United States Environmental Research EPA/600/
Environmental Protection Laboratory September 1991
Agency Corvallis, OR 97333
Research and Development
Workshop Proceedings
The Role of Created and
Natural Wetlands in
Controlling Nonpoint
Source Pollution
-------
Workshop Proceedings
The Role of Created and Natural Wetlands
in Controlling Nonpoint Source Pollution
10-11 June 1991
Arlington, Virginia
Sponsored by:
U.S. Environmental Protection Agency
Office of Research and Development
and
Office of Wetlands, Oceans, and Watersheds
Proceedings Editor:
Mr. Richard K. Olson
ManTech Environmental Technology, Inc.
USEPA Environmental Research Laboratory
200 SW 35th Street
Corvallis, Oregon 97333
Technical Editor:
Ms. Kay Marshall
Technical Resources, Inc.
3202 Tower Oaks Blvd. Suite 200
Rockville, Maryland 20852
The preparation of this Proceedings has been funded by the U.S. Environmental Protection
Agency. It was prepared at the EPA Environmental Research Laboratory in Corvallis, Oregon,
through contract number 68-C8-0006 to ManTech Environmental Technology, Inc., and
contract number 68-C0-0021 to Technical Resources, Inc. It has been subjected to peer and
administrative review and approved for publication. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
-------
LIST OF PAPERS AND AUTHORS
Evaluating the role of created and natural wetlands in controlling nonpoint source pollution
Mr. Richard K. Olson
ManTech Environmental Technology, Inc.
USEPA Environmental Research Laboratory
200 SW 35th Street
Corvallis, OR 97333
Introduction to nonpoint source pollution and wetland mitigation
Dr. Lawrence A. Baker
Water Resources Research Center
University of Minnesota
c/o USEPA Environmental Research Laboratory
200 SW 35th Street
Corvallis, OR 97333
Landscape design and the role of created, restored, and natural riparian wetlands in controlling
nonpoint source pollution
Dr. William Mitsch
School of Natural Resources
Ohio State University
2020 Coffey Road
Columbus, OH 43210
Designing constructed wetlands systems to treat agricultural nonpoint source pollution
Dr. Donald Hammer
Tennessee Valley Authority
2F 67B Old City Hall Building
Knoxville, TN 37902-1499
Developing design guidelines for constructed wetlands to remove pesticides from agricultural
runoff
Dr. John H. Rodgers, Jr.
Dr. Arthur Dunn
Department of Biology
Biological Field Station
University of Mississippi
University, MS 38677
-------
Ancillary benefits and potential problems with the use of wetlands for nonpoint source pollution
control
Dr. Robert Knight
CH2M Hill
7201 NW 11th Place
PO Box 1647
Gainesville, FL 32602-1647
Regulations and policies relating to the use of wetlands for nonpoint source pollution control
Ms. Sherri Fields
Office of Wetlands, Oceans, and Watersheds
US Environmental Protection Agency
401 M St, SW
Washington, D.C. 20460
The role of wetland water quality standards in nonpoint source pollution control strategies
Ms. Doreen Robb
Office of Wetlands, Oceans, and Watersheds
US Environmental Protection Agency
401 M St, SW
Washington, D.C. 20460
Recommendations for research to develop guidelines for the use of wetlands to control rural
nonpoint source pollution
Dr. Arnold van der Valk
Department of Botany
Iowa State University
Ames, 1A 50011-1020
Dr. Robert Jolly
Department of Economics
Iowa State University
Ames, IA 50011-1020
Research and information needs related to nonpoint source pollution and wetlands in the
watershed: An EPA perspective
Ms. Beverly Ethridge
US Environmental Protection Agency
1445 Ross Avenue, Suite 1200
Dallas, TX 75202-2733
Mr. Richard K. Olson
ManTech Environmental Technology, Inc.
USEPA Environmental Research Laboratory
200 SW 35th Street
Corvallis, OR 97333
-------
Federal programs for wetland restoration and use of wetlands for nonpoint source pollution
control
Dr. Gene Whitaker
US Fish and Wildlife Service
Division of Habitat Conservation
440 N. Fairfax Drive, Suite 400
Arlington, VA 22003
Dr. Charles R. Terrell
Ecological Sciences Division
Soil Conservation Service
P.O. Box 2890
Washington, D.C. 20013
-------
EVALUATING THE ROLE OF CREATED AND NATURAL WETLANDS
IN CONTROLLING NONPOINT SOURCE POLLUTION
Richard K. Olson
ManTech Environmental Technology, Inc.
USEPA Environmental Research Laboratory
200 SW 35th Street
Corvallis, Oregon 97333
ABSTRACT
Nonpoint source (NPS) pollution control and wetlands protection are two overlapping scientific
and policy issues of the U.S. Environmental Protection Agency. Created, restored, and natural
wetlands can contribute significantly to watershed water quality, but at the same time must be
protected from degradation by NPS pollution. Effective use of wetlands in NPS control requires
an integrated landscape approach including consideration of social, economic, and government
policy issues as well as scientific knowledge.
This article has been prepared with funding from the U.S. Environmental Protection Agency. It
was prepared at the EPA Environmental Research Laboratory in Corvallis, Oregon, through
contract #68-C8-0006 to ManTech Environmental Technology, Inc. It has been subjected to the
Agency's peer and administrative review and approved for publication.
1
-------
INTRODUCTION
This Proceedings addresses two issues that are major environmental concerns and major policy issues
for the U.S. Environmental Protection Agency (EPA):
1. Nonpoint source (NPS) pollution contributes over 65% of the total pollution load to U.S.
inland surface waters (US EPA, 1989). Sources include urban stormwater, diffuse
agricultural runoff from pastures and row crops, concentrated agricultural wastes from
feedlots, runoff from building sites, forestry activities, and drainage from mining
activities.
2. Over half of the wetlands in the lower 48 states have been lost during the past 200
years (Dahl, 1990), with some states losing more than 85 percent of their wetlands.
Remaining wetlands are frequently degraded through physical alteration, hydrologic
modification, and exposure to pollutants.
Addressing these two issues in an integrated manner makes sense in terms of both science and policy.
Wetlands occupy depressions in the landscape and therefore are often recipients of waterborne
pollutants. This exposure, coupled with the inherent ability of wetlands to sequester or transform
many pollutants, gives them an important role in water quality improvement. This role can be
consciously used in strategies to control NPS pollution by creating, restoring, or preserving wetlands
in appropriate locations in the landscape. The other side of this issue is that wetlands can be
degraded by NPS pollution. Preventing degradation of the full range of wetland functions may be
at odds with maximizing their water quality functions.
The policy linkages result from Sections of the Clean Water Act (CWA) (33 U.S.C. 12S1 as
ammended) that give EPA responsibilities in wetland protection and NPS pollution control (US
EPA, 1990), Under the CWA, most wetlands are considered to be "waters of the U.S." (Bastian et
2
-------
al., 1989) and are included within the CWA's objective to "restore and maintain the chemical,
physical, and biological integrity of the Nation's waters." Section 404 of the CWA regulates the
discharge of dredge and fill materials, and Sections 401 and 402 the discharge of waterborne
pollutants into "waters of the U.S..." Under Section 404, restoration or creation of wetlands may
be required to mitigate wetland losses.
Section 319 provides a mechanism for integrating Federal and State programs for controlling NPS
pollution. States are required to perform, under EPA oversight and approval, assessments of the
status of NPS pollution, and to develop management programs to control NPS pollution. Wetland
protection, creation and restoration may be included in State 319 programs. Thus, the 404
program and related efforts may protect and restore wetlands and their water quality functions,
to the benefit of NPS control programs. The NPS programs can protect wetlands from
degradation by pollutants, while also identifying areas where wetland creation and protection will
optimize wetland water quality functions.
WORKSHOP
In order to further define the scientific and policy linkages between NPS pollution and wetlands
issues, the EPA Office of Research and Development and the Office of Wetlands, Oceans and
Watersheds held a workshop on 10-11 June, 1991, in Arlington, Virginia. Within the overall
theme of evaluating the role of created and natural wetlands in the control of rural NPS
pollution, the workshop objectives were:
1. To review the state of knowledge; and
2. To identify research needs and approaches for developing guidelines for the
inclusion of wetlands in NPS control strategies.
3
-------
The focus of the workshop was rural NPS pollution, exclusive of acid mine drainage. Urban
stormwater and acid mine drainage are both important pollution sources for which wetlands can
provide treatment. However, urban stormwater not only differs somewhat from rural NPS in its
chemical constituents, but by definition occurs in a landscape from which many of the options
for creation and use of wetlands have been lost. Acid mine drainage is very different chemically
from most rural NPS pollution, and its treatment by wetlands requires consideration of a number
of different issues.
Restricting the workshop scope in this way still left for consideration a variety of nonpoint
sources ranging in scale from local (e.g., swine farms) to watershed and regional, and a large
number of scientific and policy questions concerning the use of wetlands to treat NPS pollution.
Presentations during the first part of the workshop provided background on these issues, and
papers corresponding to each presentation are included in this volume.
WORKSHOP RESULTS
Participants in the workshop included staff from EPA wetlands and NPS programs;
representatives of the U.S. Department of Agriculture, U.S. Fish and Wildlife Service, and the
U.S. Army Corps of Engineers; and wetlands scientists from universities and consulting firms.
This diversity of backgrounds and perspectives led to a corresponding diversity of ideas and
opinions. A number of themes emerged, however, as unifying threads throughout the discussions
and the papers in this volume. These include:
• Natural wetlands should not be used as wastewater treatment systems. In most cases,
natural wetlands are considered "waters of the U.S.," and are entitled under the CWA to
protection from degradation by NPS pollution. Natural wetlands do function within the
watershed to improve water quality, and protection or restoration of wetlands to maintain
or enhance water quality are acceptable practices. However, NPS pollutants should not be
4
-------
intentionally diverted to these wetlands, and wetlands receiving NPS loadings that will
degrade the wetland should be protected by establishing upland buffer strips or other best
management practices (BMPs).
• Wetlands must be part of an integrated landscape approach to NPS control. Created,
restored, and natural wetlands can contribute significantly to watershed water quality, but
they must be sited correctly and not be overloaded. Wetlands cannot be expected to
compensate for insufficient use of BMPs such as conservation tillage, grassed waterways,
and exclusion of livestock from riparian areas.
