• No results found

TREATMENT OF HIGH ETHANOL CONCENTRATION WASTEWATER BY CONSTRUCTED WETLANDS: ENHANCED COD REMOVAL AND BACTERIAL COMMUNITY DYNAMICS

N/A
N/A
Protected

Academic year: 2021

Share "TREATMENT OF HIGH ETHANOL CONCENTRATION WASTEWATER BY CONSTRUCTED WETLANDS: ENHANCED COD REMOVAL AND BACTERIAL COMMUNITY DYNAMICS"

Copied!
15
0
0

Loading.... (view fulltext now)

Full text

(1)

1

TREATMENT OF HIGH ETHANOL CONCENTRATION

WASTEWATER BY CONSTRUCTED WETLANDS: ENHANCED

COD

REMOVAL AND BACTERIAL COMMUNITY DYNAMICS

A. Rodriguez Caballero*, J-B. Ramond**, P.J. Welz***, D.A. Cowan**, M. Odlare*, S.G. Burton***

*School of Sustainable Development of Society and Technology, Mälardalens University, Västerås (Sweden)

**Institute for Microbial Biotechnology and Metagenomics, University of the Western Cape, Cape Town (South Africa)

***Biocatalysis and Technical Biology Research Group, Cape Peninsula University of Technology, Cape Town (South Africa)

Abstract Winery wastewater is characterized by its high chemical oxygen demand

(COD), seasonal occurrence and variable composition, including periodic high ethanol concentrations. In addition, winery wastewater may contain insufficient inorganic nutrients for optimal biodegradation of organic constituents. Two pilot-scale constructed wetlands (CWs) were used to treat artificial wastewater: the first was amended with ethanol and the second with ethanol, inorganic nitrogen (N) and phosphorus (P). A number of biochemical parameters involved in the degradation of pollutants through CW systems were monitored, including effluent chemistry and bacterial community structures. The nutrient supplemented CW showed efficient COD, N and P removal. Comparison of the COD removal efficiencies of the two CWs showed that N and P addition enhanced COD removal efficiency by up to 16%. Molecular fingerprinting of CW sediment samples using denaturing gradient gel electrophoresis (DGGE) showed that amendment with high concentrations of ethanol destabilized the microbial community structure, but that nutrient supplementation countered this effect.

Keywords: Bacterial community structures; Chemical Oxygen Demand; Constructed wetland; Ethanol; Nutrients; Wastewater.

(2)

INTRODUCTION

The wine industry is an important sector of the South African economy, especially in the Western Cape area, where most wine-lands and wine cellars are situated. The quantity of wastewater produced by the winery industry in South Africa is greater than 109 L year-1, representing a considerable threat to the environment (Sheridan, 2007). Due to the characteristics of the winery industry itself, i.e. many small cellars with low volume and localized wastewater release, and the high investment and maintenance costs of conventional treatment systems, effluent treatment is not a common practice for the winery sector, contributing to environmental degradation in drainage basins in South Africa (Burton et al., 2007). Untreated winery wastewater discharged into water-bodies or land areas can produce detrimental effects on the environment. When released in excess, the organic matter contained in winery effluent may cause oxygen depletion (due to bacterial consumption) in aquatic environments, which impacts on the function and dynamics of the ecosystems. In addition, the high phenolic content of winery wastewaters can result in phytotoxicity (Mena et al., 2009). Winery wastewater treatment must be applied locally due to the variation in composition presented by effluents from different sources, taking into account the costs as well as the quality of the effluent water.

The high organic content, variable composition, lack of inorganic nutrients and the large seasonal flow variations of winery wastewater may result in operational difficulties associated with treatment. Chemical oxygen demand (COD) values around 5000 mg L-1 are common, increasing up to 25000 mg L-1 depending on harvest loads or processing activities (Malandra et al., 2003). The organic composition of winery wastewater varies significantly in different locations (Malandra et al., 2003). However, in some cases, ethanol constitutes up to 90% of the COD content in winery wastewater (Sheridan, 2007). In addition, winery wastewater may contain sub-optimal quantities of inorganic nutrients necessary for biodegradation (Ganesh et al., 2009). Thus, nitrogen (N) and phosphorus (P) have been recommended as supplements in nutrient deficient biological treatment processes (Andreottola et al., 2002; Bories et al., 2005). Finally, the seasonal flow variations of winery wastewater have to be taken into account and the selected treatment system must be capable of adapting to vacillating flow-rates (Shepherd et al., 2001). Numerous aerobic and anaerobic suspended and attached growth systems for the treatment of winery wastewater have been evaluated (Andreottola et al., 2009). For example, high COD removal performances (98 % or more) were achieved in an aerobic activated sludge system and in an anaerobic sequencing batch reactor (Fumi et al., 1995; Ruiz et al., 2002), but these processes are costly, energy intensive and require lengthy start-up periods. As the production of winery wastewater is intermittent, conventional systems need to be periodically re-acclimated (usually bi-annually during the crush and bottling seasons) (Grismer et al., 2003; Eusébio et al., 2004). Constructed wetlands (CWs) have low maintenance costs, minimal operational energy requirements, are capable of adapting to periodic variations in flow and organic content and have been recognized as being suitable for treating various wastewaters (for review, see Vymazal, 2009). Providing sufficient land is available, CWs provide an attractive wastewater treatment alternative for small to medium sized wineries, with proven efficiencies in different countries and a variety of conditions (Shepherd et al., 2001; Masi et al., 2002; Grismer et al., 2003; Sheridan, 2007), but more information and understanding regarding the chemical and biological parameters of the process are needed.

(3)

In this study, pilot-scale CWs were amended with artificial wastewaters containing high concentrations of ethanol (79 mM,

the basis of the hypothesis that

to an increase in the COD removal efficiency in

The overall aim of this research project was to increase the understanding of the chemical and biological aspects involved in CW systems in response to ethanol, and to build the foundations for further investigations on the potential use of CWs for the trea

COD removal performance and the biotransformation of N, as well as N and P effluent contents and the evolution of bacterial community structures were assessed in this study

MATERIALS AND METHODS

Experimental set-up

Three unplanted pilot-scale CWs (A

inoculated in a ratio of 1:4 with sediment from a local wetland treating winery wastewater and sand

from the Malmesbury river area

~0.5 m³. The CWs were operated in a hybrid mode The general experimental set-up is shown in Fig. 1

Figure 1 Set-up and mode of operation of the experimental constructed we

The establishment of the bacterial population 0.3 g of glucose and 0.3 g of yeast extract months before the start of the experiment

weekly feeding regime was established for each of the in Table 1.

