IMAGE
Wetland Management

IMAGE

Photo courtesy of USDA NRCS

Photo Courtesy of USDA NRCS

Photo Courtesy of USDA NRCS

OVERVIEW

TERMINOLOGY

NATURAL WETLAND PROTECTION

Protection of Wetlands through Assignment of a Designated Use
The Challenge of Protection
The Relationship of Natural Wetlands to Water Pollution
Buffers and Other Protective Measures for Wetlands
Management Issues

NATURAL WETLANDS AND RIPARIAN AREAS AS BUFFERS
The Water Treatment Role of Natural Wetlands
Effectiveness of Natural Wetlands as Treatment Features
Effectiveness of Riparian Areas as Treatment Features
Wetland and Riparian Buffer Regulatory Issues

 
MANAGEMENT OF EXEMPT WETLAND ACTIVITIES
Silvicultural Exemption
Ranching Exemption
Innovative Management

WETLAND RESTORATION AND CREATION
Management for Wildlife
Landscape Considerations in Wetland Restoration/Creation
Riparian Restoration Guidelines for Water Quality
Coastal Wetland Restoration
Mine Reclamation
Urban Wetland Restoration
Innovative Commercial Wetland Creation

CONSTRUCTED WETLANDS
Overview
Constructed Wetlands for Animal Wastewater Treatment
Constructed Wetlands for Nonpoint Source Treatment
Constructed Wetlands for Mine Drainage Treatment


OVERVIEW

Wetland management generally involves activities that can be conducted with, in, and around wetlands, both natural and man-made, to protect, restore, manipulate, or provide for their functions and values. This discussion of wetland management is divided into issues associated with: 1) natural wetland protection; 2) activities, involving natural wetlands, that are specifically exempted from regulatory requirements; 3) wetland creation and restoration; and 4) wetland construction for water quality improvement.

The values of wetlands are by now well recognized (see Introduction, Functions and Values, and Protection sections) . The stated national goal for natural wetlands in the U.S. is one of no net loss, or protection of existing functions, as well as restoration of degraded functions. This protection goal involves not only buffering wetlands from direct human pressures, but also maintaining important natural processes that operate on wetlands from the outside and that may be altered by human activities. Management toward this goal should emphasize long-term sustenance of historical, natural wetland functions and values.

To support the national "no net loss" goal, many activities affecting natural wetlands must be conducted within the framework of government regulatory and other protection programs (see Wetland Protection section) . Manipulation of natural wetlands, within regulatory jurisdiction, is typically limited to restoration of degraded habitats. The use of natural wetlands for primary water quality treatment of either point or nonpoint pollution sources is inappropriate (Fields, 1993).

Exceptions to the rule of protection at the federal level are identified by specific Section 404 regulatory exemption categories, although such exemptions generally require the maintenance of some level of function in affected wetlands. Other exceptions include activities below minimum regulatory thresholds of applicability, and activities allowed by loopholes in the Act's construction. As a result of these caveats, by most estimates, Section 404 regulates only about 20 percent of the activities that destroy wetlands (GAO, 1991).

Wetland creation by man outside of any regulatory requirements presents opportunities for development of wildlife habitat and other valued functions as well as for capitalizing on a rapidly expanding technology for water quality improvement of both point and nonpoint pollution sources within the watershed.

Effective wetland management requires knowledge on a range of wetland subjects. Other sections within this Wetlands portion of the Education component provide current wetland information and lead to other materials that can assist wetland and watershed managers. This information can help a decision-maker evaluate wetland resources in a watershed to determine their functions, values, and roles in the watershed, assess risks, and prioritize protection. See the Wetlands Information Table of Contents to locate such information.

TERMINOLOGY

The management of wetlands and their use for water quality purposes has resulted in the introduction of a number of terms. Though definitions have not been standardized, the USEPA (Fields, 1993) recently established definitions for some of these variably applied terms, which we will follow for the rest of this discussion. Other terms used in this section are also provided here:

Natural Wetlands - wetlands that do not exist as the result of man's activities.

Wetlands - those areas that are inundated or saturated by surface or ground water at a frequency and duration sufficient to support, and that under normal circumstances do support, a prevalence of vegetation typically adapted for life in saturated soil conditions. Wetlands generally include swamps, marshes, bogs, and similar areas (40 CFR 232.2(r)).

Wetland construction - creation of wetlands built specifically for water quality improvement purposes; this typically involves controlled outflow and a design that maximizes certain treatment functions (Fields, 1993).

Wetland creation - bringing a wetland into existence, whether by accident or intentional, where none existed previously; this includes creation of wetlands for mitigation, habitat, and water quality purposes (Fields, 1993).

Wetland enhancement - the modification of a natural or created wetland to enhance one or more functions. Enhancement of some wetland functions may negatively affect other functions.

Wetland restoration - the reestablishment of a disturbed or altered wetland as one with greater function or acreage. This may involve reestablishing original vegetation, hydrology, or other parameters to reestablish original or closer-to-original wetland functions (Fields, 1993).

NATURAL WETLAND PROTECTION

The management goal for natural wetlands is generally constrained by regulatory and other government program requirements to the protection of existing functions or restoration of degraded functions. Our discussion of natural wetlands is divided into issues of protection, the role of wetlands as buffers for other receiving waters, unregulated or exempt activities, and restoration activities. For more detailed, case- specific guidelines and information on regulatory requirements, refer to the Wetland Protection section and contact the applicable regional office of the U.S. Army Corps of Engineers, the U.S. Environmental Protection Agency, and state and local environmental agencies. Throughout this discussion, references to natural wetlands are assumed to apply as well to man-made wetlands created under regulatory mitigation requirements, unless specified otherwise.

The management goal for undisturbed natural wetlands is typically to perpetuate existing functions. Functions are particular to a wetland's type and its position in the landscape (see Types of Wetlands section). Two major facets of managing wetlands for protection include buffering wetlands from direct human pressures, and maintaining natural processes in surrounding lands that affect wetlands and that may be disrupted by human activities.

Protection of Wetlands through Assignment of a Designated Use

The level of protection provided should conform with the designated use established for a wetland,. for example, aquatic life support or recreation. These coincide with two basic levels of protection recognized by environmental planners, preservation and conservation. Aquatic life support and wetland preservation connote a greater degree of protection, and involve, at most, passive use by humans (e.g., aesthetic enjoyment, wildlife observation). The recreation designated use, and wetland conservation status, connote a lesser degree of protection than do aquatic life support and preservation, on the level of protecting essential functions while allowing compatible human uses, such as recreational uses.

Factors to consider in setting the designated use and developing a management strategy for a wetland include:

It is not important to protect only rural or wilderness area wetlands. Urban wetlands can provide multiple values for suburban and city dwellers (Kusler et al. 1988). The aesthetic and recreational amenities of urban wetlands, and their value as wildlife habitat, can be significant. The capacity of a functional urban wetland in flood control can also be very important.

The Challenge of Protection

The simple goal of protecting a wetland's existing functions can prove to be incredibly complex in the modern landscape. It involves minimizing the human-induced changes affecting the natural forces that shape and sustain a wetland, such as hydrology, climate, biogeochemical fluxes, fire, and species movement. Pressures created by human activities include (see the Wetland Loss and Degradation section for a fuller review):

Other pressures that affect wetland functions operate less directly and are less apparent. These include:

The Relationship of Natural Wetlands to Water Pollution

While wetlands play a role in reducing pollutant levels of inflowing water, they also require protection as water resources. The USEPA states that the use of natural wetlands for water quality treatment for either point or nonpoint pollution sources is inappropriate (Fields, 1993). At the same time, it must be recognized that wetlands have in the past treated and continue to treat both point and nonpoint source discharges. Untreated point source discharges to wetlands have largely been eliminated through the Section 402 NPDES program. Remaining point source discharges are essentially of secondarily treated effluent, which still typically contains elevated levels of biochemical oxygen demand, suspended solids, and nutrients relative to natural inputs. Nonpoint sources have not been commensurately improved. Natural wetlands receive largely untreated runoff from much of the developed urban and agricultural area in this country. However, the USEPA (Fields, 1993) states that proper management dictates that they be protected from such inputs using water quality standards promulgated by each state. Water quality standards specifically for wetlands are gradually being adopted by states. Progress is slow in this area, but NPS pollution control is gaining momentum. Although significant NPS loading to wetlands is undesirable, it will take time to address, and measures taken to curtail it will likely result in reduced but not eliminated loadings to wetlands.

