society of Agricultural and
Biological Engineers: St. Joseph, Michigan.
Jayawardane, N. S.; Biswas, T. K; Blackwell, J.; Cook, F. J (1999) Research
Project on Land Treatment of Effluent from the Griffith City Council
Sewage Work—Report 9. CSIRO Land and Water Consultancy Report:
Griffith, NSW, Australia.
Jayawardane, N. S.; Blackwell, J.; Nicoll, G.; Wallett, D. J. (1997) Research
Project on Land Treatment of Effluent from the Griffith City Council
Sewage Work—Report 7. Final report on pollutant removal by the
FILTER system during the period November 1994 to November 1996.
Report prepared for GCC, DL&WC (PWD) and DPIE. Consultancy
Report No. 97/40. CSIRO: Canberra, Australia.
Kadlec, R. H.; Knight, R. L. (1996) Treatment Wetlands; Lewis Publishers:
Boca Raton, Florida.
Lichtenthaler, H. K. (1987) Chlorophylls and Carotenoids: Pigments of
Photosynthetic Biomembranes. Methods Enzymol., 148, 350–382.
Perez, M.; Romero, L. I.; Rodriguez-Cano, R.; Sales-Marquez, D. (2005)
Effect of the pH Influent Conditions in Fixed-Film Reactors for
Anaerobic Thermophilic Treatment of Wine-Distillery Wastewater.
Water Sci. Technol., 51, 183–189.
Petruccioli, M.; Duarte, J. C.; Eusebio, A; Federici, F. (2002) Aerobic
Treatment of Winery Wastewater Using a Jet-Loop Activated Sludge
Reactor. Process Biochem., 37, 821–829.
Quayle, W. C.; Christen, E. W.; Jayawardane, N. S.; Blackwell, J.; Corell,
R.; Fattore, A. (2006) Innovation in Management of Wastewater in the
Australian Wine Industry. Australian & New Zealand Grapegrower &
Winemaker, Annual Technical Issue, 41–49.
Rayment, G. E.; Higginson, F. R. (1992) Australian Laboratory Handbook
of Soil and Water Chemical Methods; Inkata Press: Melbourne,
Australia.
Rozema, J.; Visser, M. (1981) The Applicability of the Rooting Technique
Measuring Salt Resistance in Populations of Festuca Rubra and Juncus
Species. Plant Soil, 62, 479–485.
Saadi, I.; Borisover, M.; Armon, R.; Laor, Y. (2006) Monitoring of Effluent
DOM Biodegradation Using Fluorescence, UV and DOC Measure-
ments. Chemosphere, 63, 530–539.
Seki, K.; Miyazaki, T.; Nakano, M. (1998) Effects of Microorganisms on
Hydraulic Conductivity Decrease in Infiltration. Eur. J. Soil Sci., 49,
231–236.
Tam, N. F.; Wong, Y. S. (1996) Retention and Distribution of Heavy Metals
in Mangrove Soils Receiving Wastewater. Environ. Poll., 94, 283–291.
Tare, V.; Bokil, S. D. (1982) Wastewater Treatment by Soils: Role of
Particle Size Distribution. J. Environ. Qual., 11, 596–602.
Zenjari, B.; Nejmeddine, A. (2001) Impact of Spreading Olive Mill
Wastewater on Soil Characteristics: Laboratory Experiments.
Agronomie, 21, 749–755.