3

scale CWs were amended with artificial wastewaters containing high concentrations of ethanol (79 mM, theoretical COD 7587 mg L-1). The study was designed

that the addition of nutrients (N and P) in an appropriate to an increase in the COD removal efficiency in CWs due to the enhancement of

he overall aim of this research project was to increase the understanding of the chemical and biological aspects involved in CW systems in response to ethanol, and to build the foundations for further investigations on the potential use of CWs for the treatment of winery wastewater. The removal performance and the biotransformation of N, as well as N and P effluent contents and the evolution of bacterial community structures were assessed in this study

MATERIALS AND METHODS

scale CWs (A, B and C) consisting of identical polyethylene tanks

of 1:4 with sediment from a local wetland treating winery wastewater and sand

Malmesbury river area (Western Cape, South Africa) to give a final sediment volume of The CWs were operated in a hybrid mode of vertical and horizontal sub

up is shown in Fig. 1.

up and mode of operation of the experimental constructed wetlands.

ent of the bacterial population in each CW was induced by bi 0.3 g of glucose and 0.3 g of yeast extract (base feed) in 12.5 L of tap water

of the experiment (Ramond et al., submitted). Once

weekly feeding regime was established for each of the CWs over a period of 50 days, as outlined scale CWs were amended with artificial wastewaters containing high ). The study was designed under n appropriate ratio can lead s due to the enhancement of biodegradation. he overall aim of this research project was to increase the understanding of the chemical and biological aspects involved in CW systems in response to ethanol, and to build the foundations tment of winery wastewater. The removal performance and the biotransformation of N, as well as N and P effluent contents and the evolution of bacterial community structures were assessed in this study.

identical polyethylene tanks were

of 1:4 with sediment from a local wetland treating winery wastewater and sand

to give a final sediment volume of and horizontal sub-surface flow.

by bi-weekly addition of in 12.5 L of tap water over the period of 3 Once equilibrated, a bi-over a period of 50 days, as outlined

(4)

Table 1 Bi-weekly feeding regimes and artificial wastewater composition for CW A, B and C during the experimental period. Time period CW A (V= 12.5 L water) CW B (V= 12.5 L water) CW C (V= 12.5 L water) Day 0 (end of equilibration)

Base feed Base feed Base feed

Days 1 – 8 Base feed Base feed Base feed + 13.54 g

KN0₃ + 1.64 g KH₂P0₄ Days 9 – 35 Base feed Base feed

Ethanol (0.46 %)

Base feed + 13.54 g KN0₃ + 1.64 g KH₂P0₄ Ethanol (0.46 %) Days 36 – 50 Base feed Base feed

Ethanol (0.46 %)

Base feed + 20.31 g KN0₃ + 2.46 g KH₂P0₄ Ethanol (0.46 %)

CW A was used as control, and the bacterial population was maintained by supplying the same concentration of glucose and yeast extract on the same levels during the experimental period. Additionally, the artificial wastewater prepared for CWs B and C contained 0.46% ethanol from day 9, equivalent to 7587 mg L-1 of COD, a level which was found in winery wastes from a South African cellar (data not shown). Potassium nitrate (KN0₃) and potassium di-hydrogen phosphate (KH₂PO₄) were added only to CW C, at two different ratios as follows: A COD/N/P ratio of 300/5/1 was applied during the first part of the experiment (days 9 to 35), and the concentrations of KNO₃ and KH₂PO₄ were then increased until the ratio reached 200/5/1 (days 36 to 50). These values are approximately in the range of commonly reported COD/N/P ratios in anaerobic wastewater treatment processes (Metcalf and Eddy, 2003).

Sampling and monitoring of CWs performance

Composite core sediment samples were taken on a weekly basis from the inlet and outlet areas, at two different depths (0 to 3 cm and 10 to 15 cm). Effluent water samples were taken twice a week, between 1 and 2 hours after the initiation of each feed. Nitrate reductase activity was assessed weekly in sediment samples, and results were expressed in terms of the amount of nitrite produced from nitrate per dry weight gram of sediment within the experimental period (Abdelmagid and Tabatabai, 1987, modified by Kandeler, 1996). COD, total nitrogen (TN), nitrate nitrogen (N0₃⁻-N), ammonium nitrogen (NH₄⁺-N) and total phosphorus (TP) concentrations were measured on effluent samples on a bi-weekly basis using colorimetric methods according to manufacturer’s instructions (Hanna Instruments, Inc., USA).

Data treatment

Significant differences between the removal performances and effluent chemical compositions of the experimental CWs were evaluated by one-way ANOVA followed by Tukey’s HSD multiple

(5)

5

comparison tests using the software package PASW Statistics 18.0 (SPSS Inc., Chicago, IL, USA).

Metagenomic DNA extraction and quantification

Metagenomic DNA was extracted from sediment samples using a modified version of the method described by Wang et al. (1996). Approximately 500 mg of sediment was resuspended in 1 ml lysis buffer (25 mM Tris-HCl pH 8; 50 mM glucose; 10 mM EDTA; 25 mg ml-1 lysozyme). Microbial cells were lysed by both mechanical and chemical methods. Mechanical shearing involved three cycles of vortexing at maximum speed for 1 min followed by rapid cooling on ice for 2 min. The samples were then incubated overnight at 37°C. SDS was added to a final concentration of 1% and the samples were incubated at 65°C for 30 min. Samples were extracted twice with 1 vol equilibrated phenol (pH 7.6), followed by 1 vol chloroform:iso-amyl alcohol (24:1, vol/vol). Nucleic acid was precipitated with 1 vol ice-cold isopropanol at room temperature and harvested by centrifugation at 14 000 g for 5 min. The resulting pellet was resuspended in 50 µl TE buffer (10 mM Tris-HCl; 1 mM EDTA, pH 7.8). The concentration of the DNA-samples was measured with a NanoDrop spectrophotometer (NanoDrop Technologies, Montchanin, DE, USA).

PCR amplifications

All the polymerase chain reactions (PCR) were carried out in a Perkin Elmer Thermocycler (Gene Amp PCR system 6700). Bacterial 16S rRNA encoding genes were amplified by PCR using the universal primers E9F GAGTTTGATCCTGGCTCAG-3′) and U1510R (5′-GGTTACCTTGTTACGACTT-3′). PCR was carried out in 50 µl reaction volumes. Each reaction contained 1X PCR buffer, 0.2 U DreamTaq™ polymerase (Fermentas, USA), 200 µM of each dNTP, 0.5 µM of each primer, 0.1% BSA and between 5 to 10 ng of metagenomic DNA. PCR amplification was carried out as follows: 4 min at 94°C for denaturation; 30 cycles of 30 s at 94°C, 30 s annealing at 52°C and 105 s at 72°C; and a final elongation step of 10 min at 72°C. To perform denaturing gradient gel electrophoresis (DGGE), a nested-PCR was undergone using 1 µL of the amplicon obtained with the 16S rRNA primer set E9F/U1510R with the primer set 341F-GC (5’-CCTACGGGAGGCAGCAG-3’)/514R (5’-ATTACCGCGGCTGCTG-3’) as follows (Muyzer et al., 1993): 94°C for 4 min; 20 cycles - 94°C for 45 s; 65°C for 45 s; 72°C for 60 s additional 20 cycles - 94°C for 30 sec; 55°C for 30 sec; 72°C for 60 sec; and a final elongation step at 72°C for 10 min. PCR amplification with 341f-GC/534r was performed by using a 50 µl total volume mixture containing 0.2 U DreamTaq™ polymerase (Fermentas, USA), 1X PCR Buffer, 200 µM of each dNTP, 0.5 µM of each primer and 0.1% BSA. For the DGGE

analysis, a 40mer GC-clamp

(CGCCCGCCGCGCGCGGCGGGCGGGGCGGGGGCACGGGGGG) was added to the 5’ ends of the forward primers 341F.