Given the potential impacts of the myriad forces acting on wetlands, it is important to develop and implement strategies for the long-term protection of these ecosystems. A key element of any protection strategy is the establishment of physical buffers to minimize edge effects and to mitigate water quality impacts.

Buffers and Other Protective Measures for Wetlands

A buffer typically consists of a band of vegetation along the perimeter of a wetland or water body, preferably natural habitat, but including previously altered, stable native or introduced species. Once the need for a buffer is recognized, establishment of a suitable width is the critical task. In reality, many government agencies establish buffer requirements based on political acceptability and/or assumed aquatic resource functional value. Nevertheless, a fully informed buffer design must consider the nature of the encroaching activity, the buffer itself, the resource to be protected, and the buffering function to be performed. Castelle et al. (1994) identify four criteria for determining adequate buffer size to protect wetlands and other aquatic resources:

Relative to the last criterion, buffers can perform the following functions: In addition, Brown et al. (1987) describe how the transition area or ecotone between what is characteristically identified by regulatory agencies as wetland and undisputed upland typically supports higher species diversity and is more ecologically important than either adjacent habitat. This transition zone is of great importance to many wetland-dependent species and in riverine systems serves a vital role in maintaining regional species diversity as part of a habitat corridor.

A literature search by Castelle et al. (1994) of studies on specific buffer performance found that for sediment removal, necessary widths ranged from 10 to 60 m; for nutrient and metals removal, widths ran from 4 to 85 m; for species distribution and diversity protection, from 3 to 110 m was required; and for water temperature moderation, requirements ranged from 15 to 28 m.

Castelle et al. found that buffers less than 5 to 10 m provide little protection of aquatic resources under most conditions. They recommended minimum buffer widths of 15 to 30 m under most circumstances, with the lower end of this range providing basic physical and chemical buffering, and the upper end being the minimum needed for maintenance of biological components of wetlands and streams. They noted that fixed-width buffer approaches are easier to enforce, but that variable-width buffers are more likely to provide adequate protection on a specific-case basis. States have guidelines on desirable buffer widths and a number of states have buffers that range from 45 m to 300 m (Buchsbaum, 1994). A minimum ninety-meter buffer around state and federal wildlife refuges and conservation areas has been recommended.

Modelling can be performed to determine the width of a buffer that will reduce loading of suspended solids and bacteria from stormwater. These models relate buffer soil permeability, slope characteristics, width, and surface roughness to the surface flow (Phillips and Phillips, 1988, cited in Buchsbaum, 1994).

Although a narrow buffer may provide significant water quality benefits, the capacity for a narrow buffer to provide habitat or to act as a corridor for species is negligible. Optimal corridor widths for water quality purposes vary from 50 feet (16 m) to over 100 (34 m), with the wider corridor providing better conditions for management of wildlife (Davis, 1993).

When buffer acreage is not available or greater protection is called for, other measures can be employed. Wetlands in urban areas often require a greater level of protection. Degrading activities can include: off-road vehicle use (a problem in rural areas as well); pedestrian access; mowing; landscaping; solid waste dumping; domesticated animal access and resultant wildlife decimation, herbivory, vegetation trampling, soil compaction, and waste deposition; and others. Off-road vehicle access can be prevented by using post and cable barriers (Zentner, 1994). Pedestrian and pet access can be directed, discouraged, or eliminated through placement of shrub hedges, fences, open water buffers, signs, or a combination of these measures on the perimeter of a wetland. Common use piers and boardwalks over marshes or through swamps can be used to reduce degradation from recreational activities (Buchsbaum, 1994).

The measures noted above can be implemented by a local government agency, a wetland regulatory authority, a homeowners association, a concerned citizens' group, private individuals, or others. Community support can be developed for wetland protection. Volunteers to implement protective measures can be found in conservation organizations, volunteer water quality monitoring groups, and citizens' groups (USEPA, 1993c). Schools may value the opportunity for hands-on environmental education and involvement.

Permits issued by regulatory agencies for development around wetlands should include conditions requiring the permittee to inform future lot owners of restrictions on their use of wetlands located on, partially on, or abutting their lots. Permits can explicitly require full disclosure to potential lot purchasers; deed restrictions can be placed on such lots; permit conditions can require similar disclosure of responsibilities to subsequent lot owners. See the Wetland Protection section for a full discussion of regulatory tools.

Management Issues

As outlined above, in addition to buffering wetlands from human impacts, protective management involves maintaining important natural processes that operate on wetlands from the outside and that may be altered by human activities. One of these processes is fire. Many wetland types are adapted to periodic burns, but development interrupts natural fire patterns. Controlled burning is a management strategy that mimics the natural process in developed landscapes. It promotes marsh plant diversity and eliminates undesirable vegetation (Kent 1994b). Burns result in improved feeding and nesting for a variety of species.

Construction Impacts: For unavoidable road alignments through wetlands, it is possible to reduce impacts through "end-on" construction (USDOT 1994). Instead of driving heavy equipment in the wetland or building fill causeways or embankments, equipment is placed on work platforms mounted on concrete piles. A crane drives the piles and adds the bridge viaduct bay by bay. Waterfowl species do not seem to be disturbed by this construction process.

Mosquito control: Mosquito control is one reason that wetlands have historically been drained and it remains a cause of wetlands loss today. Natural wetlands, as well as restored and created wetlands, are habitat for mosquitoes. Constructed wetlands in particular may stagnate and increase breeding of mosquitoes because they lack a hydroperiod or do not contain predatory fish species.

Mosquito control does not have to cause wetland impacts or loss. However, pesticides such as organophosphates (e.g., malathion) that are used to control mosquitoes may be toxic to wetlands fish and aquatic invertebrates. Other more natural pesticides or bacteria can provide a more directed approach to mosquito control (Buchsbaum, 1994). Bacillus thuringiensis israelensis (Bti) is one bacterium that is more specific and less toxic than malathion. Careful application can avoid impacting other chironomid larvae that form the base of the food web in wetlands (Buchsbaum, 1994). An Integrated Pest Management approach to mosquito control should be used rather than drainage or non IPM-application of pesticides. Allowing predators of mosquitoes such as mosquito fish (Gambusia affinis), and killifishes (Fundulus spp.) access to breeding areas or introducing these fish should be part of an IPM mosquito control program.

Another method of mosquito control is to ensure that created and restored wet meadows and marshes have a hydroperiod which includes dry conditions during the mosquito egg- laying or hatching season (Zentner, 1994). The dry conditions will prevent egg-laying and hatching.

NATURAL WETLANDS AND RIPARIAN AREAS AS BUFFERS

The Water Treatment Role of Natural Wetlands

As discussed above, while USEPA states that the use of natural wetlands for water quality treatment for either point or nonpoint pollution sources is inappropriate (Fields, 1993), it is recognized that wetlands have in the past treated and continue to treat both point and nonpoint source (NPS) discharges. It will take time to curtail NPS pollutant loading to wetlands and measures taken to do so will likely result in reduced but not eliminated loadings to wetlands. Therefore, it is important to understand not only the long-term effects of such elevated loadings on wetlands, but the ability of wetlands to further treat these loadings prior to discharge into receiving waters. This ability of wetlands and riparian areas to process NPS pollutant loads has received significant study.

