Producing Native - Ornamental Wetland Plants in Constructed Wetlands Designed to Reduce Pollution from Agricultural Sources

Final Report for LNE98-100

Project Type: Research and Education
Funds awarded in 1998: $72,840.00
Projected End Date: 12/31/2001
Matching Non-Federal Funds: $60,663.00
Region: Northeast
State: Rhode Island
Project Leader:
Brian Maynard
University of Rhode Island
Expand All

Project Information


A demonstration project was designed and implemented to provide farmers with an economic incentive for the treatment of non-point source pollution, through the production and sale of ornamental or native wetland plants grown in a wetland biofilter. Wetland plants were grown, harvested and sold, from constructed greenhouse wetlands which also served to cleanse nutrient-laden agricultural waters. The demonstration wetland has been in place for two years and along with economic data derived from two growing seasons, will continue to serve as a demonstration model for ornamental treatment-production wetland biofiltration.

Project Objectives:

To demonstrate an economical solution to treating nursery runoff by growing, harvesting, and selling native and ornamental wetland plants produced in a constructed wetland treating runoff from a commercial nursery.

To evaluate the economic impact of converting nursery production space into treatment wetland production space.

To research the possibility of enclosing treatment wetlands in passively heated polyhouses to facilitate year around treatment of agricultural runoff.

To distribute the results of this study to farmers and other agricultural businesses or professionals interested in treating runoff with created wetlands via an extension/outreach program consisting of seminars, workshops, field trips, slide shows and fact sheets.


Click linked name(s) to expand/collapse or show everyone's info
  • Jose Amador
  • Thomas Holt
  • Dixon Hoogendoorn
  • William Johnson


Materials and methods:

A demonstration wetland was constructed at a commercial nursery in Middletown, Rhode Island. The wetland consisted of a single wetland cell (4m wide x 30m long x 0.2 m deep) filled with gravel (2 – 6 mm). Ten ornamental plant taxa were grown in this wetland and supplied with a commercial fertilizer for the purpose of evaluating growth and N and P recovery. Test species included Canna x generalis 'Aflame', Glyceria maxima 'Variegata' (variegated manna grass), Phalaris arundinacea 'Feecy', Phalaris arundinacea 'Picta' (variegated canary reed grass), Phragmites australis 'Variegata' (variegated common reed), Spartina pectinata 'Aureo-marginata' (variegated prairie cord grass), Colocasia esculenta (green taro), Iris pseudacorus (yellow flag), Pontederia cordata 'Alba' (pickerel weed) and Typha latifolia (broad-leaf cattail).

Cattail was included as a test species because of its wide use in constructed wetlands throughout the country and as a point of comparison with the ornamental taxa. Biomass production, tissue N and P concentrations, N and P recovery, and number of divisions produced were determined. The results of this investigation were compared with findings from other studies using the wild type plants from which these ornamental taxa were selected or bred. After the first and second years of operation a portion of the plants were harvested from the wetland and sold through the nursery.

We researched the year around treatment of agricultural runoff by enclosing treatment wetland mesocosms in three polyhouses. Each polyhouse was maintained at different minimum winter temperatures (10?C, 4?C, or -10?C). Plant growth and nutrient removal from solution was compared for three species at five different times of the year (August, November, February, April, and July) for each of the polyhouses.

Research results and discussion:

The demonstration wetland incorporated on average 87 g N per m2 and 11 g P per m2 into plant biomass. The revenue generated from plant sales from the first year of operation ($3,500) covered the cost of the construction of the wetland. Second year production was reduced because of lower demand for ornamental grass taxa, but wetland plant production still generated more than $2,000 in revenue.

By enclosing wetlands in a polyhouse and supplying them with minimal heat, we were able to increase nitrogen removal 100% (80 mg/l vs. 40 mg/l) at 10?C, and by 50% (60 mg/l vs. 40 mg/l) at 4?C. Phosphorus removal was increased 160% (13 mg/l vs. 5 mg/l) at 10?C, and by 120% (11 mg/l vs. 5 mg/l) at 4?C.

Participation Summary


Educational approach:

This project resulted in the production of three thesis chapters (Holt, 2001, University of Rhode Island Doctoral Thesis), as well as presentations and published papers at several national level meetings. An overview of the work was published in Yankee Nursery Quarterly, a major nursery extension publication in the northeast. Tours of the demonstration site were presented in conjunction with a Rhode Island Nursery and Landscape summer meeting, and a field day of the New England Nursery Association. It is estimated that over 1,000 nursery and green industry professionals viewed the project over the first two years.

No milestones

Project Outcomes

Impacts of Results/Outcomes

The demonstration wetland attracted a great deal of positive attention from the farmer and green industry professionals/customers. The plants in the wetland were in bloom throughout the summer while treating nutrient laden water. Biomass increase was tremendous. Approximately 400 nursery growers, landscapers, and wetland mitigation experts viewed the wetland during scheduled tours coinciding with the Rhode Island Nursery and Landscape Association summer meeting. The wetland was used to evaluate the growth potential of 10 ornamental plant species. N and P recovery, divisions produced, and N and P recovery per division were determined for each planting. N and P recovery of the taxa ranged between 49 to 125 g N per m2 and 6 to 16 g P per m2. Canna x ‘Aflame’ and Phragmites communis ‘Variegata’ removed the most nutrients from the wetland. Phragmites, and Phalaris arundinacea cultivars ‘Feecy’ and ‘Picta’, produced the most divisions on an area basis (200 to 240 divisions per m2). Iris pseudacorus and Pontederia cordata ‘Alba’ produced the fewest divisions (100 to 140 divisions per m2).