Denaturing gradient gel electrophoresis (DGGE)

PCR amplicons obtained with the nested primer sets (341F-GC/534R) were analyzed by DGGE. Amplicons were separated on 16.5/16.5 cm, 1 mm 9% (wt/vol) polyacrylamide (37.5:1 acrylamide:bisacrylamide) gels with varying denaturing gradients (100% denaturant was 7 M urea and 40% (vol/vol) formamide). Gels were prepared using a gradient former and were cast according to manufacturer’s specifications (BioRad Laboratories, Inc. USA). Electrophoresis was performed using the DCode DGGE system (BioRad Laboratories, Inc. USA) and was carried out

(6)

at 100 V for 16 hrs at 60 °C (1600 Vh) in 1X TAE buffer. Gels were stained in 0.5 µg/ml EtBr in 1X TAE for 20 min and visualized on an AlphaImager 3400 imaging system.

DGGE gel pictures were processed with Gelcompar II 5.0 software (Applied Maths, Belgium). The DGGE banding patterns were compiled and band matching was performed in order to obtain large data matrixes based on the presence/absence of DGGE bands. The complete banding pattern of each profile was considered for comparison. The presence/absence matrix was used to perform Principal Component Analysis (PCA). On the PCA diagrams, the complex DGGE patterns of each sample was reduced to one point in a three dimensional space. In PCA diagrams, the relative importance of each axis is given by the percentage of variance explained by that axis. The proximity of two points on the diagram is related to the community structure: points which are close together represent similar communities and vice versa.

RESULTS AND DISCUSSION

Enzyme activity and effluent nitrogen content

The reduction of N compounds through the experimental CWs was assessed by means of chemical characterization of effluent samples, measuring the content of TN, N0₃⁻-N and NH4+-N. In addition, nitrate reductase activity in sediment samples was monitored in order to evaluate the biological N0₃⁻-N reduction in the experimental CWs. No enzyme activity was registered in the deep sediments (data not shown). Enzyme activity measured in surface sediments of the three experimental CWs is shown in Fig. 2a. As expected, CW A (control) showed the lowest nitrate reductase activity values throughout the experimental period, ranging from 0.05 to 0.19 µg N0₂⁻-N gdw-1day-1. CW B (ethanol) showed relatively high nitrate reductase activity as compared with CW A (control) from day 8, reaching 15.63 µg N0₂⁻-N gdw-1day-1 on day 21. In the case of CW C (ethanol/nutrients) the nitrate reductase activity was in the same range as CW B (ethanol) after the first addition of artificial wastewater, while after day 8, the enzymatic activity decreased to values under 2 µg N0₂⁻-N gdw-1day-1. Finally, during the last two weeks of measurements (days 28 to 35), CW C (ethanol/nutrients) showed no measurable nitrate reductase activity. Therefore, the nitrate reductase measurements were discontinued at day 35. However, influent N0₃⁻-N from the artificial wastewater was degraded in CW C as shown by the stability of the effluent N0₃⁻-N concentrations (Figs. 2b and 2c).

Figs. 2b, 2c and 2d show the effluent TN, N0₃⁻-N and NH₄⁺-N concentrations of the CWs. TN concentrations in the effluent of CW A (control) and CW B (ethanol) did not differ significantly throughout the experiment (Tukey, p = 0.05). In contrast, the effluent of CW C (ethanol/nutrients) presented a peak TN (39 mg L-1) coinciding with the first addition of KNO3 (day 8), when ethanol was not yet included in the influent, and a second smaller increase towards the end of the experimental period (Fig. 2b). A similar pattern could be observed in the concentration of N0₃⁻-N. A peak of N0₃⁻-N in the effluent from CW C was shown on day 8 (no ethanol added). After the addition of ethanol, the N0₃⁻-N concentration in the effluent from CW C decreased substantially and reached values as low as 0.8 mg L-1 by day 38 (fig.2c). The N0₃⁻-N concentrations in the effluent from CW C (ethanol/nutrients) were very similar (approximately 1 mg L-1) to those from CW A (control) and CW B (ethanol) from day 35 to 50 (last 2 weeks of experiment). In terms of NH₄⁺-N, fluctuations from 0 to 2 mg L-1 were observed in the effluent samples of the experimental CWs, while concentrations in CW A (control) were more stable, ranging from 0.24 to 0.63 mg L-1. Generally, results from all three CWs were similar until the last

(7)

third of the experiment (day 35 to 50) when the CW C to peak values of 0.86 and 1.83 mg L

with NH₄⁺-N toxicity in aquatic ecosystems has been recommended

aquatic animals as follows: 0.05

long-term exposures (US EPA, 1986). In this context, the effluent experimental CWs of this study can be considered toxic if released into water need to be subjected to further treatment.

Figure 2 (a) Nitrate reductase activity in superficial sediments, and effluent concentrations of (b) Total nitrogen (TN), (c) N0₃⁻-N and (d) NH₄⁺-N from CWs A. B and C during the experimental period.

The effective N0₃⁻-N reduction achieved by the nutrient supplem with the low nitrate reductase

nitrate degradation in CW C were denitrification (DN) and dissimilatory nitrate reduction to ammonia (DNRA) possibly occurring si

Bonin et al., 1998; Fazzolari

dominance of DN or DNRA bacterial populations is dictated by the competition for carbon sources (Tiedje, 1988; Kelso et al.

which N0₃⁻-N is reduced to nitrite (

N02⁻-N metabolism follows different reduction pathways, either being reduced to nit (NO) during DN, or to NH4+

(ethanol) and C (ethanol/nutrients) shown in Fig.2d may have been formed, at least to some degree, via the DNRA pathway.

7

third of the experiment (day 35 to 50) when the NH₄⁺-N concentration increased in CW B and CW C to peak values of 0.86 and 1.83 mg L-1 respectively (fig.2d). Water quality

toxicity in aquatic ecosystems has been recommended in order

0.05-0.35 mg L-1 for short-term exposures and 0.01

(US EPA, 1986). In this context, the effluent NH experimental CWs of this study can be considered toxic if released into water need to be subjected to further treatment.

