Monitoring performance of constructed wetlands in California

A Technical Guide
for Wine Producers

RECENT ARTICLES | WYNBOER HOME

Monitoring performance of constructed wetlands in California

Reckson Mulidzi

by Reckson Mulidzi, ARC Infruitec-Nietvoorbij, Stellenbosch
Abstract
World-wide, the use of constructed wetlands is considered to be the cheapest alternative for the purification of cellar effluent. During a study tour by the author to California, five different wineries that use subsurface constructed wetlands were visited. Objectives of the study tour included an investigation into the practical implementation of constructed wetlands, as well as obtaining the latest research results regarding winery wastewater treatment. Effluent samples were taken at the different wineries and results showed that wineries with a pre-treatment system such as aerators prior to the wetlands, performed better than those without a pre-treatment system. The benefits of using constructed wetlands for winery wastewater treatment include low cost and maintenance, but the ultimate performance depends on effective management.
Introduction
World-wide, the use of constructed wetlands is considered to be the cheapest alternative for the purification of cellar wastewater. Constructed wetlands offer a relatively new approach for the treatment of winery wastewater and have been explored for such applications in only the past ten years. The objectives of the visit were:

To investigate the practical implementation of constructed wetlands.
To learn relevant technologies related to winery effluent disposal and management.
To obtain the latest research results about constructed wetlands directly from the researcher/consultant.
To investigate the possibility for collaborative research in treatment of winery effluent.

Fig 1: Combination of plants in a constructed wetland at Benziger Winery.

During a visit by the author to California, USA, the author was hosted by Heather Shepherd (senior consultant at Wallace Swanson International). Five different wineries that use subsurface constructed wetlands were visited. The performance of wetlands was monitored by taking effluent samples at selected sampling points.
Materials and methods
Effluent sampling and analysis
Effluent samples were collected from five different wineries, namely Benziger, Imagery, Villa Toscano, Chateau Felice and Fetzer over a two week period. The effluent was collected from different sampling points before and after it enters the constructed wetland. The samples were analysed for pH, total dissolved solids (TDS), total suspended solids (TSS), nitrate and chemical oxygen demand (COD). A portable Accumet AP 63 meter was used to analyse the effluent pH, a Hach conductivity/TDS meter to analyse the effluent TDS and a Hach DR/2000 direct reading spectrophotometer for TSS, nitrate and COD analysis.
Description of wetlands
Benziger Winery
This winery crushes 1200 tons of grapes per annum. A constructed subsurface wetland is used to treat the wastewater. The wetland is 40 metres long, 8 metres wide and 1 metre deep. The wetland has been filled with doleritic gravel with 35% porosity. Cattails, bulrush and common reeds have been planted together (Figure 1). Combination of plants in a wetland system helps to improve the performance because some plant species are able to resist higher COD or Na levels than others. The designed retention time of the wetland is four days. During the sampling period, the retention time was 3.3 days, i.e. 6546 litres per day. The effluent from the winery went into an aeration pond where pre-treatment took place before entering the wetland system (Figure 2). After the wastewater has been aerated, it flows into the constructed wetland through a floating pipe.

Fig 2: Aerators at Benziger settling pond, where effluent is treated before entering the wetland.

Imagery Winery
This winery crushes 2000 tons of grapes per year. Two subsurface constructed wetlands with an aerated pond (Figure 3) upfront are used to treat the wastewater. Both wetlands are 19 metres long, 8 metres wide and 1 metre deep. They are filled with dolerite gravel (5 mm diameter) with 35 % porosity. More than one type of plant was planted. The wastewater is pumped from the pond to the wetland. After being treated in the first wetland it flows to the second wetland where final treatment takes place.
Chateau Felice Winery
This winery is very small, crushing only 160 tons of grapes per year. It uses a subsurface constructed wetland to treat wastewater. The wetland is 17 metres long, 4 metres wide and 0.9 metres deep. The wetland is filled with gravel and established with plants such as bulrush, cattails and others. The design retention time for this wetland was 10.5 days, assuming that 300 tons of grapes will be crushed.
Fetzer Winery
This winery uses a constructed wetland with two aerated ponds upfront to treat their wastewater. The wetland is 90 metres long, 51 metres wide and 1 metre deep. Bulrush and cattails plants were established. A designed retention time of 10 days is used. The sampling at this winery was done only once, therefore the analysis cannot give an average performance of the wetland.
Results and discussion
Benziger Winery
Chemical oxygen demand (COD)
The aerators placed on the pre-treatment pond are playing an important role in the treatment of effluent (Figure 2). The constructed wetlands in this winery are performing very well as a secondary treatment system (Table 1). The average COD value of effluent at the outlet from the wetland was 106 mg/l, which is clean enough to be used for irrigation.