The most important forested wetlands to manage and protect as stream quality buffers may be those along first- and other low-order streams (Brinson, 1993). Wetlands along first-order streams are very efficient at nitrate removal from groundwater and runoff, and sediment removal from surface water and runoff (Whigham et al., 1988); they protect streambanks from erosion, and moderate stream temperatures by shading the water, which benefits aquatic life. Wetlands (floodplains) along higher-order streams influence water quality to a much smaller degree, since the upland runoff that passes through them and joins the stream is a much smaller fraction of the total stream flow than it is for headwater wetlands. Wetlands along large streams do, however, provide water quality benefits during flood events, a function that headwater wetlands do not provide.

Effectiveness of Natural Wetlands as Treatment Features

Water quality processes in natural wetlands are much more challenging to study than those in constructed systems. One main reason is that their water sources, rainfall and runoff, are climatically driven, making them highly variable hydrologically. It is also frequently a challenge to quantify all of the input sources and output paths. As a result, researchers tend to use differing approaches to study different systems, making their results more difficult to compare than those for the more controlled environments of constructed wetlands. Treatment efficiencies measured in natural wetlands have proven to be more widely variable than those in constructed systems, probably due only in part to differences in experimental methods, and more so to the diversity in natural system structure, function, and historical loading trends.

A substantial amount of research has focused on the biogeochemical role of wetlands in undisturbed landscapes with relatively natural levels of inputs (Nixon and Lee, 1986). Of greater interest, however, is the ability of wetlands to improve the quality of waters polluted by human activity. Significant work has been done, much of it within the last 15 years, on treatment of various polluted water sources by natural wetlands. Stormwater and wastewater have received significant attention. As reported in a literature search by Phillips et al. (1993a), natural wetlands treating domestic and municipal wastewater have removed 70% to 90% of organic matter, 26% to 70% of nitrogen, 12% to 70% of phosphorus, and high percentages of some metals. Natural wetlands treating stormwater have been somewhat more variable and less efficient. Suspended solids removal has ranged from 40% to 85%, and metals removal has been somewhat lower than in wetlands treating wastewater (Carr and Rushton, 1995; Phillips et al., 1993a). However, in one case, inorganic nitrogen removals greater than 85% were reported, as well as phosphorus reductions of greater than 70% for a natural marsh treating stormwater (Carr and Rushton, 1995).

Several researchers have looked at the efficacy of natural wetlands in treating agriculturally derived nutrient, sediment, and other pollutant loads. Two coastal forested peatlands receiving pumped cropland drainage over two years differed in nitrogen removal, reducing Kjeldahl nitrogen concentrations an average of 69% and 29% from 3 and 2.2 mg/l, and lowering nitrate concentrations 71% and 100%; phosphorus concentrations were lowered 93% and 63% from .36 and .13 mg/l; sediment removal was more consistent, with reductions of 97% and 92% (Chescheir et al., 1987). An Irish peatland that had received dairy wastewater for several decades showed high levels of nitrogen and phosphorus removal, lowering ammonia levels 88% from 15 mg/l, nitrate levels 92% from 20 mg/l, and ortho-phosphorus levels 73% from 8 mg/l (Costello, 1989). A restored prairie pothole in Minnesota showed promise for cropland runoff nitrogen and sediment removal in its first years of operation, lowering nitrate levels 70% from 4 mg/l and total suspended solids levels 92% from 1036 mg/l. Phosphorus removal was not good, with a 9% reduction from .44 mg/l (Jacobson, 1994). Several agricultural operations in Florida have used natural marshes and sloughs to treat drainage from citrus, pasture, and rangeland with variable success (Fall and Hendrickson, 1988; Goldstein, 1986; Federico, 1978). Low inflow total nitrogen levels (1 to 2 mg/l) were marginally improved or contributed to net export, while phosphorus removals ranged from 2% to 72%. A coastal creek floodplain swamp in North Carolina reduced phosphorus loads derived from cropland and animal operations by 43% over two years (Kuenzler et al., 1980).

Effectiveness of Riparian Areas as Treatment Features

Riparian areas, which include floodplain uplands as well as wetlands, are considered perhaps the most important buffer areas for protecting receiving water quality (Gilliam 1994). A number of researchers have quantified the effectiveness of both forested and grassed riparian areas for removing sediment, nitrogen, phosphorus, organic matter, and some pesticides from both surface water and ground water. Much work has focused on elevated constituent levels due to agriculture. Removal processes include deposition, absorption, adsorption, plant uptake, denitrification, and others (Welsch, 1991).

In terms of sediment, riparian zones along small streams in Coastal Plain North Carolina trapped an estimated 84 to 90 % of sediment eroding from cropland over a 25-year period (Cooper et al., 1987). Much of the coarse sediment was deposited very soon after entering the riparian area, with more than 50% of deposition occurring within 100 m of field edges. Lowrance et al (1988) used radioisotope dating in the Georgia Coastal Plain to determine that more sediment was deposited in a riparian forest over the same 25-year period than left adjacent agricultural fields, the difference being attributed to upstream inputs.

Several researchers have quantified reductions in problematic nitrate-nitrogen levels carried in cropland runoff traveling in shallow ground water across riparian areas. In one experiment, nitrate was reduced from 15 mg/l to 2 mg/l in the first 10 to 15 meters (30 to 50 ft) of riparian forest as it moved from a field toward a stream (Evans et al. 1993). Similar nitrate reductions were observed in a riparian forest in Maryland, most of the removal occurring within the first 19 m of the zone (Correll and Weller, 1989; Peterjohn and Correll: 1984; 1986). Nitrate load reduction was estimated at 45.5 kg/ha/yr. A grass riparian area between 18 and 27 m wide in Pennsylvania removed about 51% of nitrate entering at 21 mg/l, while an equal width of forested riparian zone on the opposite streambank lowered nitrate concentrations 83% from 4.3 mg/l (Schnabel, 1986). Haycock and Pinay (1993) in England measured nitrate load reductions of 99% across 26 m of riparian forest and 84% across 16 m of riparian grassland during the winter. Nitrate from dairy wastewater land application averaging 8 mg/l was reduced 89% by 30 m of reforesting grass riparian area in Coastal Plain Georgia over a 3-year period (Vellidis et al., 1995). Denitrification rates over this period averaged 68 kg/ha/yr. Overall, most other researchers have had similarly positive results with nitrate, and most believe it is largely removed from the system in gaseous forms through denitrification.

Removal of phosphorus (P) from cropland runoff by riparian areas has been somewhat less extensively researched. Cooper and Gilliam (1987) measured P deposited with the sediment in North Carolina riparian areas (see above). They estimated that over a 25-year period 50% of incoming P from agricultural areas was deposited in the riparian area. Phosphorus removal required significantly more area than a similar percentage of sediment removal, since P was concentrated in the finer sediment particles that take longer to settle. Lowrance et al. (1984) estimated that 30% of incoming P was retained in a Georgia Coastal Plain riparian forest over 3 years. In Maryland, 40 to 75 m of riparian forest retained 81% of surface water P entering at 5 mg/l and was a net exporter of dissolved ground water P (Peterjohn and Correll, 1984; 1986). Gilliam (1994) stated that riparian buffers do a reasonably good job of removing P attached to sediment, but are relatively ineffective in removing dissolved P.

Little information is presently available on removal of pesticides and fecal bacteria by riparian areas. Preliminary data from research in Kentucky on fecal bacteria indicates that removals are highly variable (Gilliam, 1994). Preliminary results of atrazine and alachlor dosing studies from Coastal Plain Georgia show reductions in surface water concentrations of 84% to 87% below conservative tracer levels, while ground water atrazine concentrations were lowered 41% and ground water alachlor 6% below tracer levels (Vellidis et al., 1995).