(a) Nitrate reductase activity in superficial sediments, and effluent concentrations of (b) Total nitrogen N from CWs A. B and C during the experimental period.

reduction achieved by the nutrient supplemented CW C did not correlate activity levels that were measured. The most likely mechanisms for nitrate degradation in CW C were denitrification (DN) and dissimilatory nitrate reduction to ammonia (DNRA) possibly occurring simultaneously at varying rates (Stevens

et al., 1998; Kelso et al., 1997). When N0₃

dominance of DN or DNRA bacterial populations is dictated by the competition for carbon et al., 1999). The DN and DNRA pathways share a reaction step in is reduced to nitrite (N02⁻-N) by the nitrate reductase enzymatic system. However, metabolism follows different reduction pathways, either being reduced to nit

+-N by DNRA (Zumft, 1997). Therefore, the

(ethanol) and C (ethanol/nutrients) shown in Fig.2d may have been formed, at least to some degree, via the DNRA pathway.

concentration increased in CW B and ater quality criteria related in order to protect sensitive and 0.01-0.02 mg L-1 for NH₄⁺-N content in the experimental CWs of this study can be considered toxic if released into water-bodies and would

(a) Nitrate reductase activity in superficial sediments, and effluent concentrations of (b) Total nitrogen

ented CW C did not correlate ctivity levels that were measured. The most likely mechanisms for nitrate degradation in CW C were denitrification (DN) and dissimilatory nitrate reduction to multaneously at varying rates (Stevens et al., 1998; ₃⁻-N is not limited, the dominance of DN or DNRA bacterial populations is dictated by the competition for carbon , 1999). The DN and DNRA pathways share a reaction step in enzymatic system. However, metabolism follows different reduction pathways, either being reduced to nitric oxide ). Therefore, the NH₄⁺-N in CWs B (ethanol) and C (ethanol/nutrients) shown in Fig.2d may have been formed, at least to some

(8)

While the reduction of N02⁻-N to NH4+-N (DNRA) is driven by a nitrite reductase system encoded by nrfA, the reduction of N02⁻-N to NO (DN) is catalyzed by different enzymes encoded by nirK or nirS, (for reviews see Zumft, 1997; Simon, 2002). In this study, the assay for nitrate reductase activity was based on the inhibitory effects of 2,4 dinitrophenol (2,4-DNP) on a generic N02⁻-N reduction, which allows the measurement of the nitrate reductase activity by means of N02⁻-N accumulation (Abdelmagid and Tabatabai, 1987). However, the sensitivity of the different enzymatic systems employed in DN and DNRA towards 2,4-DNP may be different, representing a possible source of error during the measurement of the nitrate reductase activity in the sediment samples taken from CW C (ethanol/nutrients). The maintenance of higher pH values in sediment samples from the nutrient supplemented CW C (up to 1.3 units higher than CW B, data not shown) supports the hypothesis of occurrence of DNRA, as this process has been suggested to be favored by relatively high pH values (Stevens et al., 1998). In a similar case, failure of the nitrate reductase activity measurement method used in this study was also reported by Deiglmayr (2006), suggesting that 2,4-DNP was unable to inhibit the N02⁻-N reductase activity of a N0₃⁻-N supplemented soil, resulting in an underestimation of the nitrate reductase activity. Therefore, for further studies, optimization of this methodology in cases in which additional N0₃⁻-N is applied is strongly recommended.

Effluent phosphorus content

Not only N, but also P compounds released into the environment can lead to eutrophication, which impacts the dynamics and functioning of natural ecosystems due to different effects such as changes in plant, animal and microbial community structures and fluctuations in pH or oxygen depletion among others (Smith et al., 1999). Both N and/or P have been reported as key factors in eutrophication processes in different locations (Howarth, 1988). There is a general agreement on the fact that eutrophication is more often limited by N in terrestrial areas and marine environments, and by P in the case of freshwater ecosystems (Howarth, 1988; Smith et al., 1999). Thus, the presence of P in the effluents of the experimental CWs was closely followed by chemical analyses and needs to be discussed. TP concentrations in effluent samples from CW B (ethanol) and CW C (ethanol/nutrients) were significantly higher than those from CW A (control), specially from day 35 to the end of the experiment (Fig. 3), showing 0.549 (CWs A and B) and 0.088 (CWs A and C) significant differences (Tukey, p = 0.05). On day 42, CW B presented a peak TP concentration of 4.06 mg L-1, while on day 45, effluent TP concentrations in CW C increased up to 10.9 mg L-1. Increasing concentrations of TP released from CW B and CW C could have been due to the process of draught and rewetting of the CWs. As reported by Song et al. (2007), the occurrence of drying and rewetting causes the release of P by desorption of previously adsorbed phosphate and by the higher mineralization of organic P due to the activation of the phosphatase. However, other studies have suggested that the release of P in CWs can be a consequence of anaerobic (reducing) conditions as a result of the reduction of P retaining iron compounds (Masscheleyn et al., 1992; Richardson and Craft, 1993).

(9)

Figure 3 Total Phosphorous (TP) concentration in effluent samples from CW A, B and C during the experimental period.

Variations in the COD/N/P ratios could res

control and monitoring is required. The treatment of winery wastewater with supplementation of N and P prior to discharge could result in eutrophication and/or toxicity being developed in the surrounding areas. Thus, mechanisms to enhance P and N removal would be necessary, based on the observation that the concentration of TN and

experiment in CW C (ethanol/nutrients) (Figs. 2b and 3). In this context, the add nutrients to CW systems is questionable and must be subject of further research.

COD removal

The COD removal performance from the experimental CWs was evaluated taking into account the influent COD value employed in the artificial wastewa

fig.4a), both CW B (ethanol) and CW C (ethanol/nutrients) differed significantly from CW A (control) (Tukey, p = 0.05). In CW

throughout the experiment. In

first addition of ethanol on day 14 (Fig. 4a)

1600 mg L-1 while in the case of CW C, COD peaked at a value of 1076 mg L

then exhibited a decreasing trend. At the end of the experiment, the effluent values were similar to those from the CW A (control) (81 mg L

COD removal efficiency of nutrient non-nutrient amended CW B

resulted in enhanced COD removal in CW C.

9

Total Phosphorous (TP) concentration in effluent samples from CW A, B and C during the experimental

Variations in the COD/N/P ratios could result in differences in the effluent quality

control and monitoring is required. The treatment of winery wastewater with supplementation of N and P prior to discharge could result in eutrophication and/or toxicity being developed in the areas. Thus, mechanisms to enhance P and N removal would be necessary, based on the observation that the concentration of TN and TP increased substantially at the end of the experiment in CW C (ethanol/nutrients) (Figs. 2b and 3). In this context, the add

nutrients to CW systems is questionable and must be subject of further research.