• The technical and scientific issues involved in defining the role of wetlands in NPS
control will be relatively easy to resolve compared to the social and economic issues.
Large-scale wetland creation or restoration efforts are expensive. Owners of key
restoration sites may be unwilling to participate. Watersheds are a scientifically logical
unit for NPS/wetlands programs, but rarely correspond with existing administrative units
(e.g., farms or counties).
• Knowledge of technical issues is uneven. Although more research is needed, design
criteria for constructed wetland treatment systems have been fairly well worked out.
Methods manuals for wetland creation and restoration are being developed by several
Federal agencies, and much is known about the fate and effects of nutrients in wetlands.
Topics with less sufficient information include water quality functions at the landscape
scale. For example, improved models of sources and movement of NPS pollutants, and
the role of wetlands in altering that movement, are needed to guide siting decisions for
wetland creation and restoration. The fate and effects of toxics (e.g., pesticides) in
wetlands are not well known. Watershed-level demonstrations of the effects of wetland
5
-------
restoration and creation on water quality are lacking, but are needed both as a framework
for research programs and as a technology transfer tool.
CONCLUSIONS
The combination of wetlands and NPS pollution makes for complex scientific and policy issues.
Technical issues must be addressed within the broader social and economic context if the research
results are to contribute to improvements in water quality and wetlands protection. Mechanisms
need to be developed to provide a strong link between science and policy development.
Successful implementation of NPS control strategies involving wetlands will require participation
of citizens, landowners, scientists, and government officials at every stage of the process.
6
-------
REFERENCES
Bastian, R. K., P. E. Shanaghan, and B. P. Thompson, 1989. Use of wetlands for municipal
wastewater treatment and disposal—regulatory issues and EPA policies, pp. 265-278. In: D. A.
Hammer (ed)., Constructed Wetlands for Wastewater Treatment: Municipal, Industrial and
Agricultural. Lewis Publishers, Inc., Chelsea, MI.
Dahl, T. E., 1990. Wetlands Losses in the United States: 1780's to 1980's. U.S. Department of
the Interior, Fish and Wildlife Service, Washington, DC. 21 pp.
U.S. Environmental Protection Agency, 1989. Focus on nonpoint source pollution. The
Information Broker, Office of Water Regulations and Standards, Nonpoint Sources Control
Branch, Washington, DC., November, 1989.
U.S. Environmental Protection Agency, 1990. National Guidance: Wetlands and Nonpoint Source
Control Programs. Office of Water Regulations and Standards, and Office of Wetlands
Protection, Washington, DC.
7
-------
INTRODUCTION TO NONPOINT SOURCE POLLUTION AND WETLAND MITIGATION
Lawrence A. Baker
Water Resources Research Center
University of Minnesota
c/o USE PA Environmental Research Laboratory
200 SW 35th Street
Corvallis, Oregon 97333
ABSTRACT
Nonpoint source (NPS) pollution is the major cause of impairment of U.S. surface waters. The
dominant source of NPS pollution is agricultural activity, and "traditional" pollutants—nutrients,
sediments, and pathogens—are the main detrimental constituents. Erosion from cropland has been
declining and is expected to decline further in the 1990s, but it is unclear how this will translate
into changes in sediment yields in streams. Pollution by nitrogen is of particular concern in
eutrophication of estuaries, as a contaminant of groundwater, and as an acidifying agent in
atmospheric deposition. Nitrogen fertilizer and emissions of nitrous oxides are major
contributors to the problem. The outlook on pesticides is mixed: bans on organochlorine
pesticides in the 1970s have resulted in decreasing concentrations in fish tissue; however,
herbicides are now a problem for some surface and groundwater sources of drinking water,
especially in the Upper Midwest. Metals in NPS pollution are primarily a concern in mining
areas and in urban runoff. Declining use of leaded gasoline has resulted in decreased lead in fish
tissues, sediments, and surface waters around the nation. New directions in controlling NPS
pollution include: (1) a greater emphasis on risk assessment, (2) a move toward regulatory or
quasi-regulatory approaches, and (3) a trend toward source reduction. The potential for using
wetlands to control agricultural NPS pollution is discussed by contrasting cropland runoff with
secondary wastewater effluent.
The preparation of this article has been funded by the U.S. Environmental Protection Agency.
This document has been prepared at the U.S. EPA Environmental Research Laboratory in
Corvallis, Oregon, through cooperative agreement CR8I3999 with the University of Minnesota.
It has been subjected to the Agency's peer and administrative review and approved for
publication. Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
1
-------
INTRODUCTION
During the past 20 years, most of our effort to control water pollution has been directed at
reducing point source discharges to surface waters. Today, 144 million people, twice as many as
in 1972, are served by municipal wastewater treatment plants that provide treatment at the
secondary level or better (U.S. EPA, 1990a). Less than 1 percent of municipal wastewater is now
discharged with no treatment. This upgrading of sewage treatment plants, generally from
primary to secondary treatment, has resulted in a 46 percent reduction in the discharge of
oxygen-consuming pollutants from sewage treatment plants between 1972 and 1982. Had these
improvements not occurred, discharges of oxygen-consuming pollutants would have increased by
191 percent due to population growth (ASIWPCA, 1985). But having made progress in this area,
we are left with the problem of nonpoint sources of pollution—contaminated runoff from urban
areas, agricultural fields, animal feedlots, roadways, abandoned mines, silviculture, and
construction activities. Nonpoint source (NPS) pollution has been identified as the major
remaining cause of surface water impairment (ASIWPCA, 1985; U.S. EPA, 1989), but resources
allocated for control of NPS pollution account for only about 4 percent of our national water
pollution control expenses (Farber and Rutledge, 1988; Figure 1).
This paper presents an overview of the status of NPS pollution in the United States and briefly
compares the potential of wetlands for removing pollutants from secondary wastewater with their
potential for reducing pollutant loadings from cropland. It begins with a review of several
national assessments, including recent reports developed by the U.S. Environmental Protection
Agency (EPA) and the states as mandated by sections 305(b) and 319 of the Clean Water Act.
This is followed by a closer examination of several types of NPS pollution—sediments, nutrients
(particularly nitrates), pesticides, salinity, and metals. Where possible, trends for several
pollutants are examined to address the question: Is NPS pollution getting better or worse? New
directions in the control and monitoring of NPS pollution are discussed briefly. The second part
of the paper considers several issues related to the use of constructed or managed wetlands to
weat NPS pollution. This section emphasizes some fundamental differences between the use of
wetlands for tertiary treatment of municipal wastewater, an area in which we have considerable
2
-------
experience, and their use in mitigating NPS pollution from rural lands, particularly agricultural
land, an area in which we have little experience.
NONPOINT SOURCE POLLUTION: STATUS AND TRENDS
About 30 percent of assessed U.S. surface waters do not "fully support" their designated uses
(U.S. EPA, 1990a; Table 1). The EPA has concluded that for roughly two-thirds of impaired
waters, the cause of impairment is NPS pollution (U.S. EPA, 1986; Figure 2). The EPA's most
recent biennial report on NPS pollution (U.S. EPA, 1990b), which is summarized in Figure 3 and
Table 2, reports NPS pollution impacts for 206,179 river miles, 5,300,000 acres of lakes, and
5,800 square miles of estuary. One conclusion reached by these reports is that the traditional
pollutants, particularly nutrients and sediments, are the primary causes of surface water
impairment.
Although the methodologies used in these reports differ considerably, they concur in the finding
that agriculture is the largest single cause of use impairment in assessed rivers and lakes (Table
2). A recent National Academy of Sciences report on alternative agriculture (NRC, 1989)
emphasizing the negative influence of current national agricultural policies on environmental
problems has further heightened interest in agricultural pollution.
Other sources of NPS pollution are more important than agricultural sources in several areas of
the country, although they contribute to <10 percent of the impairment of the assessed national
aquatic resources. These sources include urban runoff, land disposal, hydromodification, and
mining. A substantial percentage of systems impaired by NPS pollution have an unknown source
(Table 2).
The methodologies used in these reports present several problems: (1) only a small portion of the
total resource was assessed, (2) there is no consistent methodology for designating water uses (e.g.,
warmwater fisheries, coldwater fisheries, etc.), (3) there is no uniform methodology for
determining use attainment, and (4) because of reporting differences between the 305(b) report
(U.S. EPA, 1990a) and the 319 report (U.S. EPA, 1990b), results from the two cannot be directly
3
-------
compared (see U.S. EPA, 1990a). Nevertheless, these reports and the earlier assessment by the
ASIWPCA (1985) give a clear impression that NPS pollution, particularly agricultural pollution, is
a major source of water quality impairment in the United States.
Several recently completed national studies focus on particular pollutants and resources and give
us finer resolution of the scope of the NPS problem. These include two major studies of trends
in surface water quality at several hundred stream sites in the U.S. Geological Survey's (USGS)
National Stream Water Quality Accounting Network (NASQAN) and National Water Quality
Surveillance System (NWQSS), henceforth referred to in aggregate as the USGS network (Smith et
al., 1987; Lettenmaier et al., 1991); the National Contaminant Biomonitoring Survey conducted by
the U.S. Fish and Wildlife Service (USFWS) (Schmitt et al., 1990; Schmitt and Brumbaugh, 1990);
and the EPA's National Groundwater Pesticide Survey (U.S. EPA, 1990c).
SEDIMENTS
Sedimentation has been identified as a major source of NPS impairment of U.S. rivers and lakes
(Figure 3). Excessive sedimentation results in destruction of fish habitat, decreased recreational
use, and loss of water storage capacity. The U.S. Department of Agriculture (USDA) has
estimated that annual offsite costs of sediment derived from cropland erosion alone are $2-6
billion, with an additional $1 billion arising from loss in cropland productivity (USDA, 1987).
Erosion from agricultural lands has declined since the dust bowl years, from more than 3.5 billion
tons/year in 1938 to 3.0-3.1 billion tons/year in the 1980s (USDA, 1990). Lee reported that
cropland erosion declined by 12 percent between 1982 and 1987 (Lee, 1990). This occurred
primarily as a result of decreased erosion from land that was continuously cropped throughout the
study period. In comparison, conversion of land (cropped land to noncropped land) was
relatively unimportant in the overall decline in erosion during 1982-1987. The use of
conservation tillage increased during the 1980s, from around 40 million acres in 1980 to 88
million acres in 1988, when it accounted for 30 percent of all cropland. Conservation tillage is
expected to increase in the future (USDA, 1990) and when combined with expanded implemen-
tation of other conservation compliance programs should result in further reductions in cropland
erosion during the 1990s (USDA, 1990).