Phillips (1989) used a model, the Riparian Buffer Delineation Equation (RBDE), to estimate buffer widths needed to effectively treat agricultural runoff in the Coastal Plain of North Carolina. The equation uses properties found in soil surveys (slope gradient, soil moisture storage capacity, surface roughness, and soil saturated hydraulic conductivity) along with proposed buffer width to assess potential treatment capacity. Phillips found a wide variation in buffer effectiveness, with widths ranging from 5 to 93 m needed to remove nitrate from runoff volumes typical of 50 acres of row crop on relatively poorly-drained soils. He found slope gradient to be the most important variable bearing on effectiveness.

Wetland and Riparian Buffer Regulatory Issues

Although wetlands and riparian areas along low-order streams can provide effective water quality improvement, as well as habitat, floodwater storage, ground and surface water recharge, and critical amphibian breeding and reproduction sites, first order streams are often not protected as rigorously as more visible, larger streams. Low order streams and their associated wetlands and riparian areas are not protected by the nationwide Permit No. 26 within the Section 404 program, which allows up to 10 acres (4.0 hectares) of wetland impact. Laney (1988) determined that losses of isolated and limited-flow wetlands due to this authorization in North Carolina were significant, both individually and cumulatively, and these losses appeared inconsistent with the objective of the Clean Water Act to maintain the physical, chemical, and biological integrity of the nation's waters, including wetlands. Therefore, wise watershed management would include the use of other means to ensure the protection of headwater wetlands (see Wetland Protection section).

MANAGEMENT OF EXEMPT WETLAND ACTIVITIES

Exceptions to the rule of protection described above for natural wetlands include specific regulatory exemption categories, which generally require the maintenance of some level of function in affected wetlands, and activities below minimum regulatory thresholds of applicability. All of these activities can significantly impact wetlands if not conducted conscientiously.

Traditional land use activities exempted from federal and many state regulations include silviculture, agriculture, ranching, and sometimes mining. Most exemption language, including that in the Clean Water Act Section 404, stipulates that such operations in wetlands must be ongoing and established in nature to qualify for exemption. Another frequent caveat in state wetland rules is that the activity must make appropriate use of best management practices, or BMPs, for the exemption to hold. Exemption language may require individuals to obtain an approved conservation compliance plan from the National Resources Conservation Service (NRCS) or other approved plan to support their activities.

Section 404(f) of the Clean Water Act describes exempt agricultural activities and BMPs required for forestry operations to prevent adverse effects (at 40 CFR 232.2). The USEPA has also published more detailed guidance on agricultural activities and Section 404 (USEPA, 1991a). State departments of forestry often publish forestry BMP manuals with detailed information.

Silvicultural Exemption

To minimize impacts to wetlands, harvesting must be managed carefully and BMPs must be implemented. "Minor drainage" is permitted under Section 404, but ditches that significantly alter the hydrology of a wetland may not be constructed.

In terms of specific practices, road and skid trails should be minimized in number, width, and length. They should be located sufficiently far from water flow, or be bridged or culverted, so as not to impede or increase water flow or contribute to stagnation (USEPA, 1993a; Siegal and Haines, 1990). Trails should be maintained to prevent erosion. Low ground pressure vehicles and aerial logging reduce the soil compaction and hydrologic modifications resulting from heavy equipment and road construction (USEPA, 1993a; Vowell and Olszewski, 1989). Pesticides with high toxicity to aquatic life should be avoided, and slow release fertilizer formulations based on soil tests should be used (USEPA, 1993).

Maintaining riparian buffers along streams will enhance forest regeneration as well as provide wildlife habitat. At least a few snags and cavity trees should be left for habitat and tree stumps should be 12 inches high or less (USEPA, 1993; Vowell and Olszewski, 1989). The same species that existed on the site prior to harvest should be replanted afterward. Changing from mixed hardwoods to pine, for example, may change site hydrology because of the differences in evapotranspiration and growth rate of the species (Richardson and McCarthy, 1994; Skaggs, 1991). Discharges related to land clearing for silviculture may be regulated and it is advisable to contact a Corps or USEPA regional office (USEPA, 1994c).

Ranching Exemption

The CWA Section 404(f)(1)(A) exempts normal ranching activities from wetland permitting.

Under the ranching exemption, controlled grazing by livestock in winter and spring can improve herbaceous wetland nesting habitat and promote plant growth and seed production (Kent, 1994b). It has been recommended that livestock be removed after 50% of forage plants have been grazed.

The Bureau of Land Management, ranchers, and private landowners are utilizing beavers to restore degraded riparian areas in the West (USEPA, 1993b; SCS, 1989; Stuebner, 1992). The beavers create ponds that raise the water table and reestablish a wetland habitat, which wetland plants re-colonize. The plants can then be grazed in late winter and early spring.

After employing beavers to restore a riparian area and instituting a managed grazing regime there, one rancher increased weaning weight of his calves by 150 lbs., and increase his cow/calf numbers by 50% (USEPA, 1993b). The restored areas provide important habitat for many game and nongame species, and recreational opportunities as a result.

Innovative Management

Degraded, prior converted wetlands may offer opportunities for innovative management approaches (which may require permitting). In the southeastern coastal plain, a mixed- use, aquaculture-silviculture (crayfish-timber) enterprise can be quite successful (Mitsch and Gosselink, 1993). The hydrologic cycle of a bottomland hardwood forest can be simulated by winter impoundment of a prior converted or degraded swamp or area planted in flood-tolerant tree species. The crayfish are harvested in the spring and summer. Such a system can restore bottomland hardwood community structure and provide water quality benefits of nutrient removal. Inflow of toxic compounds must be monitored closely, however, because crayfish accumulate them. Since timber rotations are long, generally 20 - 50 years, this system can provide wildlife habitat as well, particularly if it is not intensively managed.

It is recommended that water depths of such managed impoundments not exceed 8 inches (15 cm) (Dugger and Frederickson, 1992). Inundation adversely affects terrestrial species and flood-intolerant tree species (King 1995). The forest should not be impounded for 3 years after acorn germination so that seedlings can become established (Kent, 1994b). Drawdown must be completed prior to the beginning of the growing season as trees and plants will be adversely impacted even if artificial inundation lasts only a few days into the growing season (King, 1995). Seedlings must not be inundated. Impoundment should not be conducted where an area is an important corridor for animal movement or where rare species occur.

Some rice farmers have found that they can take advantage of the annual flooding cycle of farmed wetlands to combine rice farming with crayfish production (Mitsch and Gosselink, 1993). Crayfish forage in the rice fields when they are re-flooded after the summer-fall rice harvest; the crayfish are harvested in spring before draining and replanting of rice. Pesticides can not be used on the fields because of the crayfish; however, flooding eliminates much of the need for pesticides.

Other farmers are converting flood-prone farmland into wildlife refuges in cooperation with federal and state agencies (Deterling, 1994). The farmers receive direct payments or tax deductions. Runoff of nutrients, agrichemicals, and eroded soil into nearby water resources is minimized or eliminated, and the wetland can provide functions in the watershed again.

Contact the Ducks Unlimited Private Lands Program 1-800-453-8257

WETLAND RESTORATION AND CREATION

While an implicit part of the national goal of no-net-loss involves mitigation for unavoidable impacts to wetlands,an explicit part of the goal is the restoration of wetlands where possible to recover the historical quality of the remaining acreage base (Conservation Foundation, 1987). Restoration may be required as part of a permitting process, but restoration efforts may also be prompted by environmental resource management goals for habitat or water quality improvement in keeping with the net- recovery clause of the national no-net-loss goal. In either case, degraded wetlands present restoration opportunities for improvements to water quality, habitat, water storage and other functions, and these opportunities can be particularly useful for watershed-scale environmental planning. The goal of restoration is typically to reestablish wetland ecosystems to levels that existed prior to human influence. Wetland creation can include regulatory mitigation or commercial and private creation efforts outside of regulatory requirements. A useful volume was recently released by the National Academy of Sciences addressing restoration of wetlands and other aquatic ecosystems from a management standpoint (National Research Council, 1992).