The COD removal performance from the experimental CWs was evaluated taking into account the influent COD value employed in the artificial wastewater (Fig.4a and 4b). As shown in fig.4a), both CW B (ethanol) and CW C (ethanol/nutrients) differed significantly from CW A (control) (Tukey, p = 0.05). In CW A (control), the COD remained constantly close to zero

n the effluent from CW B and CW C, the COD on day 14 (Fig. 4a). The effluent COD from CW in the case of CW C, COD peaked at a value of 1076 mg L

then exhibited a decreasing trend. At the end of the experiment, the effluent values were similar to those from the CW A (control) (81 mg L-1), reflecting close to 100% removal efficiency. The COD removal efficiency of nutrient-amended CW C was consistently 5%

(Fig. 4b). These results demonstrate that the addition of nutrients resulted in enhanced COD removal in CW C.

Total Phosphorous (TP) concentration in effluent samples from CW A, B and C during the experimental

ult in differences in the effluent quality and thus, control and monitoring is required. The treatment of winery wastewater with supplementation of N and P prior to discharge could result in eutrophication and/or toxicity being developed in the areas. Thus, mechanisms to enhance P and N removal would be necessary, based on increased substantially at the end of the experiment in CW C (ethanol/nutrients) (Figs. 2b and 3). In this context, the addition of external nutrients to CW systems is questionable and must be subject of further research.

The COD removal performance from the experimental CWs was evaluated taking into account ter (Fig.4a and 4b). As shown in fig.4a), both CW B (ethanol) and CW C (ethanol/nutrients) differed significantly from CW A (control), the COD remained constantly close to zero , the COD increased after the The effluent COD from CW B ranged from 730 to in the case of CW C, COD peaked at a value of 1076 mg L-1 on day 28 but then exhibited a decreasing trend. At the end of the experiment, the effluent values were similar ), reflecting close to 100% removal efficiency. The 5% to 10% greater than (Fig. 4b). These results demonstrate that the addition of nutrients

(10)

Figure 4 COD removal performance in terms of (a) effluent COD (CW A, B and C) (CW B and C) during the experimental period.

In general, COD removal efficiencies of around 90% from winery wastewater have been achieved in aerobic treatment systems such as activated sludge, aerobic sequencing batch react and membrane bioreactors (Torrijos and Moletta, 1997;

2002; Brucculeri et al., 2005;

processes such as anaerobic filters, up

fixed bed reactors have reported lower COD removal efficiencies (80% 2003; Yu et al., 2006; Ganesh

an anaerobic sequencing batch

Evolution of the bacterial community structure

The evolution of the bacterial community structure was assessed by DGGE fingerprinting and PCA analysis of the CW sediments. Figs. 5a and 5b

of the evolution of the bacterial community structure in the surface and the deep sediments respectively. In the surface sediment samples, 47.7

with PC1, PC2 and PC3 explaining respectively 20.1 5a). In the deep sediment samples, 41.1

and PC3 explaining respectively 16.1%, 14.4% and 10.6% of the variance (Fig. 5b

baseline communities in the superficial and deep sediments from both inlet and outlet samples showed that following a three month equilibration period, communities from all CWs were similar (Fig. 5a and 5b) (Ramond

allowed evaluation of subsequent community changes in response to amendment.

In the surface sediments of CW A (control) and CW C (ethanol/nutrients) the microbial communities evolved similarly throughout the experiment, (essentially along PC2 and PC3 (explaining 15.6% and 11.4% of the total variance, respectively), and were closely located on the 3D-PCA representation at each sampling time (Fig. 5a).

and final (day 49) samplings, the microbial community fingerprint

separated from those of CW A and CW C, and evolved essentially along PC1 and PC3 (explaining 20.1% and 11.4% of the variance respectively) (Fig. 5a). It was concluded that at the surface, amendment with ethanol resulted in a seve

structure and that supplementation with nutrients (N/P) ameliorated this effect. This occurred

COD removal performance in terms of (a) effluent COD (CW A, B and C) and (b) COD removal efficiency (CW B and C) during the experimental period.

In general, COD removal efficiencies of around 90% from winery wastewater have been achieved in aerobic treatment systems such as activated sludge, aerobic sequencing batch react

Torrijos and Moletta, 1997; Fumi et al., 1995;

Artiga et al., 2007). Several studies evaluating anaerobic treatment such as anaerobic filters, up-flow activated sludge blanket processes and anaerobic fixed bed reactors have reported lower COD removal efficiencies (80%

Ganesh et al., 2009), an exception being reported by Ruiz an anaerobic sequencing batch reactor which achieved 98% COD removal.

Evolution of the bacterial community structure

The evolution of the bacterial community structure was assessed by DGGE fingerprinting and PCA analysis of the CW sediments. Figs. 5a and 5b show the 3D-PCA graphical

of the evolution of the bacterial community structure in the surface and the deep sediments In the surface sediment samples, 47.7% of the spatial variations were explained, C3 explaining respectively 20.1%, 15.6 % and 11.4% of the variance (Fig. In the deep sediment samples, 41.1% of the spatial variations were explained, with PC1, PC2

ctively 16.1%, 14.4% and 10.6% of the variance (Fig. 5b

the superficial and deep sediments from both inlet and outlet samples showed that following a three month equilibration period, communities from all CWs were

Ramond et al., submitted) These similarities in communit evaluation of subsequent community changes in response to amendment.

In the surface sediments of CW A (control) and CW C (ethanol/nutrients) the microbial communities evolved similarly throughout the experiment, (essentially along PC2 and PC3 aining 15.6% and 11.4% of the total variance, respectively), and were closely located on the PCA representation at each sampling time (Fig. 5a). In contrast, for the mid

) samplings, the microbial community fingerprints of CW

separated from those of CW A and CW C, and evolved essentially along PC1 and PC3 (explaining 20.1% and 11.4% of the variance respectively) (Fig. 5a). It was concluded that at the surface, amendment with ethanol resulted in a severe modification of the microbial community structure and that supplementation with nutrients (N/P) ameliorated this effect. This occurred

and (b) COD removal efficiency

In general, COD removal efficiencies of around 90% from winery wastewater have been achieved in aerobic treatment systems such as activated sludge, aerobic sequencing batch reactors ., 1995; Andreottola et al., ). Several studies evaluating anaerobic treatment sludge blanket processes and anaerobic fixed bed reactors have reported lower COD removal efficiencies (80%-90%) (Keyser et al., , 2009), an exception being reported by Ruiz et al. (2002) on reactor which achieved 98% COD removal.

The evolution of the bacterial community structure was assessed by DGGE fingerprinting and PCA graphical representations of the evolution of the bacterial community structure in the surface and the deep sediments % of the spatial variations were explained, % of the variance (Fig. % of the spatial variations were explained, with PC1, PC2 ctively 16.1%, 14.4% and 10.6% of the variance (Fig. 5b). Clustering of the superficial and deep sediments from both inlet and outlet samples showed that following a three month equilibration period, communities from all CWs were community composition evaluation of subsequent community changes in response to amendment.