4
-------
Does declining erosion from agricultural land result in improved conditions in rivers and lakes?
From a water quality standpoint, the relationship between gross erosion (defined as a loss of soil
from a parcel of land) and sediment yield (defined as the loss of suspended solids from the
watershed) is complex; changes in erosion do not necessarily translate simply into changes in
sediment yield. This discrepancy is best illustrated by a classic historical analysis of erosion and
sediment yield in the agricultural watershed of Coon Creek, Wisconsin (Trimble, 1981). During
the period 1853-1938, annual sheet and rill erosion was 630 x 10s tons. This amount declined to
456 x 10s tons as a result of improved soil conservation, a decrease of 28 percent (Figure 4).
However, sediment yield remained virtually constant (40-42 x 10s tons/year) throughout the
period of record because stream bank and channel erosion increased when sheet and rill erosion
declined (Figure 4). From a broader perspective, it has been estimated that nationally 25 percent
of sediment yield occurs from stream bank erosion (Figure 7-10 in Van der Leeden et al., 1990).
Stream modifications, such as channelization and reservoir construction, also affect sediment
yield. For example, the completion of reservoirs on the Missouri River during the 1950s and
1960s was probably a major contributing factor in an observed 50 percent decline in sediment
discharges by the Mississippi River to the Gulf of Mexico (Meade and Parker, 1985). Finally,
other land uses account for about half the overall sediment loading to U.S. surface waters, placing
an upper limit on the potential for reducing sediment yields by controlling cropland erosion (Van
der Leeden et al., 1990). The USDA has estimated that a reduction of one billion tons in gross
erosion from the Nation's cropland would cause a 13-18 percent decline in the production of
sediments, with concomitant reductions in total phosphorus and total organic nitrogen of 5-7
percent and 7-9 percent, respectively (USDA, 1990).
With this perspective, it is perhaps not surprising that concentrations of suspended solids (SS)
have not changed at most USGS network stations since the 1970s (Table 3), despite declining
cropland erosion. Smith et al. (1987) and Lettenmaier et al. (1991) both showed that
concentrations of suspended solids did not change significantly at most of the USGS network
sites. Among remaining sites, roughly the same number have shown an increase in suspended
solids as have shown a decrease (Table 3). Smith et al. (1987) observed that increasing SS
concentrations tend to -occur in areas with high rates of soil erosion.
5
-------
NUTRIENTS
Phosphorus
Nutrients have been identified as the dominant cause of impairment by NPS pollution in lakes
and estuaries (Figure 3). In the majority of freshwater lakes, phosphorus (P) is the limiting
nutrient for algal growth (Chiaudani and Vighi, 1974; Miller et al., 1974), an exception being
lakes highly enriched in municipal wastewater, which typically have low N:P ratios (Miller et al.,
1974; Baker et al., 1985b). There is no national data base on trends in lake eutrophication. In the
USGS stream trend studies of Smith et al. (1987) and Lettenmaier et al. (1991), total phosphorus
(TP) concentrations paralleled SS concentrations: the majority of stations (75-80 percent)
exhibited no change; of those that changed, more decreased than increased (Table 3). However,
Smith et al. (1987) concluded that reductions in TP concentrations were probably associated with
declines in point sources. For the Great Lakes, municipal point sources of phosphorus were
reduced by 51-67 percent between 1975 and 1985 (CEQ, 1990), resulting in major reductions in
lake phosphorus levels. Further reductions will depend upon reductions in NPS phosphorus
loadings, which in 1986 comprised 59-88 percent of total phosphorus loadings in the Great Lakes
(CEQ, 1990).
Where increases in phosphorus occurred, Smith et al. (1987) found statistical associations between
TP increases and measures of fertilized acreage and cattle population. Thus, there is some
suggestion that phosphorus loadings from agricultural areas are increasing.
^itroeen
NPS nitrogen pollution is important in at least three arenas: (1) eutrophication of surface waters,
(2) groundwater contamination in agricultural areas, and (3) acidification of forested watersheds.
Nitrogen is particularly important in regard to the eutrophication of estuaries, since algal growth
in estuaries is usually nitrogen limited (reviewed in Stoddard, 1991). An interesting aspect of this
problem in the Chesapeake Bay is that atmospheric deposition appears to be a major source of
nitrogen (Stoddard, 1991). Direct deposition of nitrogen (NH4+ and NOs" ) to the water surface
of the Bay accounts for about 12 percent of the total N loading; additional atmospheric nitrogen
loading comes from deposition to the watershed. Although about 90 percent of the nitrogen
deposition to the Chesapeake Bay watershed is retained in plants and soils, the remainder passes
6
-------
through the watershed and into the Bay. This accounts for an additional 22 percent of the Bay's
nitrogen budget, which together with direct deposition to the Bay surface means that 35 percent
of the Bay's total nitrogen loading is derived from atmospheric deposition (Stoddard, 1991; Table
4). Other nonpoint sources (e.g., animal waste, fertilizer fluxes, etc.) account for 16 percent of
total nitrogen inputs. These findings suggest that reductions of nitrous oxide emissions, which
are generated primarily by automobiles and other vehicles, would ameliorate the eutrophication
problem in the Chesapeake Bay.
One of the greatest NPS pollution concerns in the Midwest and other agricultural areas is nitrate
contamination of groundwater. Nielsen and Lee (1987) used USGS well water records in
conjunction with an analysis of sensitivity factors to estimate the potential for nitrate
contamination in groundwater. In 474 counties out of 1,663 agricultural counties with adequate
data, measured nitrate levels were >3 mg/L in more than 25 percent of the wells; 87 of these
counties had measured nitrate levels greater than the EPA's maximum contaminant level (MCL)
of 10 mg/L in more than 25 percent of the wells. These counties are located primarily in the
Great Plains, the Corn Belt, the Southwest, and the Northwest. More recently, the National
Pesticide Survey, a statistically designed survey of pesticides and nitrate in drinking water wells
of the United States, showed that more than half had detectable levels of nitrate; in 2.4 percent of
domestic rural wells (254,000 wells) and 1.2 percent of community supply wells (1,130 wells),
nitrate exceeded 10 mg/L (U.S. EPA, 1990c; Table 5).
Finally, elevated atmospheric nitrogen inputs may also result in nitrogen saturation of forested
watersheds. Nitrate levels are elevated in atmospheric deposition throughout the eastern United
States, occurring in a S04s" :N03* ratio of about 2:1 (reviewed in Baker et al., 1991). Nitrate is
generally thought to be efficiently retained in watersheds, neutralizing inputs of HNO, (Baker et
al., 1991). However, recent evidence suggests that high inputs of nitrate may exceed the demand
by watershed plants and microbes, resulting in nitrate saturation and subsequent breakthrough of
HNOs into streams and lakes (Stoddard, 1991). This process is potentially important because the
additional inputs of HNOs would exacerbate lake and stream acidification. There is conclusive
evidence of nitrogen saturation in parts of northern Europe. However, in the United States,
where atmospheric nitrate inputs are generally lower, the evidence for nitrogen saturation is
7
-------
sketchy. Some evidence that nitrogen saturation may be occurring has been found in the Catskill
and Adirondack Mountains of New York and in the Mid-Appalachian and Smoky Mountains in
the Southeast (Stoddard, 1991).
Is nitrogen pollution getting better or worse? Several lines of evidence suggest that it is getting
worse, at least for surface waters. First, surface water concentrations increased in the USGS
network stations: far more stations exhibited increased nitrate concentrations during 1974-1981
than decreases, and far more stations exhibited increases in total nitrogen during 1978-1984 than
declines (Table 3). It is perhaps even more important that rivers discharging into estuaries on the
East Coast and the Gulf of Mexico exhibited increases in nitrate concentrations of 20-46 percent
(Smith et al., 1987). The probable cause of this increase is the increased use of nitrogen
fertilizers, from around 6.5 million tons in 1970 to 9-12 million tons in the 1980s (USDA, 1990).
Smith et al. (1987) found strong associations between increased nitrate and measures of
agricultural activity, supporting the contention that fertilizer nitrogen is a major cause of the
uptrend.
SALINITY
Excessive salinity is a problem primarily in arid regions, where it can lower crop yields, add to
water treatment costs, and increase the maintenance cost of water supply systems. The USGS
network studies (Table 2) show that 2-3 times more stations have experienced increases in
chloride (a good surrogate for salinity) than have experienced declines. In the western United
States, salinity problems are caused largely by irrigated agriculture, which both adds salt load and
causes reduction in flow through evaporation. In the Colorado River Basin, damages due to
salinity are conservatively estimated at S311 million/year (USDI, 1989). Salinity in the Colorado
Basin declined during the early 1980s due to unusually high flows, but salinity levels are expected
to increase over the next 20 years as a result of continued development and a return to more
normal hydrologic conditions. Salinity control has reduced the salt load by 156,000 tons/year, but
an additional million-ton reduction will be needed by the year 2010 to keep salinity at Imperial
Dam (near the U.S.-Mexico border) below the criterion level of 879 mg/L.
8
-------
Increased chloride concentrations in many eastern U.S. streams are probably associated with
increased use of road salt (Smith et al., 1987). In several regions of the National Stream Survey (a
randomized sampling of streams in the eastern United States) 15-30 percent of the lower stream
reaches had chloride concentrations more than 10 times higher than would be expected from
natural sources (A. T. Herlihy, Utah State University, pers. comm.), indicating widespread
chloride contamination in this part of the country.
PESTICIDES
As with nitrate, there is widespread public concern about contamination of groundwater by
pesticides in agricultural areas. Of particular interest in recent years has been the potential for
herbicide contamination, because herbicide use has increased fourfold since 1966. However, con-
cerns about increased use of herbicides in conjunction with expanded use of conservation tillage
may be overstated, as several studies indicate that total herbicide use may not change appreciably
with the acceptance of conservation measures (Fawcett, 1987; Logan, 1987). A compilation of
existing data by Williams et al. (1988), summarized in NRC (1989), shows that 46 pesticides have
been detected in groundwater from 26 states; atrazine, aldicarb, and alachlor were the most
commonly detected. Nielsen and Lee (1987) concluded that the potential for pesticide contam-
ination was greatest in the Eastern Seaboard, the Gulf States, and the Upper Midwest. In the
EPA's National Pesticide Survey (U.S. EPA, 1990c), 12 of the 126 pesticides and pesticide
metabolites analyzed were found in detectable quantities. One or more of these pesticides were
detected in 10.4 percent of the nation's community water supply wells and in 4.2 percent of rural
domestic wells. None of the community water supply wells and only 0.6 percent of the rural
domestic wells exceeded maximum contaminant levels (MCL) or health advisory limits (HAL) for
pesticides. Five pesticides exceeded the MCL/HAL limits in rural domestic wells: atrazine,
alachlor, DCPA acid metabolites, lindane, and EDB (Table 5). Thus, from a national perspective,
a small number of pesticides has contributed significantly to contamination of groundwater. Two
of these pesticides, DBCP and EDB, are no longer in use in the United States.