Fig 3: Aerators that serve as primary treatment at Imagery Winery.

During the sampling period, the average COD of effluent at the inlet to the aerated pond was 4433 mg/l resulting in the average inlet to the wetland to be 631 mg/l. This clearly shows that the aerated pond had already treated the wastewater, with the wetland doing the polishing work.
TDS, TSS, Nitrate and pH
Although constructed wetlands are known to be ineffective for the treatment of TDS, the wetland in this winery had an average lower TDS at the outlet compared to that at the inlet (data not shown). The good results of this system could be probably attributed to effective management. It must also be emphasised that compared to South Africa most, if not all, wineries in California does not use caustic soda. This results in wastewater with low TDS and sodium values.
The pre-treatment system (Figure 2) and wetlands (Figure 1) also reduced the total suspended solids. The water at the outlet was very clean with an average TSS value of 12 mg/l. Although wetland systems are not known to be effective in removing nitrates from the effluent, the treatment system in this winery was able to reduce the average nitrate level from 4.2 mg/l to 0.7 mg/l (data not shown). The constructed wetland also improves the colour of wastewater (Figure 6).
Imagery Winery
COD, TDS, TSS, Nitrate and pH
The overall average COD removal in this winery was ineffective, even though it was not yet peak season. The possible reason for the poor removal of COD can be attributed to poor management. This was evident during the sampling period when the holes in the step feeding pipes were full of sludge due to lack of cleaning. The outlet TDS, TSS and nitrate was higher than the inlet on some days. As mentioned previously, this is the result of a neglected wetland system. In general, wetlands are very sensitive treatment systems, only performing well when looked after.

Fig 4: A waterlogged wetland at Villa Toscano Winery due to excessive influx of effluent.

Villa Toscano Winery
COD
Average COD removal in this winery was 55% (data not shown) during the sampling period. As there is no pre-treatment system, constructed wetland is working as a primary treatment. The wetland plants at this winery were still very young during the sampling period. Due to poor management, the wastewater applied was also more than the designed retention time (Figure 4). Consequently, the system at some stage received more than it could handle. Under these conditions, the system still managed to remove 55% COD. If a pre-treatment system can be put in place in combination with effective management, the wetland will probably perform well.
Chateau Felice Winery
Average COD removal by the wetland at this winery was 25% (data not shown). As the winery is very small, effluent production is low and the wetland mostly operates on clean water. The inclusion of floating plants in the wetland increased its aesthetic value (Figure 5).
Fetzer Winery
As this winery was remote, only one sampling was possible. Results from this winery showed good performance with 71% COD removal (data not shown).
Effluent composition: USA (California) vs. South Africa
The composition of winery wastewater from cellars in South Africa and USA is similar, except for high sodium in South African effluent. The reason for low sodium content in the USA effluent is absence of caustic soda (sodium hydroxide) in the cleaning process. Potassium hydroxide is used instead, which is beneficial for wastewater treatment by constructed wetlands, as potassium is an essential element needed by plants.

Fig 5: Constructed wetland with a combination of plants at Chateau Felice Winery, indicating their aesthetic value.