Management for Wildlife

Wetlands are especially critical habitats for wildlife, and exceed all other land types in wildlife productivity (Payne, 1992). Historically, wetland wildlife management was overwhelmingly concerned with maximizing production of waterfowl and furbearing mammals, and was focused largely on game species. By the 1970's, scientific and public perspective had shifted and resulting laws codified a concern for managing wildlife for diversity, emphasizing non-waterfowl and non-game species (Kent, 1994c). Kent (1994c) summarized a range of approaches to managing for habitat diversity developed in response to this evolution in perspective.

During this time, the larger question of whether wetland and other habitat should be managed at all, and if so, in what sense, gained high visibility. Kent (1994c) holds that the practice of wetland habitat management in the sense of active manipulation should be limited to degraded and created wetlands, as discussed below.

"It is vainglorious to expect that managers can improve on the complex dynamic processes of natural undisturbed wetlands. Active management will by necessity enhance habitat for some species while degrading habitat for other species. Management may fail because of inadequate or inaccurate information, imprecise water control, colonization and modification by nuisance species, or even political or public pressure to terminate or modify management techniques or goals (Fredrickson, 1985). Therefore, it seems reasonable to reserve active management for wetlands known to be degraded and created wetlands."

For this reason, the subject of wildlife management is located here under wetland restoration and creation. Some additional wildlife management discussion occurs under the Management of Exempt Wetlands subsection for the same reason, since exempt systems are typically degraded and offer the possibility of improvement through active management.

Marsh creation or restoration is thus a good opportunity to manage wetlands for broad wildlife habitat goals. Not only can a restored marsh provide enhanced wildlife benefits, but other functions can be improved concurrently. Whether created or restored, wetlands designed for wildlife should take into consideration: minimum habitat area of anticipated species, their tolerance for disturbance, and the system's functional relationship to other water resources and adjacent ecosystems (Kent 1994b). It should be noted that while created wetlands can be suitable for some species, such as waterfowl, other, particularly threatened and endangered, species do not colonize artificially created wetland systems as readily or consistently as they do restored natural wetlands (Kent 1994b).

While management of restored or created wetlands should as a rule emulate the functions of undisturbed marshes, there may be times when single- or priority-objective management is appropriate. For a given wetland site, a restoration or creation management strategy must involve determination of the most important values to be obtained, and of whether a single, exclusive value outweighs the suite of values to be obtained from historic restoration. If a single-purpose wildlife use is sought, such as certain fish utilization, management may result in manipulation of marsh hydrology at the expense of other species and wetland functions. For example, game fish species require consistently deep water, yet shallow, emergent-plant-depth water levels provide the highest plant species diversity and greatest overall wildlife use of marshes (Mitsch and Gosselink 1993; Kent 1994b). At the same time, waterfowl require different structural conditions depending on species needs for feeding (divers versus dabblers), nesting, or staging (Weller 1981; Kent 1994b). In general, a ratio of no more than 1:1 open water to emergent vegetation maximizes waterfowl use (Weller 1981). Thus, tradeoffs are inevitable when structural components of a wetland, such as water level, are artificially manipulated. Any management strategy beyond reestablishment of historical functions must weigh these tradeoffs in light of management goals.

Hydrologic control can involve passive, climatically driven designs that emulate some natural ecotype, or managed designs using operable weirs, control gates, and pumps. The following table illustrates general relationships between water level and marsh characteristics.

Water level management for marsh species

Summer water level      Moist soil(mudflat)     15 cm     > 30 cm

Plant species diversity            fair         excellent  fair
Wildlife use and diversity         fair         excellent  good
Fish abundance                     none         good       excellent
Migratory bird use                 excellent    good       fair
Invasion by nuisance species       high         low        low
adapted from Mitsch and Gosselink (1993).

In much of the continental U.S., emulating the hydrology of a natural marsh would involve drawdown of water levels in the spring and gradual re-flooding in the fall. This pattern can stimulate primary productivity (Kent 1994b). Creation of a marsh adjacent to agriculture will likely provide elevated nutrient levels that will stimulate productivity and, if not too great, facilitate establishment of the wetland community while improving downstream water quality over previous levels.

Landscape Considerations in Wetland Restoration and Creation

Created wetlands for nonpoint source pollution control are advocated as an important part of any watershed or floodplain restoration plan (Mitsch, 1994). Location of constructed wetlands in the landscape is an important factor in determining their role. As discussed in the Riparian Wetlands subsection, the most important wetlands to manage and protect as stream quality buffers may be those along first- and other low-order streams (Brinson 1993). Wetlands along first-order streams are very efficient at nitrate removal from groundwater and runoff, and sediment removal from surface water (Whigham et al., 1988). Constructed wetlands bordering agricultural fields can be designed to intercept tile drainage with high nutrient levels that otherwise often flows directly into receiving streams, bypassing even riparian areas. Placing wetlands in a distributed pattern high in the watershed may incur less total runoff and erosion for the entire watershed than the same acreage put into large wetlands low in the watershed (van der Valk and Jolly, 1993).

Mitsch (1993) observed in a comparison of experimental systems using phosphorus as an example that retention as a function of nutrient loading will generally be less efficient in downstream wetlands than in smaller upstream wetlands. Wetlands (floodplains) along higher-order streams influence water quality to a much smaller degree, since the upland runoff that passes through them and joins the stream is a much smaller fraction of the total stream flow than it is for headwater wetlands. Wetlands along large streams do, however, provide water quality benefits during flood events, a function that headwater wetlands do not provide. Mitsch (1993) cautioned that the downstream wetlands could retain more mass of nutrients than upstream systems, and that a placement tradeoff might be optimum. From a management standpoint, creating many smaller wetlands around a watershed would mean dealing with more landowners, but taking less land out of production on any one farm than creating a few large wetlands, and is more fair in terms of not asking any landowner to contribute more than what is needed to treat the runoff from their land (van der Valk and Jolly, 1993).

Hammer (1992) envisions a holistic watershed wetland management approach involving a hierarchical arrangement of restored or created wetlands within a watershed landscape. Following conventional on-farm BMP systems, first-order control involves constructed wetlands designed specifically for animal wastewater, processing facility wastewater, or septic tank effluent treatment. Second-order control also occurs at the individual farm level, and consists of constructed wetland/upland systems, such as the nutrient/sediment control system described above, for treating cropland runoff or discharge from animal wastewater treatment systems, and providing some ancillary benefits as well. Third-order control requires a larger, watershed picture, and involves nutrient/sediment control systems, constructed wetland/pond complexes, and restored or created wetlands and riparian areas along many small streams higher in the watershed, providing water quality, hydrologic buffering, life support, and other values. Finally, fourth-order control uses large wetlands low in the watershed primarily for hydrologic buffering and habitat support values in addition to limited water quality benefits. First- and second-order systems are located within the bounds of individual farms and require active operation to maintain optimum treatment performance, while third- and fourth-order elements provide water quality benefits to runoff from numerous farms or entire watersheds, and function without intervention.

Instream wetlands can be created on small streams by impounding or adding a control structure to the stream. Mitsch (1993) observed that creation of in-stream wetlands is a reasonable alternative to upland locations only in lower-order streams and that such wetlands are susceptible to reintroduction of accumulated pollutants in large flow events as well as being unpredictable in terms of stability. Such systems would also likely involve higher maintenance and management costs than off-stream designs.