In the Midwest, surface water concentrations of herbicides commonly exceed proposed stan-
dards, at least during part of the year. Baker et al. (1985a) reported seasonally elevated
concentrations of atrazine in two Ohio rivers; peak concentrations of several pesticides in finished
9
-------
tap water derived from surface water exceeded what are now proposed MCL/HAL limits. In an
ongoing study, the USGS measured herbicide concentrations in 150 rivers in the Midwest, an area
that accounts for about 60 percent of the pesticides (mostly herbicides) used in the nation
(Goolsby and Thurman, 1990). Concentrations of all measured herbicides were low during the
pre-application period, but increased following herbicide application (May-June). During this
period, nearly half the samples had atrazine concentrations above the HAL limit of 3 fig/L
(median « 3.8 /
-------
criteria for the protection of aquatic life in more than half the collected runoff samples (U.S.
EPA, 1983). Probably of greater significance in terms of the total number of miles of impaired
streams and rivers is the leaching of metals from abandoned coal mines in the Appalachian region
and abandoned metals mines in the West (Moore and Luoma, 1990). Studies by the U.S. Fish and
Wildlife Survey and the Appalachian Regional Commission (reviewed in Herlihy et a!., 1990)
indicate that there are approximately 10,000 km of acidic mine drainage streams in the
Appalachian region. Pollution by metals, particularly selenium, is a problem in some wetlands
receiving irrigation return flow, but the problem appears to be limited to closed-basin systems in
the West (Deason, 1989).
Undoubtedly, the most important NPS problem of metals has been the atmospheric deposition of
lead resulting from combustion of leaded gasoline. Perhaps one of the most successful efforts to
control NPS pollution was the removal of lead from gasoline during the early 1970s. The decline
in lead use resulted in decreased concentrations in surface waters at the USGS network stations
(Smith et al., 1987; Lettenmaier et al., 1991), in Mississippi River delta sediments (Treffry et al.,
1985), and in fish tissues in the National Contaminant Biomonitoring Program (Figure 7).
NEW DIRECTIONS
In the nearly 20 years since passage of the Clean Water Act, there has been uneven progress in
controlling NPS pollution. However, past experiences have led to a recent refocusing of efforts
in several areas, including (1) development of a risk assessment approach to pollution control, (2)
a trend towards source reduction as a key to reducing pollution, and (3) a shift from voluntary
efforts to regulation and monetary incentives.
RISK ASSESSMENT FRAMEWORK
EPA is gradually developing a risk assessment approach to dealing with pollution problems, so
that the effort to control a particular type of pollution will have some relation to the risk posed to
humans or ecosystems (SAB, 1990). As a starting point, there is a need to move beyond chemical
monitoring to measures of ecological condition (U.S. EPA, 1989; Hughes and Larsen, 1988;
Larsen et al., 1988). This is a high priority if we are to determine impacts from diffuse sources,
-------
which are often caused by multiple pollutants, commonly in conjunction with habitat alteration.
States are currently required to develop biological criteria ("biocriteria") under Sections 303 and
304 of the Clean Water Act, although research to support this effort has been limited (GAO,
1990).
A second component of risk assessment is comparative risk analysis. On a national scale, the
305(b) and 319 program reports are intended to provide a comparative risk assessment of water
quality problems, although as noted earlier, there are severe deficiencies in the methodologies
used in these reports and in most other monitoring programs used for assessment of pollution
(Hren et al., 1990). EPA's Ecological Monitoring and Assessment Program (EMAP; U.S. EPA,
1991), now under development, will provide a statistically based representation of ecological
conditions in the Nation's waters. The goal of this national-scale sampling effort is to evaluate
status and trends in ecological conditions, but EMAP also is designed to address specific issues,
such as causes and sources of impairment by NPS pollution, through more detailed diagnostic
studies. These studies should be useful in developing policies that would provide the greatest
improvement in ecological condition per dollar invested.
At the watershed scale, a key component of risk assessment is targeting major sources of
pollution. This trend has emerged from a consensus that many past watershed efforts have taken
a fragmented and inefficient approach to controlling NPS pollution (NWQEP, 1988; Water Quality
2000, 1990; U.S. EPA, 1989; Humenick et al,, 1987). Efforts to geographically target NPS
pollution reduction are based on a growing recognition that small parts of watersheds often
contribute a disproportionately large share of NPS pollutants and that these areas need to be
targeted to reduce overall watershed pollutant loadings (NWQEP, 1988; CBP, 1990). A good
example is the USDA's targeting of highly erodible lands for inclusion in its Conservation
Reserve Program. Efforts to improve watershed NPS models should greatly enhance prospects for
geographic targeting.
EMPHASIS ON SOURCE REDUCTION
Increasingly, we are finding that the best way to reduce pollution is not to produce it. As
previously discussed, source reduction has proven effective in reducing NPS inputs of lead and
12
-------
organochlorine pesticides. Bans on detergents containing phosphorus have proven to be a cost-
effective method for reducing phosphorus ievels by 50 percent in wastewater effluent. In
farming practices, reduction of fertilizer inputs has been identified as a cost-effective approach
for reducing nutrients in runoff (Magleby et al.r 1990). Approaches to prevent groundwater
pollution, such as wellhead protection programs, are vastly less expensive than remediation
efforts.
SHIFT FROM VOLUNTEER APPROACHES TO REGULATORY AND MONETARY
INCENTIVES
Low voluntary participation rates have been a problem in many watershed NPS pollution
reduction efforts; voluntary efforts alone are now thought to be insufficient to control major NPS
pollution (U.S. EPA, 1989; NWQEP, 1988; CBP, 1990; GAO, 1990). Controls on local land use,
animal waste management, and other measures have been suggested as regulatory complements to
conventional NPS abatement methods (GAO, 1990; CBP, 1990). Between regulatory control and
purely voluntary efforts are quasi-regulatory efforts and cost-share approaches, such as the
conservation compliance measures (the "sodbuster" and "swampbuster" provisions) in the 1990
Farm Bill.
There also has been a gradual move from the "command-control" strategy of pollution reduction,
which dictates specific pollution reduction methods, toward the use of market incentives. Some
market incentive tools, such as the trading of pollution credits, may not be feasible for control of
NPS pollution. More realistic may be "pollution taxes" on fertilizers (now in use in Iowa) and
pesticides, which would tend to increase the efficiency of application and thereby decrease the
likelihood of these chemicals moving offsite as pollutants. There has also been considerable
interest in eliminating disincentives for reducing agricultural pollution caused by the current crop
price support system used in the United States, which tends to maximize crop production at the
expense of environmental concerns (NRC, 1989).
USING WETLANDS TO CONTROL NONPOINT SOURCE POLLUTION
The goal of the papers in this volume is to evaluate the prospects for using constructed or natural
wetlands to ameliorate the effects of NPS pollution in rural areas. There is relatively little
13
-------
published information on the efficacy of using wetlands to control rural NPS pollution but
considerable experience in using wetlands as tertiary treatment for municipal wastewater (Kadlec
and Alford, 1989; Knight, 1990; Richardson, 1985). Therefore, a useful starting place in this
exercise is to address the question: How is cropland runoff different from the effluent of a
secondary wastewater treatment plant, and what are the implications of these differences in
considering the potential of wetlands to remove pollutants from cropland runoff? For this
comparison, it is useful to employ data from streams in agricultural areas of the "Corn Belt and
Dairy Region," which were compiled by Omernik (1976). This region has a high potential for
wetlands control of NPS impacts, having both extensive agricultural areas and numerous wetlands
and sites suitable for wetland restoration and creation. Data on secondary treatment plant
effluent is based on a nationwide compilation of nutrient levels in wastewater treatment plant
effluent (Gakstatter et al., 1978).
As a first consideration, note that the chemical composition of secondary effluent from
wastewater treatment plants is relatively constant, with standard deviations around 10 percent of
the mean (Table 6). Although data are not available for making a similar statistical calculation of
uncertainty for cropland runoff, it appears that annual pollutant loadings from cropland may vary
by one order of magnitude (reviewed in Novotny and Chesters, 1981). Thus, one could design a
wetlands system for tertiary treatment of municipal wastewater from published effluent data, but
site-specific loading data would be needed to design an efficient wetland treatment system to
treat cropland runoff. Second, whereas total nitrogen concentrations in secondary effluent and
cropland runoff are similar, total phosphorus concentrations in cropland runoff are only 1 /20 of
the concentrations in secondary effluent. Therefore, wastewater effluent is strongly nitrogen
limited, with a mean N:P ratio of 2.4:1, whereas cropland runoff is generally phosphorus limited,
with a mean N:P ratio of 31:1 (Table 6). Third, both N and P occur primarily in the soluble form
in wastewater treatment plant effluent; this is also true for nitrogen in cropland runoff (the
predominance of soluble N increases with increasing fertilizer nitrogen use by agriculture; see
Omernik, 1977) but not for phosphorus, which occurs primarily in association with particulate
matter. Fourth, concentrations of suspended solids are usually <30 mg/L in secondary treatment
plant effluent, compared with values of 100-1000 mg/L for cropland. Perhaps the most critical
difference, however, is that pollutant loading from croplands is largely event-driven, with
14
-------
extreme variations in both flow and pollutant concentrations, whereas the flow and composition
of effluent from a wastewater treatment plant is relatively stable. Concentrations of suspended
solids can vary by 2-3 orders of magnitude in cropland runoff within a year, with peak flows
carrying the great majority of sediment loading (Figure 8).
Because of these differences, the relative importance of pollution removal processes would be
different in wetlands receiving cropland runoff than in wetlands receiving tertiary wastewater.