In the surface sediments of CW A (control) and CW C (ethanol/nutrients) the microbial communities evolved similarly throughout the experiment, (essentially along PC2 and PC3 aining 15.6% and 11.4% of the total variance, respectively), and were closely located on the In contrast, for the mid-term (day 28) CW B (ethanol) were well separated from those of CW A and CW C, and evolved essentially along PC1 and PC3 (explaining 20.1% and 11.4% of the variance respectively) (Fig. 5a). It was concluded that at the re modification of the microbial community structure and that supplementation with nutrients (N/P) ameliorated this effect. This occurred

(11)

independently from the nutrient load which changed during the course of the experiment in CW C (Table 1). Points representing the inlet and outlet surface samples of each individual CW remained close on the 3D-PCA graphical representation throughout the experiment (Fig. 5a), indicating that amendment induced minimal spatial modification in the community structure. In contrast, points representing the deep inlet and outlet sediment samples of each CW diverged on the PCA diagram (Fig. 5b), indicating spatial modifications of the microbial community structure in all CWs during the course of the experiment. At the inlet, by d

experiment), the community of CW A (control) was well separated along PC1/PC2 (30.7% of the variance) and PC3 (10.6% of the variance) from that of CW B (ethanol) and CW C (ethanol/nutrients) respectively (Fig. 5b). These results sugg

experiment, ethanol amendment, with or without nutrient supplementation, modified the microbial community structure

structure of CW C (ethanol/nutrients) was closer

These results suggest that at that stage, ethanol had impacted and modified the community structure, an effect that was no more evident when nutrients were added simultaneously with a COD/N/P ratio of 200/5/1 (Table 1). In the deep outlet sediments, the points representing CW A (control) and CW C (ethanol/nutrients) were very close on the 3D

experiment (Fig. 5b), indicating similar microbial community structures. In contrast, the poi representing CW B (ethanol) were distant at each sampling, and evolved strongly along PC1 between day 0 and day 28 (16.1% of variance), and along PC1, PC2 and PC3 between day 28 and day 49 (41.1% of variance). This demonstrated that, throughout the exp

sediments ethanol induced a modification of the microbial community structure that was not apparent when nutrients were added.

Figure 5 Three-dimensional principal component analysis (PCA) representing the evolution of the

communities in the (a) surface sediments and (b) deep sediments of the constructed wetlands. Capital letters close to the points designate the respective constructed wetland: “i” and “o” respectively indicating inlet and outlet samples. ; Initial sampling (day 0). ; Sampled

The aim of this section of the study was to evaluate the impact of on the microbial community structure of the CWs.

important role in the removal of organic matter, nitrogen and phosphorous (Truu

11

independently from the nutrient load which changed during the course of the experiment in CW senting the inlet and outlet surface samples of each individual CW PCA graphical representation throughout the experiment (Fig. 5a), indicating that amendment induced minimal spatial modification in the community structure.

trast, points representing the deep inlet and outlet sediment samples of each CW diverged on the PCA diagram (Fig. 5b), indicating spatial modifications of the microbial community structure in all CWs during the course of the experiment. At the inlet, by d

experiment), the community of CW A (control) was well separated along PC1/PC2 (30.7% of the variance) and PC3 (10.6% of the variance) from that of CW B (ethanol) and CW C (ethanol/nutrients) respectively (Fig. 5b). These results suggest that at this stage of the experiment, ethanol amendment, with or without nutrient supplementation, modified the microbial community structure However, by the end of the experiment (day 49), the community structure of CW C (ethanol/nutrients) was closer to that of CW A (control) than CW B (ethanol). These results suggest that at that stage, ethanol had impacted and modified the community structure, an effect that was no more evident when nutrients were added simultaneously with a (Table 1). In the deep outlet sediments, the points representing CW A (control) and CW C (ethanol/nutrients) were very close on the 3D-PCA diagram throughout the experiment (Fig. 5b), indicating similar microbial community structures. In contrast, the poi representing CW B (ethanol) were distant at each sampling, and evolved strongly along PC1 between day 0 and day 28 (16.1% of variance), and along PC1, PC2 and PC3 between day 28 and day 49 (41.1% of variance). This demonstrated that, throughout the experiment, in the deep outlet sediments ethanol induced a modification of the microbial community structure that was not apparent when nutrients were added.

dimensional principal component analysis (PCA) representing the evolution of the

communities in the (a) surface sediments and (b) deep sediments of the constructed wetlands. Capital letters close to the points designate the respective constructed wetland: “i” and “o” respectively indicating inlet and outlet samples.

Sampled on day 28. ; Final sampling (day 49).

aim of this section of the study was to evaluate the impact of nutrients (N and P)

on the microbial community structure of the CWs. In CWs, microbial communities play a important role in the removal of organic matter, nitrogen and phosphorous (Truu

independently from the nutrient load which changed during the course of the experiment in CW senting the inlet and outlet surface samples of each individual CW PCA graphical representation throughout the experiment (Fig. 5a), indicating that amendment induced minimal spatial modification in the community structure.

trast, points representing the deep inlet and outlet sediment samples of each CW diverged on the PCA diagram (Fig. 5b), indicating spatial modifications of the microbial community structure in all CWs during the course of the experiment. At the inlet, by day 28 (mid-term of the experiment), the community of CW A (control) was well separated along PC1/PC2 (30.7% of the variance) and PC3 (10.6% of the variance) from that of CW B (ethanol) and CW C est that at this stage of the experiment, ethanol amendment, with or without nutrient supplementation, modified the However, by the end of the experiment (day 49), the community to that of CW A (control) than CW B (ethanol). These results suggest that at that stage, ethanol had impacted and modified the community structure, an effect that was no more evident when nutrients were added simultaneously with a (Table 1). In the deep outlet sediments, the points representing CW A PCA diagram throughout the experiment (Fig. 5b), indicating similar microbial community structures. In contrast, the points representing CW B (ethanol) were distant at each sampling, and evolved strongly along PC1 between day 0 and day 28 (16.1% of variance), and along PC1, PC2 and PC3 between day 28 and eriment, in the deep outlet sediments ethanol induced a modification of the microbial community structure that was not

dimensional principal component analysis (PCA) representing the evolution of the microbial communities in the (a) surface sediments and (b) deep sediments of the constructed wetlands. Capital letters close to the points designate the respective constructed wetland: “i” and “o” respectively indicating inlet and outlet samples.

nutrients (N and P) and ethanol In CWs, microbial communities play an important role in the removal of organic matter, nitrogen and phosphorous (Truu et al., 2009;

(12)

Glick, 2010). Microbial activities induce biogeochemical transformations in natural, managed and engineered ecosystems, especially in response to contamination. Additionally, organic pollutants such as ethanol have been proven to induce structural changes on microbial communities (Feris et al., 2008). However, in this study the addition of N and P with a COD/N/P ratio of 300/5/1 helped to maintain the microbial community structure in CW C (ethanol/nutrients) at all locations and sampling times except in the deep inlet sediments. These results thus confirmed that the addition of inorganic nutrients at an appropriate ratio to nutrient deficient biological wastewater treatment processes enhances the COD removal efficiency which could be linked to the maintenance of a stable microbial community structure.