The Integrated Production of Wine (IPW) system in South Africa now also discourages the usage of caustic soda and encourages cellars to use potassium hydroxide instead. Furthermore the South African wine industry as a whole is trying to discourage the use of all sodium related detergents in the cellar. This trend is encouraging in terms of the whole concept of environmental management as potassium is less harmful when compared to sodium.
Wetland plants
The most commonly used wetland plants throughout the world include cattails (Scirpus, spp.), bulrush (typha, spp.) and reeds (phragmites, spp.). Other plants that can be used in constructed wetlands are currently being investigated at ARC Infruitec-Nietvoorbij. Floating species, such as duckweed, may be added to the free water surface of constructed wetlands to control algae growth. When selecting plants for constructed wetlands, pH and temperature tolerances for the plants need to be considered. Shepherd (2002) presented a partial list to assist in plant selection (Table 2).
Reddy and De Busk (1987) listed the following characteristics as being desirable for plants in wetland treatment systems:

Adaptability to local climate.
High photosynthetic rates and oxygen transport capability.
Tolerance to high concentrations of nutrients and pollutants.
Ability to assimilate pollutants.
Tolerance to adverse climatic conditions.
Resistance to pests and diseases.
Deep root penetration.
In some cases flowering species like Arum Lily (Varklelie) provide a more aesthetic look to the system. However, during design, care and caution must be used as bulrush, cattails and reeds are very invasive and frequently overgrow any other species present.

Fig 6: Liquid showing the quality of wastewater after being treated by wetlands at Benziger Winery. On the left-hand side is wastewater from the winery to the pond and in the middle is wastewater from the pond to the wetland. The wastewater on the right-hand side was taken at the outlet of the wetland (COD only 106 mg/l and TSS = 12 mg/l).

Monitoring and maintenance of constructed wetland
A wetland is a biological filter and like all filters, requires some occasional cleaning and maintenance. A big advantage of wetland systems is that less maintenance is needed than with anaerobic digesters. Routine removal of excess plant material and captured sediments will allow the wetland to continue to function. Regular harvesting of the crop can prevent the system from becoming clogged.
Conclusions
The performance of constructed wetlands depends on the management at each winery. Results also showed that wineries with a pre-treatment system, such as prior the wetlands, performed better than those without a pre-treatment system, e.g. Benziger Winery vs. Villa Toscano Winery. Land, funds and effective management incorporated with some creativity and flexibility is needed to construct a wetland for wastewater treatment. While wetlands do not need intensive management, potential problems (e.g. plants that begin to yellow may indicate that a high organic load has gone through the system or sulphur levels are too high) must be identified and addressed immediately. Should such problems arise, alternatives must be available (e.g. re-circulation or aeration of wastewater). The benefits of using constructed wetlands for winery wastewater purification include low cost, aesthetic wastewater treatment system and low maintenance. However, constructed wetlands should be seen as a secondary treatment system. The removal of solids is therefore essential before effluent enters the wetland system.
Recommendations
When constructed wetlands are considered for the treatment of winery wastewater, the following recommendations should be considered during the designing stage.

A compulsory pre-treatment system for solids removal, as solids contains more than 40% of the COD load. The solids will also cause clogging of the system.
Rainwater should be included during the design phase. The system should be designed so that its capacity is bigger than required as rainwater will decrease the retention period by pushing untreated effluent to the outlet.
A 10% safety factor should be incorporated into a design. This will allow for unknowns such as fluctuations in effluent composition. It also allows for the removal of plants from parts of the wetland without affecting overall performance.
A combination of more than one type of plant is essential as various types of plants tolerate wastewater differently and their ability to remove nutrients from the effluent also differs.

Future research must address the treatment of conductivity and the need for plants that are able to operate under low temperatures. This is especially important where effluent is produced during winter months. A workshop on the design, construction and management of constructed wetlands should be a priority.
For further information, contact the author at tel. (021) 809-3014, fax (021) 809-3002 and e-mail MulidziR@arc.agric.za.

Table 1. COD values (mg/l) at Benziger Winery

Table 2. Plants commonly used in constructed wetlands for wastewater treatment (Shepherd, 2002)

References
Reddy, K.R. & De Busk, R.A.,1987. State of the art utilization of aquatic plants in water pollution control. Water Science Technology 19: 61-79.
Shepherd, H.L., 2002. Use of constructed wetlands in treating winery process wastewater: Design issues and field investigations. International Workshop and Seminar on Effects and Treatment of Cellar and Distillery Effluent, Stellenbosch, South Africa, 23-24 April.

Wynboer is incorporated in WineLand, magazine of the SA wine producers.
Subscribe to WineLand

Visit our sister sites:

South African wine farmers' representative organisation

Facts, figures, contact details and much more in the 2009/10 Directory