First, sedimentation of particles would be a major process for removing suspended solids and
phosphorus in wetlands receiving cropland runoff, since phosphorus is present largely in the
particulate form. The high N:P ratios in cropland runoff suggest that plant uptake would be
relatively more important as a long-term phosphorus retention process in agricultural wetlands
than in wastewater systems, where phosphorus is grossly oversupplied relative to plant nutrient
requirements. In contrast, adsorption of phosphorus to soils is a major mechanism of phosphorus
removal in wetlands receiving wastewater effluent (Richardson, 1985); sedimentation is
unimportant because phosphorus occurs as a soluble species. Denitrification would be an
important mechanism in wetlands receiving cropland runoff, perhaps even more efficient than in
wastewater wetlands (Kadlec and Ahord, 1989; Knight, 1990), since inorganic nitrogen is
oversupplied relative to phosphorus in cropland runoff. A compilation of nutrient balances
(Nixon and Lee, 1986) indicates that typically 20-80 percent of the phosphorus and 10-90 percent
of the nitrogen are retained in natural wetlands.
These observations suggest that the design limitation for maximum pollutant removal would be
based upon the need to retain sediments during peak flows. In this regard, the design of
constructed wetlands for cropland runoff would have more in common with wetlands designed
for treatment of urban runoff (Barten, 1986; Meiorin, 1989; Weidenbacher and Willenbring, 1984;
also see Loucks, 1990) than with those designed for treatment of secondary wastewater.
In exploring the feasibility of using wetlands for efficient removal of sediment-bound pollutants,
the first consideration is the size of wetland needed to control sediment movement. Very limited
information from urban systems suggests that the ratio of wetlands to drainage area needed to
achieve a reasonable reduction of suspended solids is on the order of 1:20 (Barten, 1986; Meiorin,
IS
-------
1989; Weidenbacher and Willenbring, 1984). A second consideration is the design lifetime. With
sedimentation rates in wetlands on the order of 1 cm/yr (Mclntyre and Naney, 1991), it appears
that the effective lifetime of a wetlands system designed for efficient sediment removal may be
on the order of a few decades. Several papers in this volume discuss the additional research
needed to develop design criteria for rural wetlands treatment systems.
Pesticides in cropland runoff are also a concern for cropland wetland systems. Of particular
interest is atrazine, which is the most heavily used herbicide in the United States. Goolsby and
Thurman (1990) showed that for medium-sized watersheds in the Midwest (800-2,000 km2), the
median post-application concentration of atrazine in streams was 10 pg/L, with a 75 percent
quartile value of 16 vg/L. This has two implications. First, atrazine is phytotoxic and is
probably toxic to algae at concentrations of J-10 fig/L (deNoyelles et al., 1982; Kosinski and
Merkle, 1984; Johnson, 1986), although it has a fairly short half life (ca. weeks; Huckins et al.,
1986). It is, therefore, reasonable to suspect that atrazine would inhibit growth of algae in some
wetlands, at least during part of the year. Second, if a constructed wetland were to promote flow
to the groundwater system, there would be a possibility of causing groundwater contamination.
In the preceding few paragraphs, wetlands are regarded as engineered systems designed to remove
pollutants. There are other benefits to constructing wetlands in rural areas, such as habitat for
wildlife. As engineers and wetlands scientists move from using wetlands for tertiary wastewater
treatment into the area of rural NPS control, the ancillary benefits of wetland treatment systems
may equal or exceed the benefits of the pollutant removal function. Thus, the optimal design of
a rural constructed wetland may not necessarily be the one that provides the maximum pollutant
removal or even the most cost-effective pollutant removal but the one that balances the pollutant
removal function of wetlands with other ecological functions.
16
-------
REFERENCES
ASIWPCA, 1985. America's Clean Water: The States' Evaluation of Progress, 1972-1982.
Association of State and Interstate Water Pollution Control Administrators, Washington, DC.
Baker, D. B., K. A. Krieger, R. P. Richards and J. W. Kramer, 1985a. Gross erosion rates,
sediment yields, and nutrient yields for Lake Erie tributaries: Implications for targeting. In:
Perspectives on Nonpoint Source Pollution, Proceedings of a National Conference, Kansas City,
MO. EPA/440/5-85/001.
Baker, L. A., P. L. Brezonik, and C. Kratzer, 1985b. Nutrient loading models for Florida lakes,
pp. 253-258. In: J. F. Taggart and L. M. Moore (eds.), Lake and Reservoir Management, Volume
1. North American Lake Management Society, Washington, DC.
Baker, L. A., J. M. Eilers, R. B. Cook, P. R. Kaufmann, and A. T. Herlihy, 1991. Interregional
comparisons of surface water chemistry and biogeochemical processes, pp. 567-613. In: D. F.
Charles (ed.), Acidic Deposition and Aquatic Ecosystems: Regional Case Studies. Springer-
Verlag, New York, NY.
Barten, J., 1986. Nutrient removal from urban storm water by wetland filtration: The Clear Lake
restoration project. Lake and Reservoir Management, 2: 297-305.
Capel, P. D., 1991. Atmospheric deposition of herbicides on the Mid-Continental United States
(abstract). Eos, 71: 1329.
CBP, 1990. Report and Recommendations of the Nonpoint Source Evaluation Panel. Chesapeake
Bay Program CBP/TRS 56/91. USEPA, Washington, DC. 28 pp.
CEQ, 1990. Environmental Quality: 20th Annual Report. Council on Environmental Quality,
Executive Office of the President, Washington, DC.
17
-------
Chiaudani, G. and M. Vighi, 1974. The N:P ratio and tests with Selanastrum to predict
eutrophication in lakes. Water Research, 8: 1063-1069.
Deason, J. P., 1989. Impacts of irrigation drainwater on wetlands, pp. 127-138. In: Wetlands:
Concerns and Successes. American Water Resources Association, Bethesda, MD.
deNoyelles, R., W. D. Kettle, and D. E. Sinn, 1982. The responses of plankton communities in
experimental ponds to atrazine, the most heavily used pesticide in the United States. Ecology, 63:
1285-1293.
Farber, K. D. and G. L. Rutledge, 1988. Pollution abatement and control expenses, 1983-86.
Survey of Current Business, May 1988.
Fawcett, R. S., 1987. Overview of pest management for conservation tillage systems. In: T. J.
Logan, J. M, Davidson, J. L. Baker, and M. R. Overcash (eds.), Effects of Conservation Tillage
on Groundwater Quality: Nitrates and Pesticides. Lewis Publishers, Inc. Chelsea, MI.
Fisher, D., J. Ceraso, T. Mathew, and M. Oppenheimer, 1988. Polluted Coastal Waters: The Role
of Acid Rain. Environmental Defense Fund, New York, NY.
GAO, 1990. Greater EPA Leadership Needed to Reduce Nonpoint Source Pollution.
GAO/RCED-91-10. GAO, Washington, DC.
Gakstatter, J. H., M. O. Allum, S. E. Dominiguez, and M. R. Crouse, 1978. A survey of
phosphorus and nitrogen levels in treated municipal wastewater. Journal of the Water Pollution
Control Federation, 50: 718-722.
Goolsby, D. A. and E. M. Thurman, 1990. Herbicides in rivers and streams of the Upper
Midwestern United States. In: Proceedings of the Forty-sixth Annual Meeting of the Upper
Mississippi River Conservation Committee, Bettendorf, IA.
18
-------
Grover, R, 1991. Nature, transport, and fate of airborne residues, pp. 90-117. In: R. Grover
and A. J. Cessna (eds.), Environmental Chemistry of Herbicides, Vol. 2. CRC Press, Boca Raton,
FL.
Herlihy, A. T., P. R. Kaufmann, and M. E. Mitch, 1990. Regional estimates of acid mine
drainage impact on streams in the mid-Atlantic and southeastern United States. Water, Air, and
Soil Pollution, 50: 91-107.
Hindall, S. M., 1975. Measurement and Prediction of Sediment Yields in Wisconsin Streams.
Water Resources Investigation Report, U.S. Geological Survey, Reston, VA. pp. 54-75.
Hren, J., C. J. O. Childress, J. M. Norris, T. H. Chaney, and D. N. Myers, 1990. Regional water
quality: Evaluation of data for assessing conditions and trends. Environmental Science and
Technology, 24: 1122-1127.
Huckins, J. N., J. D. Petty, and D. C. England, 1986. Distribution and impact of trifluralin,
atrazine, and fonofos residues in microcosms simulating a northern prairie wetland.
Chemosphere, 15: 563-588.
Hughes, R. M. and D. P. Larsen, 1988. Ecoregions: an approach to surface water protection.
Journal of the Water Pollution Control Federation, 60: 486-493.
Humenik, F. J., M. D. Smolen, and S. A. Dressing, 1987. Pollution from nonpoint sources.
Environmental Science and Technology, 21: 737-742.
Johnson, B. T., 1986. Potential impact of selected agricultural chemical contaminants on a
northern prairie wetland: a microcosm evaluation. Environmental Toxicology and Chemistry, 5:
473-485.
Kadlec, R. H. and H. Alvord, 1989. Mechanisms of water quality improvement in wetland
treatment systems. In: D. W. Fisk (ed.), Wetlands: Concerns and Successes. American Water
Resources Association, Bethesda, MD.
19
-------
Knight, R. L., 1990. Wetland systems. In Natural Systems for Wastewater Treatment, Manual of
Practices FD-16. Water Pollution Control Federation.
Kosinksi, R. J. and M. G. Merkle, 1984. The effect of four terrestrial herbicides on the
productivity of artificial stream algal communities. Journal of Environmental Quality, 13: 75-82.
Larsen, D. P., D. R. Dudley, and R. M. Hughes, 198S. A regional approach to assess attainable
water quality: An Ohio case study. Journal of Soil and Water Conservation, 43: 171-176.
Lee, L. K.., 1990, The dynamics of declining soil erosion rates. Journal of Soil and Water
Conservation, 45: 622-624.
Lettenmaier, D. P., E. R. Hooper, C. Wagoner, and K. B. Faris, 1991. Trends in stream water
quality in the continental United States, 1978-1987. Water Resources Research, 27: 327-340.
Logan, T. J., 1987. An assessment of Great Lakes tillage practices and their potential impact on
water quality. In: T. J. Logan, J. M. Davidson, J. L. Baker, and M. R. Overcash (eds.), Effects of
Conservation Tillage on Groundwater Quality: Nitrates and Pesticides. Lewis Publishers, Inc.,
Chelsea, MI.