CONCLUSIONS

From an economic perspective, aerobic treatment systems present disadvantages due to the high costs of aeration and sludge processing. Anaerobic processes are less costly, but generally insufficient to reach the effluent quality required for discharge in surface waters (Brito et al., 2007). Conversely, the variation in the seasonal flow and pollutant concentrations complicate treatment by traditional aerobic systems such as activated sludge (Chudoba and Pujol, 1996). Thus, CWs present a convenient, eco-friendly, cost-effective alternative method for the treatment of winery wastewaters in countries, such as South Africa, where land is available and the climate provides optimal environmental conditions. These systems are not disturbed by seasonal fluctuations and might adapt to the size of the local sources of winery wastewater (Masi et al., 2002; Grismer et al., 2003).

This study showed that the COD removal (ethanol degradation) efficiency increased with the addition of nutrients (nitrogen and phosphorus) in pilot scale CWs. Additionally, efficient NO3- -N degradation was observed in the nutrient amended CW. This could be explained by the presence of ethanol as a carbon source (electron donor) and nitrate (electron acceptor) enhancing the occurrence of dissimilatory processes (denitrification / dissimilatory nitrate reduction process). However, the nitrate reductase activity measurements presented analytical difficulties and couldn’t be used as an index of the performance of CW C. Thus, an optimization of the method is suggested. Effluent TN and TP concentrations increased at the end of the experiment in the nutrient amended CW C, and further research is needed in order to enhance the degradation of these compounds. Finally, the analysis of the bacterial community structure by means of DGGE showed that ethanol amendment increased the variability of the bacterial community structure. However, the addition of N and P together with ethanol countered this effect, resulting in a more stable bacterial consortium, similar to that of the control CW.

ACKNOWLEDGEMENTS

This work was funded by the Water Research Commission of South Africa. The authors of this paper would also like to acknowledge the work and support of the student Alaric Prins during the experimental phase of this study.

(13)

13

REFERENCES

Abdelmagid, H.M. and Tabatabai, M.A. (1987). Nitrate reductase activity of soils. Soil Biology and Biochemistry 19, 421-427.

Andreottola, G., Foladori, P. and Ziglio, G. (2009). Biological treatment of winery wastewater: an overview. Wat Sci Technol, 60(5), 1117-1125.

Andreottola, G., Foladori, P., Ragazzi, M., and Villa, R. (2002). Treatment of winery wastewater in a sequencing batch biofilm reactor. Wat Sci Technol, 45, 347-354.

Artiga, P., Carballa, M., Garrido, J.M. and Mendez, R. (2007). Treatment of winery wastewaters in a membrane submerged bioreactor. Wat Sci Technol 56, 63-69.

Bonin, P., Omnes, P. and Chalamet, A. (1998). Simultaneous occurrence of denitrification and nitrate ammonification in sediments of the French Mediterranean Coast.Hydrobiologia, 389, 169-182.

Bories, A., Sire, Y. and Colin, T. (2005). Odorous compounds treatment of winery and distillery effluents during natural evaporation in ponds. Wat Sci Technol 51 (1) 129-136.

Brito, A. G., Peixoto, J., Oliveira, J. M., Oliveira, J. O., Costa, C., Nogueira R. and Rodrigues, A. (2007). Brewery and winery wastewater treatment: some focal points of design and operation”, In: Utilization of By-Products and Treatment of Waste in the Food Industry.V. Oreopoulou and R. Winfred (Eds.), ISEKI Series, Vol. 3, Springer, NewYork.

Brucculeri, M., Bolzonella, D., Battistoni, P., and Cecchi, F. (2005). Treatment of mixed municipal and winery wastewaters in a conventional activated sludge process: a case study. Wat Sci Technol, 51, 89-98.

Burton, S., Sheridan, C., Law-Brown, J. Le Roes, M., Cowan, D., Rohr, L. and Mashaphu, N. (2007). Integrated research for use in constructed wetlands for treatment of winery wastewater, Report 1544/1/07, Water Research Commission, South Africa.

Chudoba, P. and Pujol, R. (1996). Activated sludge plant facing grape harvest period – a case study, Wat Sci Technol, 34(11), 25-32.

Deiglmayr, K. (2006). Community structure and activity of nitrate-reducing microorganisms in soils under global climate change. PhD thesis, Institute for soil science, University of Hohenheim, Germany.

Eusébio, A., Petruccioli, M., Lageiro, M., Federici, F. and Duarte, J. C. (2004). Microbial characterisation of activated sludge in jet-loop bioreactors treating winery wastewaters, J. Ind. Microbiol. Biotechnol. 31 29-34.

Fazzolari, E., Nicolardot, B. and Germon, J.C. (1998). Simultaneous effects of increasing levels of glucose and oxygen partial pressures on denitrification and dissimilatory nitrate reduction to ammonium in repacked soil cores. Eur. J. Soil Biol. 34(1), 47-52.

Feris, K., Mackay, D., De Sieyes, N., Chakraborty, I., Einarson, M., Hristova, K. and Scow, K. (2008) Effect of ethanol on microbial community structure and function during natural attenuation of benzene, toluene, and o-xylene in a sulfate-reducing aquifer. Environ Sci Technol, 42, 2289-94.

Fumi, MD., Parodi, G., Parodi, E., Silva, A., and Marchetti, R. (1995). Optimization of long-term activated sludge treatment of winery wastewater. Bioresour Technol 52, 45-51.

Ganesh, R., Rajagopal, R., Thanikal, J.V., R.A., Ramanujam., and Torrijos, M. (2009). Anaerobic treatment of winery wastewater in fixed bed reactor. Bioprocess Biosyst Eng.33, 619-628.

(14)

Grismer, M.E., Carr, M.A., Shepherd, H.L. (2003). Evaluation of constructed wetland treatment performance for winery wastewater. Water Environ. Res. 75, 412-421.

Howarth, R.W. (1988). Nutrient limitation of net primary production in marine ecosystems. Ann. Rev. Ecol. 19, 89-110.

Kandeler, E. (1996). Nitrate reductase activity. In: Methods in soil biology. Schinner F., Öhlinger, R., Kandeler, E. and Margesin, R., (eds), pp. 176-179. Springer-Verlag, Berlin Heidelberg.

Kelso, B. H. L., Smith, R.V. and Laughlin, R.J. (1999) Effects of carbon substrates on nitrite accumulation in freshwater sediments. Appl. Environ. Microbiol. 65(1), 61-66.

Kelso, B. H. L., Smith, R.V., Laughlin, R.J. and Lennox, S.D. (1997). Dissimilatory nitrate reduction in anaerobic sediments leading to river nitrite accumulation. Appl. Environ. Microbiol. 63, 4679-4685.