Wetland creation or restoration can provide significant benefits to surrounding systems in addition to water quality improvement. Diversity of wetland structural habitat in the landscape (particularly small multiple wetlands that differ in water level, plant species, and size), tends to increase species diversity and abundance (Weller 1981; Fleming et al. 1994). Similar to natural wetlands (see Natural Wetlands as Buffers subsection above), created systems can act as buffers for wildlife habitat. They protect streambanks from erosion, and moderate stream temperatures by shading the water, which benefits aquatic life. Larger riparian wetlands further downstream provide flood control and wildlife benefits. Knight (1993) noted that wetlands placed high in the watershed are likely to have more intermittent, less reliable water supplies, and thus exhibit lower primary production and lower overall food-chain benefits than those low in the watershed with perennial water supplies.

Riparian Restoration Guidelines for Water Quality

The U.S. Forest Service has published guidance on reforesting previously cleared riparian areas and renovating degraded riparian areas for the protection of receiving water quality (Welsch, 1991). The guidance is directed toward agricultural and silvicultural land uses and emphasizes that riparian buffers are meant to be used as part of a sound land management system including upland best management practices, and can be damaged and functionally impaired otherwise.

The design of the riparian buffers described above includes three zones intended to filter surface runoff and shallow groundwater flow. Beginning at the edge of the receiving water body, the first zone is a fixed 15 ft. wide, undisturbed native forest/shrub zone to provide a stable ecosystem at the water's edge, to perform nutrient buffering, to provide shade, and to contribute detritus and large woody debris to the water body. Landward of zone 1, zone 2 is the heart of the riparian buffer. A minimum of 60 ft. wide, it is composed primarily of native trees and shrubs, and it provides contact time and carbon energy source for buffering processes and for long-term sequestering of nutrients by trees. Periodic timber harvesting and stand improvement is acceptable in this zone. Livestock are to be excluded from both zones 1 and 2. At the landward margin, zone 3, a minimum of 20 ft. wide, is a graded, dense grass/forb strip for sediment control and nutrient uptake. Shaping into diversions, basins, and level spreaders toward this end is appropriate. This zone should be actively managed; mowing is recommended, grazing is acceptable, and periodic sediment removal, reshaping, and revegetating are necessary to maintain performance. Actual zone widths beyond the minimum can be determined based on USDA-defined Hydrologic Soil Groups found in the buffer; on the ratio of buffer area to source area; or on Soil Capability Classes of the buffer as shown in soil surveys. In addition, more involved buffer width estimation models utilizing properties and data found in soil surveys are available (Phillips, 1989).

Coastal Wetland Restoration

Coastal marsh restoration and creation efforts have been more successful than similar inland attempts (Redmond, 1992). This success appears to be due largely to researchers' ability to predict more accurately the key component, hydrologic patterns, in tidally influenced areas than in freshwater settings. Also, coastal restoration efforts have perhaps had a longer history than freshwater wetland restoration.

Restoration of coastal marshes and creation of salt marshes on dredge spoil has been found to facilitate shoreline aggradation, stabilize beach erosion, and protect landowners from the impacts of storms (Mitsch and Gosselink, 1993; NOAA, 1990; NOAA, 1995a). Restoration of wetlands on eroding shorelines can protect critical habitat for marine life and freshwater aquatic life (NOAA, 1995a; NOAA, 1995b), as well as reduce land subsidence (NOAA, 1995b; Duffy and Clark, 1989).

Mine Reclamation

Many wetland creation and restoration projects have been conducted at phosphate, coal, and sand and gravel mines throughout the United States. Some of these projects include creation or restoration of riparian wetlands, including expansive bottomland hardwood swamps in the central Florida phosphate mining district (Clewell, 1990). Some reclamation has been accomplished by terracing mountainsides to control erosion and treat acid mine drainage (Mitsch and Gosselink, 1993). In the Midwest, more than 5,666 hectares of wetlands have become established through natural colonization of coal mine slurry ponds. Many mining companies are reclaiming their slurry ponds through wetland creation (Levine and Willard, 1990).

Wetlands created or restored for mine reclamation may provide habitat for birds, mammals, herpetofauna, and macroinvertebrates if water is not acidic and does not contain high levels of toxic compounds. They may be stocked with fish and used for recreational activities. Water in reclaimed wetlands has been used for crop irrigation, livestock, fire protection, industrial purposes, and even as a water supply for human use (Brooks, 1990). There is extensive literature on the subject of mine reclamation as noted in Clewell (1990) and Brooks (1990). Also, individual project reports on phosphate mine reclamation are available through inter-library loan from the Florida Institute of Phosphate Research, 1855 W. Main St., Bartow, Florida 33830; (813) 533-0983.

Urban Wetland Restoration

Wetland restoration can be an important contributor to downstream habitat and water quality recovery in urbanized landscapes. Restored urban wetlands can help protect floodplains and streambeds that are otherwise degraded by urbanization forces, and can help to minimize downstream flooding that results from urbanization. Such wetlands can also reduce sedimentation of lagoons, bays, and other downstream water resources (Williams, 1990; Gale and Williams, 1988; Marcus, 1988). Larger restoration projects are more cost effective and are typically more beneficial ecologically as well (King and Bohlen, 1994; Lewis, 1988). Larger areas may provide habitat for interior species that an equivalent acreage of smaller parcels cannot support.

Upland buffer zones adjacent to urban restoration projects are important to protect them from degrading forces and provide important habitat used by many wetland species (Lewis, 1988). Such projects require other protective measures as well to sustain their functions long-term (see the Wetland Buffers subsection above).

Restoration of urban wetlands in coastal California has been fairly successful. In an evaluation of 120 such completed projects, 65 percent functioned similarly to natural wetlands, 25 percent were functional but resulted in different habitats than were originally designed, and 10 percent were failures (Zentner, 1988).

Innovative Commercial Wetland Creation

In association with Ducks Unlimited, the Natural Resources Conservation Service, and federal or state fish and wildlife agencies, farmers throughout the South are managing their rice, corn, and soybean fields as wetland habitat for economic benefit (Deterling 1994; Muzzi 1994). By flooding fields from November through February, farmers provide winter habitat along migratory flyways. Some farmers keep fields flooded longer because flooding keeps the soil soft and kills flood-intolerant weeds, thus eliminating tillage costs. Waterfowl eat flood tolerant weeds and weed seeds. Crops can then be planted earlier and faster, since crops germinate faster because tilling is not necessary and weeds are not competing for nutrients and space. Farmers who flood their fields save from $10 -$30 per acre on herbicide and tillage costs (Muzzi 1994)and there is less chemical runoff to pollute local water resources. Farmers can also receive income by leasing access rights to hunters or hunting clubs (Deterling 1994). Ducks Unlimited provides farmers with water control structures and advice in return for a 10-year agreement to conduct winter flooding and provide waterfowl habitat.

CONSTRUCTED WETLANDS

IMAGE

Photo courtesy of USDA NRCS

Overview

Interest in the use of natural physical, biological, and chemical aquatic processes for the treatment of polluted waters has increased steadily in the United States over the last two decades. This interest has been driven by growing recognition of the natural treatment functions performed by wetlands and aquatic plants, by the escalating costs of conventional treatment methods, and by a growing appreciation for the potential ancillary benefits provided by such systems. Aquatic treatment systems have been divided into natural wetlands, constructed wetlands, and aquatic plant systems (USEPA, 1988). Of the three types, constructed wetlands have received the greatest attention for treatment of point source pollution. As discussed and defined at the beginning of the Wetlands information section, constructed wetlands are a subset of created wetlands designed and developed specifically for water treatment (Fields, 1993). They have been further defined as:

engineered systems designed to simulate natural wetlands to exploit the water purification functional value for human use and benefits. Constructed wetlands consist of former upland environments that have been modified to create poorly drained soils and wetlands flora and fauna for the primary purpose of contaminant or pollutant removal from wastewaters or runoff (Hammer, 1992).