Loucks, O. L., 1990. Restoration of the pulse control function of wetlands and its relationship to
water quality objectives pp. 467-477. In: J. A. Kusler and M. E. Kentula (eds.), Wetland
Creation and Restoration: The Status of the Science. Island Press, Washington, DC.
Magleby, R. S., S. Piper, and C. E. Young, 1990. Economic insights on nonpoint pollution
control and the Rural Clean Water Program, pp. 63-69. In: Making Nonpoint Pollution Control
Work: Proceedings of a National Conference, St. Louis, MO. National Association of
Conservation Districts, League City, TX.
Mclntyre, S. C. and J. W. Naney, 1991. Sediment deposition in a forested inland wetland with a
steep-farmed watershed. Journal of Soil and Water Conservation, 46: 64-66.
20
-------
Meade, R. H. and R. S. Parker, 1985. Sediments in Rivers of the United States. National Water
Summary, 1984. Water Supply Paper 2275. U.S. Geological Survey, Reston, VA.
Meiorin, E. C., 1989. Urban runoff treatment in a fresh/brackish water marsh in Fremont,
California pp. 677-685. In: D. A. Hammer (ed.), Constructed Wetlands for Wastewater Treatment:
Municipal, Industrial, and Agricultural. Lewis Publishers, Inc., Chelsea, MI.
Miller, W. E., T. E. Maloney, and J. C. Greene, 1974. Algal productivity in 49 lake waters as
determined by algal assays. Water Research, 8: 667-679.
Moore, J. N. and S. N. Looma, 1990. Hazardous wastes from large-scale metal extraction: A case
study. Environmental Science and Technology, 24: 1278-1285.
Nielsen, E. G. and L. K. Lee, 1987. The Magnitude and Costs of Groundwater Contamination
from Agricultural Chemicals: A National Perspective. AGES870318. National Resources
Economics Division, Economic Research Service, USDA, Washington, DC.
Nixon, S. W. and V. Lee, 1986. Wetlands and Water Quality: A Regional Review of Recent
Research in the United States on the Role of Freshwater and Saltwater Wetlands as Sources,
Sinks, and Transformers of Nitrogen, Phosphorus, and Various Heavy Metals. Technical Report
Y-86-2, U.S. Army Corps of Engineers, Washington, DC. 229 pp.
Novotny, V. and G. Chesters, 1981. Handbook of Nonpoint Pollution: Sources and Management.
Van Nostrand Reinhold Company, New York, NY.
NRC, 1989. Alternative Agriculture. Committee on the Role of Alternative Farming Methods in
Modern Production Agriculture, Board on Agriculture, National Research Council, National
Academy Press, Washington, DC. 448 pp.
21
-------
NWQEP, J988. 1987 Annual Report: Status of Agricultural Nonpoint Source Projects. National
Water Quality Evaluation Project, Office of Water Regulations and Standards Division, USEPA,
WH-585, Washington, DC.
Omernik, J. M., 1976. The Influence of Land Use on Stream Nutrient Levels. EPA/600/3-
76/014. USEPA Environmental Research Laboratory, Corvallis, OR.
Omernik, J. M., 1977. Nonpoint Source-Stream Nutrient Level Relationships: A Nationwide
Survey. EPA/600/3-77/105. USEPA Environmental Research Laboratory, Corvallis, OR.
Otterby, M. A., and C. A. Onstad, 1981. Average Sediment Yields in Minnesota. ARR-NC-8.
USDA Agricultural Research Service. 9 pp.
Rappaport, R. A., N. R. Urban, P. D. Capel, J. B. Baker, B. Looney, and S. J. Eisenreich, 1984.
"New" DDT inputs to North America: atmospheric deposition. Chemosphere, 14: 1167-1173.
Richardson, C. J., 1985. Mechanisms controlling phosphorus retention capacity in freshwater
wetlands. Science, 228: J 424-1427.
SAB, 1990. Reducing Risk: Setting Priorities and Strategies for Environmental Protection. SAB-
EC-90-021. Scientific Advisory Board, USEPA, Washington, DC.
Schmitt, C. J. and W. G. Brumbaugh, 1990. National contaminant biomonitoring program:
concentrations of arsenic, cadmium, lead, mercury, selenium, and zinc in U.S. freshwater fishes,
1976-1984. Archives of Environmental Toxicology, 19: 733-747.
Schmitt, C. J., J. L. Zajicek, and P. H. Peterman, 1990. National contaminant biomonitoring
program: residues of organochlorine chemicals in U. S. freshwater fish, 1976-1984. Archives of
Environmental Toxicology, 19: 748-781.
22
-------
Smith, R. A., R. B. Alexander, and M. G. Wolman, 1987. Water quality trends in the nation's
rivers. Science, 235: 1608-1615.
Stoddard, J., 1991. Aquatic effects of nitrogen oxides, pp. 10-120 - 10-239. In: Air Quality
Criteria for Oxides of Nitrogen: External Review Draft, August, 1991. EPA/600/8/91/0496-A.
USEPA.
Treffry, J. H., S. Metz, R. P. Trocine, and T. A. Nelson, 1985. A decline in lead transport by the
Mississippi River. Science, 230: 439-441.
Tyler, M. 1988. Contribution of Atmospheric Nitrate Deposition to Nitrate Loading in the
Chesapeake Bay. VERSAR, Inc., Columbia, MD.
Trimble, S. W., 1981. Changes in sediment storage in the Coon Creek basin, Driftless Area,
Wisconsin, 1853-1975. Science, 214: 181-183.
U.S. Department of Agriculture, 1987. Agricultural Resources: Cropland, Water and
Conservation Situation and Outlook Report. Economic Research Service, USDA, Washington,
DC. 39 pp.
U.S. Department of Agriculture, 1990. Agricultural Resources: Cropland, Water and
Conservation Situation and Outlook Report. Economic Research Service, USDA, Washington,
DC. 55 pp.
U.S. Department of the Interior, 1989. Quality of Water: Colorado River Basin. Progress Report
No. 14, USDI.
U.S. Environmental Protection Agency, 1983. Results of the Nationwide Urban Runoff Program,
Volume 1 - Final Report. WH-554. Water Planning Division, USEPA, Washington, DC.
23
-------
U.S. Environmental Protection Agency, 1986. National Water Quality Inventory: Report to
Congress. USEPA, Washington, DC.
U.S. Environmental Protection Agency, 1989. Nonpoint Sources: Agenda for the Future. WH-
556. USEPA, Office of Water, Washington, DC. 31 pp.
U.S. Environmental Protection Agency, 1990a. National Water Quality Inventory: 1988 Report to
Congress. EPA/440/4-90/003. USEPA, Washington, DC.
U.S. Environmental Protection Agency, 1990b. Managing Nonpoint Source Pollution: Final
Report to Congress on Section 319 of the Clean Water Act (1989). EPA/506/9-90. USEPA,
Washington, DC.
U.S. Environmental Protection Agency, 1990c. National Pesticide Survey: Project Summary.
USEPA, Washington, DC.
U.S. Environmental Protection Agency, 1991. EMAP Monitor, January 1991. Office of Research
and Development, USEPA, Washington, DC.
Van der Leeden, F., F. O. Troise, and D. K. Todd, 1990. The Water Encyclopedia. 2nd edition.
Lewis Publishers, Inc., Chelsea, MI.
Water Quality 2000, 1990. Water Quality 2000: Phase II Report, Problem Identification. Water
Quality 2000, Alexandria, VA.
Weidenbacher, W. D. and P. R. Willenbring, 1984. Limiting nutrient flux into an urban lake by
natural treatment and diversion. Lake and Reservoir Management, 3: 525-526.
Williams, W. M., P. W. Holdren, D. W. Parsons, and M. N. Lorber, 1988. Pesticides in Ground
Water Data Base: 1988 Interim Report. Office of Pesticide Programs, USEPA, Washington, DC.
24
-------
Table 1. Attainment of Designated Use for U.S. Surface Waters
Source: U.S. EPA, 1990a
Rivers and
streams*
Lakes*
Estuaries and
coastal waters'
Total national resource
1,800,000
39,400,000
35,198
Resource within
assessed states
1,150,482
22,347,961
not given
Assessed resource
519,413
16,314,012
26,676
Assessed waters meeting
use designation, %:
Fully supporting
Threatened*
Partially supporting
Not supporting
69.6
6.9
20.1
10.3
73.7
17.8
16.6
9.8
71.6
1.3
22.8
5.6
* Resource estimates are in miles. The total resource value is from ASIWPCA (1985). The
assessed resource is for 48 reporting states only.
b Resource estimate in acres. No primary reference given for the total lake resource. The
assessed resource is for 40 reporting states.
* Resource estimates in square miles; 23 out of 30 estuarine states reporting.
* The "threatened" category is a subset of "fully supporting" in the 305(b) report. Thus,
percentages sum to more than 100%.
25
-------
Table 2. Water Quality Impairment by Nonpoint Source Pollutants
Source: U.S. EPA, 1990b
Rivers* Lakes® Estuaries*
Nonpoint source
impacts reported 206,179 5.3 x 10 5,800
Degree of impact, %*
Nonsupport 52 42 54
Partial support 28 22 36
Threatened 20 36 10
Causes, %
Agriculture 41 23 7
Urban 4 6 11
Land disposal 3 4 8
Construction 2 2
Hydromodification 6 6
Mining 8 7 16
Natural 8 10
Others 3 16 43
In-place -- — 16
Unknown 23 21 4
* Resource estimate in miles. 40 states reporting impacts; 20 reported identified use impairment;
33 identified causes of NPS pollution.
& Resource estimates in acres. The number of states reporting impacts is not given; 18 states
identified the type of use impairment; 25 states identified causes of NPS pollution.
* Resource estimate in square miles. Thirteen states reporting impacts.
* EPA (1990b) includes "threatened" in the "impaired" category, in contrast to the 305(b) report
(EPA 1990a).
26
-------
Table 3. Trends of Selected Water Quality Variables in Major Rivers
of the United States
Sources: Smith et al., 1987 (for 1974-1981); Lettenmaier et al., 1991 (for 1978-1987)
1974-1981 1978-1987
Number of stations
Number of stations
Constituent
Median*
Increasing Decreaiing
No Change
Increasing
Decreaiing
No Change
Chloride
15
101
34
154
65
32
295
Suspended solids
67
43
39
194
12
19
121
Total P
0.13
43
50
288
12
69
308
Nitrate**
0.41
116
27
240
82
24
284
Lead
4
2
23
219
2
23
325
Cadmium
< 2
16
2
231
1
16
298
Arsenic
< 1
21
4
220
1
25
286
• Median concentrations are in mg/L for major common constituents and f%/L for trace metals
(lead, cadmium, and arsenic).