Keyser, M., Witthuhn, R.C., Ronquest, L.C. and Britz, T.J. (2003). Treatment of winery effluent with upflow anaerobic sludge blanket (UASB)–granular sludges enriched with Enterobacter sakazakii.Biotechnol. Lett. 22, 1893-8.

Malandra, L., Wolfaardt, G., Zietsman, A. and Viljoen-Bloom, M. (2003). Microbiology of a biological contactor for winery wastewater treatment. Water Research 37, 4125-4134.

Masi, F., Conte, G., Martinuzzi, N. and Pucci, B. (2002). Winery high organic content wastewaters treated by constructed wetlands in Mediterranean climate. In: Proceedings of the 8th International Conference on Wetland Systems for Water Pollution Control. University of Dar-es-Salaam, Tanzania and IWA, pp. 274–282 .

Masscheleyn, P.H., Pardue, J.H., DeLaune, R.D., and Patrick Jr. W.H. (1992). Phosphorus release and assimilatory capacity of two lower Mississippi valley freshwater wetland soils. Journal of the American Water Resources Association, 28(4), 763-773.

Mena, J., Gómez, R., Villaseñor, J. and de Lucas, A. (2009). Influence of polyphenols on low-loaded synthetic winery wastewater constructed wetland treatment with different plant species. Canadian Journal of Civil Engineering, 36(4), 690-700.

Metcalf and Eddy. (2003). Wastewater Engineering Treatment and Reuse. Fourth Edition, Mc Graw Hill, New York.

Muyzer, G., de Waal, E.C. and Uitterlinden, A.G. (1993) Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl. Environ. Microbiol. 59:695-700.

Ramond, J-B., Welz, P.J., Cowan, D.A. and Burton, S.G. Microbial community structure, a key parameter in monitoring the development and stability of constructed wetland mesocosms. Submitted to Ecological Engineering.

Richardson, C. J. and Craft, C. B. (1993). Effective phosphorus retention in wetlands: Fact or fiction? In: Constructed Wetlands for Water Quality Improvement, G. A. Moshiri (ed.), Lewis Publishers, Boca Raton.

Ruiz, C., Torrijos, M., Sousbie, P., Lebrato-Martinez, J., Moletta, R. and Delgenes, J.P. (2002). Treatment of winery wastewater by anaerobic sequencing batch reactor. Wat Sci Technol, 45, 219-224.

Shepherd, H.L., Grismer, M.E. and Tchobanoglous, G. (2001). Treatment of High-Strength Winery Wastewater Using a Subsurface-Flow Constructed Wetland. Water Environment Research, 73, 394-403

Sheridan, C. (2007). Constructed Wetlands for the Primary Treatment of Winery Effluent. PhD thesis, University of Cape Town, South Africa.

(15)

15

Simon, J. (2002). Enzymology and bioenergetics or respiratory nitrate ammonification. Fems Microbiol Rev, 26, 285-309.

Smith, V.H., Tilman, G.D. and Nekola, J.C. (1999). Eutrophication: impacts of excess nutrient inputs on freshwater, marine, and terrestrial ecosystems. Environmental Pollution, 100, 179-196.

Song, K-Y., Zoh, K-D., and Kang, H. (2007). Release of phosphate in a wetland by changes in hydrological regime. Sci. Total Environ. 380, 13-18.

Stevens, R.J., Laughlin, R.J. and Malone, J.P. (1998). Soil pH affects the processes reducing nitrate to nitrous oxide and di-nitrogen. Soil Biol. Biochem. 30(8/9), 1119-1126.

Tiedje, J. M. (1988). Ecology of denitrification and dissimilatory nitrate reduction to ammonium. In: Biology of Anaerobic Microorganisms, A. J. B. Zehnder (ed), pp. 179-244. Wiley, New York.

Torrijos, M., and Moletta, R. (1997). Winery wastewater depollution by sequencing batch reactor. Wat Sci Technol, 35, 249-257.

Truu, M., Johanson, J., Truu, J. (2009). Microbial biomass, activity and community composition in constructed wetlands. Sci. Total Environ. 407, 3958-3971.

US EPA. (1986). Quality criteria for water. Washington, DC. EPA 440/5-86-001.

Vymazal, J. (2009). The use constructed wetlands with horizontal sub-surface flow for various types of wastewater. Ecological Engineering, 35, 1-17.

Wang, Y., Zhang, Z. and Ruan, J. (1996). A proposal to transfer Microbispora bispora (Lechevalier 1965) to a new genus, Thermobispora gen. nov., as Thermobispora bispora comb. nov. Int J Syst Bacteriol 46: 933-938.

Yu, H.Q., Zhao, Q.B. and Tang, Y. (2006). Anaerobic treatment of winery wastewater using laboratory-scale multi- and single-fed filters at ambient temperatures. Process Biochem, 41, 2477–2481.

Zumft, W. G. (1997). Cell biology and molecular basis of denitrification. Microbiol Molecular Biol Rev, 61, 533-616.

Figure

Figure 1 Set-up and mode of operation of the experimental constructed we
Table  1  Bi-weekly  feeding  regimes  and  artificial  wastewater  composition  for  CW  A,  B  and  C  during  the  experimental period
Figure  2  (a)  Nitrate  reductase  activity  in  superficial  sediments,  and  effluent  concentrations  of  (b)  Total  nitrogen  (TN), (c) N0₃⁻-N and (d) NH₄⁺-N from CWs A
Figure 3 Total Phosphorous (TP) concentration in effluent samples from CW A, B and C during the experimental  period
+3

References

Related documents

46 Konkreta exempel skulle kunna vara främjandeinsatser för affärsänglar/affärsängelnätverk, skapa arenor där aktörer från utbuds- och efterfrågesidan kan mötas eller

Coad (2007) presenterar resultat som indikerar att små företag inom tillverkningsindustrin i Frankrike generellt kännetecknas av att tillväxten är negativt korrelerad över

Sugarcane has been harvested and sugar produced in Brazil for hundreds of years. The cane cutters, originally slaves, have always had poor working conditions and

According to the study, nitrogen removal in FWS wetlands is strongly affected by temperature just as the previous hypothesis, and N removal rate in FWS constructed wetlands is higher

Industrial Emissions Directive, supplemented by horizontal legislation (e.g., Framework Directives on Waste and Water, Emissions Trading System, etc) and guidance on operating

The depth concentration profile (Figure 4) was analyzed for the highest average TN removal value (74%) achieved in optimized SAZ (the carbon columns at 450 mm SAZ level).. The

Like in the first experiment, 24 g fresh weight of the plant litter material have been used in each of the 20 one-liter beakers, but here exactly 600 ml of inlet water were used

Det ligger självfallet i verksamhetsutövarens eget intresse att så långt som möjligt begränsa skadorna på den egna verksamheten. Så gott som alla olyckor på en deponi utgörs