Constructed wetlands can provide many of the water quality improvement functions of natural wetlands with the advantage of control over location, design, and management to optimize those water quality functions. Constructed wetlands are not typically intended to replace all of the functions of natural wetlands, but emphasize certain features to maximize pollutant removal efficiency and to minimize point source and nonpoint source pollution prior to its entry into streams, natural wetlands, and other receiving waters. Wetlands created for habitat, water quantity, aesthetic and other functions as well as water quality functions typically call for different design considerations than those used solely for water quality improvement.

This tailored design approach to constructed systems generally makes them less suitable as wildlife habitat than natural wetlands. Nevertheless, constructed wetlands are often designed with ancillary wildlife values in mind, for example, incorporating open water for waterfowl usage. While species diversity of vegetation and microflora and fauna are lower in treatment wetlands, bird usage can be higher than that in adjacent natural wetlands because of the more eutrophic, and hence more productive, aquatic conditions in the loaded systems (McAllister, 1993, in Kadlec, 1995). A major concern with the use of constructed wetlands for wildlife habitat is the potential for concentrating accumulated pollutants up the food chain, with deleterious effects to birds and other consumers. While wildlife impacts have been observed in several instances with wetlands created for habitat (see the Wetlands Loss and Degradation section), these appear related to agricultural irrigation return flows in the West or hazardous waste site releases (Knight, 1993). So far, no similar problems are documented for constructed treatment wetlands (Kadlec, 1995; Knight, 1993), but the potential for harm exists with some metals and other compounds (Knight, 1993), and the issue requires continued evaluation.

Constructed wetlands are becoming an increasingly common method for treatment of all forms of water pollution, including confined animal wastewater, cropland runoff, urban stormwater, septic tank effluent, municipal wastewater effluent, acid mine drainage, industrial process waters, and landfill leachate (Kadlec and Knight, 1996; Kadlec, 1995: Bastian and Hammer, 1993). The beginnings of constructed wetland technology are dated to the 1950's in Germany for municipal wastewater treatment (Brix, 1994). This use is the most established and advanced, with hundreds of systems in place in Europe and the United States (Kadlec and Knight, 1996; Kadlec, 1995; Brown and Reed, 1994; Brix, 1994; Bastian and Hammer, 1993). Most constructed wetlands installed to date are used for advanced (nutrient reduction) treatment of municipal wastewater, with a large number also in place for secondary (solids and BOD) wastewater treatment. Use of these systems for primary wastewater treatment without prior or adequate settling and solids removal quickly overloads them and degrades performance capabilities, and is largely avoided (Cronk, 1995; Reaves et al., 1994). Other than primary wastewater uses, the range of potential applications for constructed wetlands is great and the record of actual applications is rapidly expanding.

Performance of constructed wetlands is good for a number of pollutants. In general, the greatest and most consistent reductions have been those of suspended solids, BOD, and fecal coliforms, with common discharge values of 10-20 mg/l for the first two and 50-100 fecal colonies/100 ml (Hammer, 1992). Phosphorus and nitrogen reductions are typically good, but less than efficiencies in the first three categories given the same conditions, with nitrogen usually more efficiently and consistently reduced than phosphorus. Strong nutrient reductions generally require greater area or lower application rates than do the first three constituents. Metals and some synthetic organic chemicals can also be reduced effectively, but results are more variable (Kadlec and Knight, 1996; Kadlec, 1995; Brown and Reed, 1994; Brix, 1994; Bastian and Hammer, 1993; Reed and Brown, 1992).

While construction costs can vary significantly, constructed wetlands provide treatment at significantly lower installation and maintenance costs than conventional municipal wastewater treatment options (Hammer, 1992). Hammer (1992) estimates that construction costs range from 1/10 to 1/2 of the cost of comparable conventional treatment systems. Constructed wetlands do, however, typically require significantly more land than conventional facilities. The major construction costs are associated with land purchase, pumping water to the wetlands, earthwork, possible impermeable liner, and planting (Kadlec, 1995; Reed et al., 1994; USEPA, 1988). Using data from municipal systems, Kadlec (1995) cites construction costs from 18 North American surface flow wetlands ranging from $4,500 to $203,000 per hectare (1994), with a mean of $68,000. Reed et al. (1994) give a range of $75,000 to $170,000 per hectare for the same type of system. Once up and running, operation and maintenance costs for constructed wetlands can be lower than for alternative treatment options, generally less than $1,000/ha/year (Kadlec, 1995), including the cost of pumping, mechanical maintenance, and pest control.

A number of information sources on constructed wetlands for water quality purposes are available (for non-water-quality-related wetland-creation references, see the Wetland Mitigation section). The first comprehensive synthesis of information on wastewater treatment wetlands was released at the end of 1995 by Kadlec and Knight (1996). Proceedings of conferences dealing exclusively with constructed wetlands for both point and nonpoint source treatment have been produced by Moshiri (1993), Cooper and Findlater (1990), Hammer (1989), and others, providing results, experience, and guidance on all aspects of conventional and alternative design, construction, operation, maintenance, and efficiencies. Other conferences have included coverage of constructed water quality wetlands, (Ross 1995); (Steele 1995). Perhaps the first conference dealing strictly with constructed wetlands for animal waste treatment was held in 1994 (DuBowy and Reaves, 1994). Schueler (1992) produced a guidance manual for constructed stormwater wetlands. A number of texts in the water quality/treatment area have also addressed constructed wetlands, such as those by Reed et al. (1995) and Novotny and Olem (1994). The USEPA and the Water Pollution Control Federation (WPCF) have both published design manuals which provide well-rounded basic coverage of design, performance, case studies with costs, and related issues for constructed wastewater wetlands (WPCF, 1990; USEPA, 1988).

For more detailed discussion of constructed wetlands, select the Best Management Practices for Non-Agricultural Nonpoint Source Pollution Control link in the Information Component subject index. Then choose the source type of interest from: industrial stormwater; mining/acid mine drainage; point sources; roads; septic systems; and urban stormwater options.

Constructed Wetlands for Animal Wastewater Treatment

Use of constructed wetland systems for confined animal wastewater has gained momentum in recent years, yet is still largely in the experimental stages. The major treatment concerns for these systems are BOD, ammonia, suspended solids, phosphorus, fecal coliforms, and sometimes metals added to feeds. The most problematic constituent seems to be ammonia; because of very high influent BOD levels, practically the entire wetland water column is essentially anoxic, inhibiting the aerobic nitrification step that must take place before denitrification and gaseous nitrogen release can occur. Very good nitrogen removals can occur with prior dilution or some form of aeration.

Although animal wastewater systems can borrow much from the municipal wastewater experience, an important difference is the need to keep capital costs and operation requirements to a minimum on the farm compared to municipal constraints. Also, minimizing wetland acreage is not as much a driving force with animal producers, since they often have significant area dedicated to lagoon waste land application, and wetlands can replace much of that disposal need in a fraction of the area. Municipal wastewater wetland design efforts toward increasing technical sophistication to maximize efficiency and minimize land requirements are of little assistance to animal facility operators.

Constructed Wetlands for Nonpoint Source Treatment

Wetlands constructed to treat stormwater runoff must be designed somewhat differently than wastewater wetlands. These constructed nonpoint source (NPS) wetlands can provide high removal efficiencies for stormwater pollutants and can be used to reduce stormwater runoff peak discharge rates. Constructed stormwater and other NPS treatment wetlands that mimic natural systems have been successful at many sites (Bingham, 1994; Mitsch and Gosselink, 1993). Constructed wetlands can also providing a pleasing natural area. Wetlands are highly valued by many landowners and can serve as attractive centerpieces to developments and recreation areas; wetlands also typically increase property values (Shaver, 1992; Schueler, 1987). Constructed wetland systems can provide ground water recharge in the area, thus lessening the impact of impervious surfaces. This recharge can also provide a ground water subsidy to the surficial aquifer, which can benefit local vegetative communities and decrease irrigation needs.