•*1978—1987 data are for total nitrogen.
27
-------
Table 4. Nitrogen Budget for the Chesapeake Bay*
Source: Stoddard (1991), modified from Tyler (1988) and Fisher et al. (1988)
Inputs to
Flux to
the watershed
the Bay
Fertilizer
11.3
0.4
Animal waste
13.9
0.4
Nitrogen deposition to land
11.4
1.1
Total for watershed
36.6
1.9
Point sources
-
2.4
Nitrogen deposition to bay surface
-
0.6
Total nitrogen input:
4.9
* Inputs and fluxes are in 10® moles/year.
28
-------
Table 5. Pesticides and Nitrate in Community Well Systems
and Rural Domestic Wells
Source: U.S. EPA, J 990c*
Constituent**
Use
% Detected
HAL/MCL
% above
HAL/MCL***
CWS
RDW
CWS
RDW
Nitrate
Fertilizer
52.1
57.0
10
1.2
2.4
Pesticides
Various
10.4
4.2
Various
0
0.6
Atrazine
Herbicide
1.7
0.7
3
X
DBCP
Nematode
0.4
0.4
0.2
--
X
Alachlor
herbicide
0
0.03
2
--
X
EDB
Insecticide
0
0.2
0.05
X
Lindane
Insecticide
0
0.1
0.2
--
X
DCPA
Herbicide
6.4
2.5
4,000
„
Prometon
7
0.5
2.6
100
Simazine
1
1.1
0.2
1.0
--
Hexachlorobenzene
Fungicide
0.5
0
1
»
Dinoseb
Herb./Fung.
0.03
0
7
...
Beniazon
Herbicide
0
0.1
20
—
* This was a statistically based survey of community water supply (CWS) wells and rural
domestic wells (RDW). During 1988-1990, 1,300 wells were sampled, representing an
estimated population of 94,600 community water supply wells and 10.5 million rural domestic
wells.
** Pesticide acronyms are as follows: EDB « ethylene dibromide; DBCP « l,2-dibromo-3-
chloropropane; DCPA - dimethyl tetrachloroterephthalate.
***For individual pesticides, percentages above HAL/MCL are not given. An "X" indicates that
some wells exceeded the HAL/MCL and a " indicates that none did. HAL » health
advisory limits; MCL - maximum contaminant levels.
29
-------
Table 6. Characteristics of Secondary Effluent from Municipal
Wastewater Treatment Plant, and Cropland Runoff
Secondary effluent*
Concentration
(mg/L)
Cornbelt cropland4
Concentration
(mg/L)
Loading
kg/ha/yr
Suspended solids*
10-30
50-1000
Total P
6.8 ± 0.4
0.14
0.31
Soluble P
5.3 ± 0.4
0.06
0.13
Total N
15.8 ± 1.2
4.4
9.5
Sol. inorg. N
8.4 ± 0.45
3.4
7.3
N:P
2.4
31.4
* From Gakstatter et al., 1978. Means and standard deviations represent effluent from 244
activated sludge treatment plants located throughout the country.
41 Source: Omernik, 1976. Total P and N loadings are means for "mostly agricultural" land (>
90% cropland) within the "Corn Belt and Dairy" region, based upon 80 sampled streams.
Soluble P and N are calculated as the product of the ratio of soluble/total nutrients, calculated
for "mostly agricultural" lands throughout the country (n » 91), and the total nutrient
concentrations in the Corn Belt and Dairy region.
* The range for suspended solids is from Table 1-3 of Novotny and Chesters (1981). The
estimate for cropland is a range for stream yields in the agricultural areas of Wisconsin and
Minnesota (Otterby and Onstad, 1981; Hindall, 1975).
30
-------
Municipal (53%)
'Wrf*'
Industrial (30%)
R&D (1%)
Other (10%)
Regulations & Monitoring (2%)
NPS Pollution (4%)
Figure 1
-------
LAKES
RIVERS
Combined Sewer
Overflows
1 Natural Causes
Industrial Point
Sources
Other/Unknown
~ Municipal
Point Sources
Nonpoint Sources
Figure 2
-------
LIST OF FIGURES
Figure I. The annual cost of water pollution control in the United States, 1986. Total
expenses in 1986 were $26.9 billion, calculated in 1982 dollars. Source: Farber
and Rutledge (1988).
Figure 2. Sources of pollution causing impairment of U.S. surface waters. Percentages for
estuaries are based on area, percentages for lakes are based on numbers of lakes,
and percentages for rivers are based on miles. All percentages are calculated on
the basis of the assessed resource. Source: U.S. EPA (1986).
Figure 3. Causes of surface water impairment by NPS pollution. Source: U.S. EPA (1990b).
Figure 4. Sediment budgets for Coon Creek, Wisconsin. The watershed has an area of 360
km2. Numbers are annual averages in 10s tons/yr. Source: Trimble (1981), as
modified in Meade and Parker (1985).
Figure 5. Herbicide concentrations during the post-application period in 150 Midwest rivers.
Median and quartiles shown. Source: Goolsby and Thurman (1990).
Figure 6. Trends in several organochlorine compounds in fish tissue in U.S. surface waters,
determined in the National Contaminant Biomonitoring Program. Fish tissue
concentrations are in /4g/g wet weight for whole fish. Source: Schmitt et al.
(1990).
Figure 7. Median lead levels in fish collected in the National Contaminant Biomonitoring
Program (source: Schmitt and Brumbaugh, 1990), compared with U.S. emissions of
lead (source: CEQ, 1990). Fish tissue concentrations are in jxg/g wet weight for
whole fish.
31
-------
Figure 8. Suspended solids and phosphorus loadings as a function of time for Honey Creek,
Ohio. Infrequent peak flow periods contribute disproportionately to the total
annual loading: > 90% of the suspended solids and phosphorus loading occurs in <
10% of the time. Source: Baker et al. (1985a).
32
-------
Sediments
Nutrients ~
Pathogens
Organic/D.O. i
Metals/pest./prior. \
Other
0 10 20 30 40 SO
% of Assessed Impaired Waters
Rivara
Eatuariaa
Lake*
Fiqure 3
-------
A. 1853-1938
Upland
sheet
and rill Upland
erosion gullies
630 ^ go
Sources of Sediment
Tributaries
46
Sediment
discharge
at mouth
>42
« 78 230 Lower
42 Middle valley
Hillslopes 269 Upland Tributary valley
valleys valleys
Sinks and Storage ol Sediment
B. 1938-1975
Upland
sheet
and rill Upland
erosion gullies
.456 ^ 71
Sources o! Sediment
T ributaries
36
30
Middle
Sediment
discharge
at mouth
40
Hillslopes 332
30 153 Lower
«2 Middle valley
Upland valley
valleys
Sinks and Storage of Sediment
Figure 4
-------
Alachlor
Atrazine
Cyanazine
Simazine
4 8 12
Concentration, ug/L
16
Figure 5
-------
Fish conc.
ug/g
1976 1978 1980
Year
1984
DDT
i—J Toxaphene
PCB
Chlordane
Figure 6
-------
Fish conc.
ug/a
Emissions
1000 metric T
0.2 i
Emissions
0.1 -
Year
Figure 7
-------
Percent of Annual Load
100
eo
60
40
20
-S- Sutpsndod solid*
"A" Totsl Phosphorus
20
40
Percent of Year
Figure 8
-------
LANDSCAPE DESIGN AND THE ROLE OF CREATED, RESTORED, AND NATURAL RIPARIAN
WETLANDS IN CONTROLLING NONPOINT SOURCE POLLUTION
William J. Mitsch
School of Natural Resources
The Ohio State University
2021 Coffey Road
Columbus, Ohio 43210
ABSTRACT
General design principles, landscape locations, and case studies of natural and constructed
riverine wetlands for the control of nonpoint source (NPS) water pollution are presented.
General design principles of wetland construction for NPS pollution control emphasize self-
design and minimum maintenance systems, with an emphasis on function over form and
biological form over rigid designs. These wetlands can be located as instream wetlands or as
floodplain riparian wetlands, can be located as several wetlands in upstream reaches or fewer in
downstream reaches of a watershed, and can be designed as terraced wetlands in steep terrain.
Case studies of a natural riparian wetland in southern Illinois, an instream wetland in a
downstream location in a northern Ohio watershed, and several constructed riparian wetlands in
northeastern Illinois demonstrate a wide range of sediment and phosphorus retention, with greater
efficiencies generally present in the constructed wetlands (63-96 percent retention of phosphorus)
than in natural wetlands (4-10 percent retention of phosphorus). By itself, this could be
misleading as the natural wetlands have much higher loading rates and actually retain an amount
of nutrients comparable to constructed wetlands (1-4 g-P m"2y_1).
1
-------
INTRODUCTION
The riparian wetlands of the United States, which once were connected to many of the streams,
rivers, and Great Lakes of the Nation, have for all intents and purposes been eliminated from the
landscape. With the wetland loss and their conversion to other land uses, our rivers and streams
no longer have the ability to cleanse themselves, and bodies of water such as the Great Lakes are
no longer buffered from upland regions. The net result has been poorer water quality,
particularly in the glaciated Midwest where nonpoint source (NPS) pollution in now pervasive. In
Midwestern states such as Ohio, Indiana, Illinois, and Iowa, where more than 80 percent of the
wetlands have been drained, partially in response to the Swamp Lands Acts of 1849, 1850, and
1860, water quality has been particularly degraded. With a 90 percent loss of wetlands, Ohio is
ranked below only California for the highest percent loss of wetlands from the 1780s to the 1980s
(Dahl, 1990). NPS water pollution problems are common in Ohio, and nutrients continue to flow
into Lake Erie to the north and the Ohio River to the south.