We also researched the possibility of enclosing treatment wetlands in passively or minimally heated polyhouses to facilitate year around treatment of agricultural runoff. N and P recovery of Iris pseudacorus, Schoenoplectus validus, and Typha latifolia were investigated in three polyhouses, maintained at different winter temperatures, to identify temperatures and time periods when wetland plants would act as sinks or sources of N and P. Both Iris and Schoenoplectus acted as sinks for N and P throughout the experiment in the warmest treatment, while Typha was a source of N and P during the winter regardless of the temperature treatment. We identified old shoots as the plant component that released the most nutrients during the winter, and new shoots as the plant component that was the largest sink for nutrients in the spring and summer.

Environmental problems associated with greenhouse and nursery runoff continue to gain attention and constructed wetlands have emerged as effective, low-cost methods of water treatment for mitigating agricultural pollutants from nursery runoff. The expense of using such systems could be offset by growing ornamental aquatic plants in treatment wetlands which, in turn, could be harvested and sold. The potential might exist to incorporate traditional nursery crops into treatment wetlands. For example, several taxa of Canna, Iris and ornamental grasses have been used in treatment wetlands for years. Plants harvested from treatment wetlands could be sold bare root, or potted into containers using traditional potting mixes. Wetlands could be reestablished by the replanting of divisions back into the wetland. Results from this study showed that all of the taxa evaluated were able to grow in a flooded gravel-based wetland.

Economic Analysis

The treatment wetland generated enough revenue through plant sales to cover the cost of the construction of the wetland after a single season of operation. The major expenses of the wetland included the liner ($600), gravel ($600), and labor (80 hrs. x $10 hr=$800). Wetland plants were purchased the previous year ($300) from local nurseries and propagated over the winter for the initial planting of the wetland. The total cost of the constructed wetland was $2,300.

Ten percent of the wetland was harvested 9 months after the initial planting. These plants were potted into 6 in. and 10 in. containers and were priced at $3.50 and $5.50, respectively. The value of the potted plants was $4,150. Sixty percent of the plants were sold for an income of $2,490. If 75% of the wetland was harvested every spring the potential revenue would be $31,125. The remaining 25% of the plants could be used to replanted the wetland.

The farmer previously used the polyhouse to grow containerized landscape plants. The polyhouse previously held 1,200 containers with a value of $10 per container. These containers required a two year growth period prior to sale, so that the farmer’s revenue potential was $12,000 every two years, or $6,000 annually. This could be matched by selling only 15% of the plant material produced in the treatment wetland. The cost of labor involved in maintaining the containerized polyhouse was similar to maintaining the wetland polyhouse.

General requirements for plants suitable for constructed wetlands and plant production include tolerance of pollutants and waterlogged conditions, quick propagation rates, rapid establishment, spread, growth, and a high pollutant removal capacity. All of the test plants in this study showed an ability to satisfy these requirements. Typha latifolia had the largest biomass and N recovery. While Canna and Typha had the largest P recovery. However, in constructing a wetland for both economic recovery as well as nutrient recovery, it is important to grow a wide variety of plants that have economic value. For example, Typha produced 162 divisions.m-2. If a 1000 m2 wetland was constructed with only Typha, could a nursery sell 162,000 divisions of a single species? It would have to diversify. Nurseries have maximized sales by growing a wide assortment of material. We feel this is also the best approach in a constructed wetland. Perhaps a large portion of the wetland could be planted with a species that removes large amounts of nutrients and a smaller portion of the wetland could be planted with a wide variety of wetland crops that may not remove as many nutrients but would have more economic value.

Farmer Adoption

The use of wetland plant taxa for nutrient removal has been known for several decades, and is accepted practice in many areas. The concept of producing ornamental plants on a commercial scale in constructed wetland biofilters is a relatively new concept. The attractiveness of this idea is that the grower can visualize cost-recovery for what otherwise might be rejected as an excessive capital outlay. Our work demonstrates the feasibility of producing 10 wetland plant taxa in shallow polyethylene covered wetlands for commercial use. The present work also presents some of the first data on nutrient (N and P) removal potential for these taxa in a demonstration ornamental wetland biofilter.

Areas needing additional study

Other promising areas of inquiry include further cost analysis of the benefits of wetland plant production in treatment wetlands. As well, further work could be done on the overwinter performance and requirements of the system, beyond the three taxa evaluated in this work. Ornamental wetland biofilters might also be useful for remediation of pesticide residues in nursery or greenhouse runoff. Though the present work provides numbers for nutrient removal on an area basis, it remains to be determined what area of treatment wetland would be needed to handle a given volume of nursery runoff on a unit, greenhouse, or acre basis.

Information Products

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.