Runoff-driven wetlands by nature experience highly variable inputs, both hydrologically and in terms of pollutant loads. As a result, pollutant removal efficiency data are challenging to collect, are often collected using varying methods, often approximate in terms of accurately representing overall loading and removal efficiency, and are ultimately highly variable both within sites and between research efforts. Overall, NPS wetlands show much more variable performance than wastewater and other constant- source constructed wetlands.

In a recent literature review of constructed systems for agricultural NPS treatment, an important information gap (common to virtually all constructed wetland studies) was a dearth of information on long-term removal efficiencies (Osmond et al., 1995). The average length of a constructed wetland study was a little more than one year, following less than a year of preliminary loading. Decreasing efficiencies with time were observed in more than one experiment, and commonly recognized as a long-term possibility.

Nutrients, sediments, pathogens, metals and organic chemicals are pollutants typically removed by NPS constructed wetlands. Suspended solids removal in NPS constructed wetlands is generally greater than 60%; total nitrogen removal ranges from 25 to 76%; metals removal is variable, but lead generally shows at least 75% reduction; and phosphorus removal ranges from 30 to 90%, with an average of 50% (Bingham, 1994; Schueler et al., 1992). NPS constructed wetlands may release dissolved phosphorus because of improper design, including reliance on biotic activity for removal of phosphorus (D'Angelo and Reddy, 1994; Oberts and Osgood, 1991).

The use of constructed wetlands for stormwater treatment is still an emerging technology, hence there are no widely accepted design criteria. However, certain general design considerations do exist. It is important to first drop stormwater inflow velocities and provide opportunity for initial sediment deposition and solids removal using facilities that can be periodically maintained and that minimize the likelihood of entraining deposited sediment in subsequent inflows (Landers and Knuth, 1991; Oberts and Osgood, 1991). It is important to provide for the removal of oil and grease and floatable debris, preferably in the pre-treatment basin. The basin's outfall can be fitted with some form of skimmer or other means to retain floating matter (Palmer and Hunt, 1989). It is important to maximize the hydraulic residence time and the distribution of inflows over the treatment area, avoiding designs that may allow for hydraulic short-circuiting. Emergent macrophytic vegetation plays a key role, intimately linked with that of the sediment biota, by providing attachment sites for periphyton, by physically filtering flows, and by serving as a major storage vector for carbon and nutrients, an energy source for sediment microbial metabolism, and a gas exchange vector between sediments and air. Thus, it is important to design for a substantial native emergent vegetative component. Anaerobic sediment conditions should be ensured to allow for long-term burial of organic matter and phosphorus. A controlled rate of discharge is the last major physical design feature. While an adjustable outfall may seem desirable for fine-tuning system performance, regulatory agencies often require a fixed design to preclude subsequent inappropriate modifications to this key feature. Plants must be chosen to withstand the pollutant loading and the frequent fluctuation in water depth associated with the design treatment volume. It is advisable to consult a wetlands botanist to choose the proper vegetation.

Florida Administrative Code 40C-42, the stormwater rule used by the St. Johns River Water Management District, recommends that a constructed wetland for stormwater have less than 70% open water, a residence time of at least 14 days, and inlet structures designed to minimize turbidity and maximize settling of sediments (Palmer and Hunt, 1989). Storage capacity should be twice the capacity of an average storm event and drawdowns should be conducted to stabilize bottom sediments and reduce the re-release of orthophosphorus from the benthic sediments (Maristany et al., 1989; Esry et al., 1989).

For agricultural NPS runoff, researchers in Maine have developed and tested a multi-step constructed "nutrient/sediment control system" for cropland runoff (Reed et al., 1995; Higgins et al., 1993), and a number of such systems have now been installed around the state. Components of the system include, in sequence: a sediment basin; a level spreader, which disperses flows across an overland grass filter; the filter, which provides fine sediment and nutrient removal; an emergent marsh that grades into open water, primarily for nutrient removal; and a final grass filter to capture solids and nutrients in the form of algae that is produced in the pond. These systems have removed 90-100% of suspended solids, 85-100% of total phosphorus, 90-100% of BOD, and 80-90% of total nitrogen from potato field runoff in northern Maine (Hammer, 1992).

Constructed Wetlands for Mine Drainage

Acid mine drainage (AMD) is a major water pollutant associated with various types of mining operations, especially coal mining. AMD characteristically has low pH and high concentrations of iron, sulfate, and trace metals. Conventional treatment of acid mine drainage with alkaline reagents is "active" in nature, costly, and must be continued indefinitely (Skouson et al., 1994; Brodie et al., 1993).

The use of constructed wetlands for treatment of AMD is a "passive" technology, and provides a potential alternative to the conventional, active methods of chemical treatment. Thriving wetland communities have been observed despite acid mine drainage inputs. Closer inspection has revealed that outflow from such wetlands was of higher pH and did not contain, or contained only low concentrations of iron, sulfate and trace metals. This rapidly led to use of constructed wetlands to treat acid mine drainage (Skousen et al., 1994). It is estimated that over 400 wetlands are now in use in the U.S. for treatment of acid mine drainage (Weider, 1994).

Wetlands treatment of AMD is still an emerging technology. Treatment effectiveness has been variable to date, and long-term treatment effectiveness data do not yet exist. Hence, there are no widely accepted design criteria. The characteristics of AMD appear to present greater design challenges than more conventional applications. Some of the AMD removal processes initially thought to occur in wetlands were not evident when detailed research was completed (Vile and Weider, 1993; Skousen et al., 1994). Nonetheless, technical understanding of AMD wetland treatment issues is improving. The two major AMD contaminants from operations that encounter pyrite, including coal mine operations, are characteristically acidity and metals, usually iron (Fe) and manganese (Mn). Significant metals removal can take place through physical/chemical cation exchange and complexation with organic matter, both of which occur in the substrate. This physical filtering function is ultimately finite, and saturation of all available sites will occur (Weider, 1994; Gambrell, 1994; Skousen et al., 1994; Stark et al., 1994; Richardson, 1985). On the other hand, oxidation/reduction reactions yielding precipitation occur in wetlands and can provide a major sink for metals. However, the biological and chemical processes that "treat" the metals and the acidity are pH dependent. If the pH of inflow is less than 3, the wetland will not function (USDI, 1990). A calcium source, such as limestone, must be added regularly to constructed wetlands to regulate pH (Weider, 1994; McMillen et al., 1994). If the pH of the acid mine drainage and the wetland can be raised to 6.0, and if loading is less than 3g/m2/day, retention of metals can remain effective (Weider, 1994; USDI, 1990).

Research shows that an aerated vertical-flow constructed wetland is very effective in manganese removal (McMillen et al., 1994). The system causes an increase in the pH of the mine drainage when the inflow water infiltrates through the soil, a filter layer, and a limestone gravel layer. Manganese precipitates with the limestone (McMillen et al., 1994; Weider, 1994), although it is not normally precipitated in natural wetland processes (McMillen et al., 1994).

If constructed wetland management goals include wildlife habitat, pH must be greater then 3.5-4.0, and the concentrations of heavy metals in the water and sediments must not be toxic (Lacki et al., 1992).

More information regarding constructed wetlands for acid mine drainage can be found through the "Best Management Practices for Non-Agricultural Nonpoint Source Pollution Control" link in the Education Component subject index. Once there, choose the mining/acid mine drainage source type option.

Return to Wetlands Menu Page