Much national attention is now focused on the design and construction of wetlands. In 1987, a
National Wetlands Policy Forum was convened by the Conservation Foundation at the request of
the U.S. Environmental Protection Agency (EPA) to investigate the issue of wetland management
in the United States (NWPF, 1988; Davis, 1989). The distinguished group of 20
members—including three governors, a State legislator, State and local agency heads, chief
executive officers of environmental groups and businesses, farmers, ranchers, and academic
experts—published a report that set significant goals for the Nation's remaining wetlands. The
report recommended a policy "to achieve no overall net loss of the nation's remaining wetlands
base and to create and restore wetlands, where feasible, to increase the quantity and quality of
the nation's wetland resource base" (NWPF, 1988).
2
-------
In his 1990 budget address to Congress, President Bush echoed the "no net loss" concept as a
national goal (Davis, 1989), shifting the activities of many agencies such as the Department of
Interior, the EPA, the U.S. Army Corps of Engineers, and the Department of Agriculture to
provide leadership toward a unified and seemingly simple goal. It was not anticipated that there
would be a complete halt of wetland loss in the United States when economic or political reasons
dictated otherwise, so implied in the "no net loss" concept is wetland construction and restoration.
This paper will present some general principles of ecological engineering for the proper design of
wetlands, some design ideas for the placement of wetlands in the landscape, and case studies of
three riparian wetland systems that have been evaluated for their role in controlling NPS
pollution.
ECOLOGICAL ENGINEERING OF WETLANDS
Ecological engineering combines basic and applied science for the restoration, design, and
construction of ecosystems, including wetlands (see Mitsch, 1988, 1991; Mitsch and Jorgensen,
1989 for details). The goals of ecological engineering and ecotechnology are: (1) the restoration
of ecosystems that have been substantially disturbed by human impacts such as environmental
pollution, climate change, or land disturbance; (2) the development of new sustainable ecosystems
that have both human and ecological value; and (3) the identification and protection of the life-
support value of existing ecosystems. Combining ecosystem function with human needs is the
foundation of ecological engineering (also called "ecotechnology"), presently defined as "the
design of human society with its natural environment for the benefit of both" (Mitsch and
Jorgensen, 1989). Ecological engineering provides approaches for conserving our natural
environment while at the same time adapting to and sometimes solving intractable environmental
pollution problems.
3
-------
CREATING AND RESTORING WETLANDS
The creation and restoration of wetlands need to be accomplished in ecologically sound ways.
Wetland creation refers to the construction of wetlands where they did not exist before. These
created wetlands are also called constructed wetlands or artificial wetlands, although the last term
is not preferred by many wetland scientists. No one has estimated the number of such wetlands
in the United States, but it is probably in the thousands. Wetland restoration refers to the
enhancement of existing wetlands, often with only the hydric soils as an indicator of former
wetlands.
Wetlands are constructed or restored in the landscape for a variety of reasons or objectives. The
three most popular reasons for wetland construction in the United States have been to treat
wastewater, to compensate a wetland loss elsewhere (mitigation), and to provide habitat for
wildlife. A more recent application under study in a number of locations around the country is
to control NPS runoff, especially from rural agricultural lands. Over the past 15 years, many
natural wetlands have been evaluated for that role as well. Wetlands built for the control of NPS
pollution (e.g., sediments and nutrients) are often part of a watershed or river floodplain
restoration project (Hey et al., 1989; Livingston, 1989; Newton, 1989; Mitsch, 1990; Mitsch and
Cronk, in press). The construction of wetlands for NPS pollution control contributes to two
national goals—cleaning up our Nation's waterways and adding to our Nation's wetlands reserves.
PRELIMINARY PRINCIPLES OF WETLAND DESIGN
Some of the principles of ecological engineering that could be applied to the construction and
restoration of wetlands for NPS pollution control are outlined below:
4
-------
1. Design the system for minimum maintenance. The system of plants, animals, microbes,
substrate, and water flows should be developed for self-maintenance and self-design
(Mitsch and Jorgensen, 1989; Odum, 1989).
2. Design a system that utilizes natural energies, such as potential energy of streams, as natural
subsidies to the system. Pulsing streams during Midwestern spring transport great quantities
of nutrients in relatively short periods.
3. Design the system with the landscape, not against it. Floods and droughts are to be
expected, not feared. Outbreak of plant diseases and invasion of alien species are often
symptomatic of other stresses and may indicate faulty design rather than ecosystem failure.
4. Design the system with multiple objectives, but identify at least one major objective and
several secondary objectives.
5. Design the system as an ecotone. This means including a buffer strip around the site, but it
also means that the wetland site itself is often a buffer system between upland and aquatic
systems.
6. Give the system time. Wetlands do not become functional overnight and several years may
elapse before nutrient retention or wildlife enhancement is optimal. Strategies that try to
short-circuit ecological succession or over-manage are doomed to failure.
7. Design the system for function, not for form. If initial plantings and animal introductions
fail but the overall function of the wetland—based on initial objectives—is intact, then the
wetland has not failed. Expect the unexpected.
8. Do not over-engineer wetland design with rectangular basins, rigid structures and channels,
and regular morphology. Ecological engineering recognizes that natural systems should be
mimicked to accommodate biological systems (Brooks, 1989).
-------
LOCATING WETLANDS IN THE LANDSCAPE
Wetlands are effective in retaining some nutrients and sediments from NPS pollution. Wetlands
should not be expected, however, to control all of the influx of sediments and nutrients from a
watershed, nor should the creation of one small wetland be expected to result in significant
improvements in downstream water quality. If wetlands are to be constructed in the watershed
for the control of NPS pollution, a variety of possible designs should be considered.
INSTREAM WETLANDS
Wetlands can, of course, be designed as instream systems by adding control structures to the
streams themselves or by impounding a distributary of the stream (Figure 1). Blocking an entire
stream is a reasonable alternative only in low-order streams, and it is not generally cost-effective.
This design is particularly vulnerable during flooding and might be very unpredictable in its
ultimate stability. It has the advantage, however, of potentially "treating" a significant portion of
the water that passes that point in the stream. Maintenance of the control structure and the
distributary might mean significant management commitments to this design.
RIPARIAN WETLANDS
The natural design for a riparian wetland primarily fed by a flooding stream (Figure 2) allows for
flood events of a river to deposit sediments and chemicals on a seasonal basis in the wetland.
Because there are both man-made and natural levees along major sections of streams, it is often
possible to create such a wetland with minimal construction. The wetland could be designed to
capture flooding water and sediments and slowly release the water back to the river after the
flood passes. This is the design of natural riparian wetlands in bottomland hardwood forest areas
(e.g., Mitsch et al., 1979). The wetland could also be designed to receive water from flooding and
retain it through the use of flap-gates.
6
-------
A riparian wetland fed by a pump (Figure 3) creates the most predictable hydrologic conditions
for the wetland, but at an extensive cost for equipment and maintenance. This design is being
used for some of the wetlands at the Des Plaines River Demonstration Project (see below), but
for the specific purpose of establishing experimental conditions for research. If it is anticipated
that the primary objective for building a constructed wetland is the development of a research
program to determine design parameters for future wetland construction in the basin, then a
series of wetlands fed by pumps is a good design. If other objectives are most important, then
the use of large pumps is usually not appropriate. Small pumps may be necessary to carry
wetlands through drought periods.
In a compromise between the designs in Figures 2 and 3, a wetland could be fed by diversion of
the stream at such a distance upstream of the wetland that it could effectively be fed by gravity
(Figure 4). In such a design, natural energies rather than pumps could be used, and natural
levees rather than impoundments would effectively hold the water on the floodplain.
UPSTREAM VERSUS DOWNSTREAM WETLANDS
The advantages of locating several small wetlands in the upper reaches of a watershed (but not in
the streams themselves) rather than fewer larger wetlands in the lower reaches should be
considered (Figure 5). Loucks (1989) argues that locating a greater number of low-cost wetlands
in the upper reaches of a watershed rather than building fewer high-cost wetlands in the lower
reaches offers a better strategy for wetlands to survive extreme events. However, a modeling
effort on flood control by Ogawa and Male (1983) suggested the opposite: the usefulness of
wetlands in decreasing flooding increases with the distance the wetland is downstream. Figure 6
shows a design where multiple wetlands are constructed in the landscape to intercept small
streams and drainage tiles. The main stream itself is not diverted, but the wetlands receive their
water, sediments, and nutrients from small tributaries, swales, and overland flow. More
7
-------
significantly, if tile drains can be located and broken upstream of their discharge into tributaries,
they can be very effective conduits for supplying adequate water to the wetlands. Because these
tile drains are often the sources of the highest concentrations of chemicals such as nitrates from
agricultural fields, the wetlands could be very effective in controlling certain types of NPS
pollution while creating a needed habitat in an agricultural setting.
WETLANDS IN STEEP TERRAIN
Wetlands are a phenomenon of naturally flat terrain. However, steeper terrain is often most
susceptible to high erosion and hence high contributions of suspended sediments and other
chemicals. One approach is to attempt to integrate "terraced" wetlands into the landscape (Figure
7). In this case, wetland basins are constructed as smaller basins that "stair-step" down steep
terrain. While there are some examples of these types of wetlands, particularly in the building of
acid mine drainage wetlands in the Appalachian mountains, few wetlands of this type have been
constructed for the control of NPS pollution.
CASE STUDIES-WETLANDS THAT CONTROL NONPOINT SOURCE POLLUTION
HERON POND, ILLINOIS-A NATURAL RIPARIAN WETLAND
Early work by Mitsch et al. (1977, 1979) on a riparian forested wetland in southern Illinois
demonstrated the ability of these systems to serve as nutrient and sediment traps. Heron Pond is
a 30 ha alluvial cypress swamp dominated by bald cypress (Taxodium distichum [L.] Rich.) and
water tupelo (Nyssa aquatica L.) in the northern extreme of the Mississippi Embayment (Figure
8). The wetland is highly productive, with a continual coat of duckweed on the water surface,
and is estimated to be flooded by the adjacent Cache River almost every year as illustrated in
Figure 2. In the case of this riparian wetland, approximately 447 g m"2 of sediments and about
3.6 g m"2 of phosphorus was retained by the wetland during a flood event (Table 1). Gross
8
-------
sedimentation rates were estimated to be 5,600 g m^y"1 from sediment traps, with most of that
due to allochthonous productivity and resuspension. The flux of phosphorus with flooding
represented the largest inflow of the nutrient during the study period. Our preliminary
phosphorus budget for the system suggested that the flooding river deposited about 33 times as
much phosphorous as entered the wetland through bulk precipitation and that the 30 ha swamp
retained about 0.4 percent of the total annual phosphorus flux of the river. For the year in which
measurements took place, the swamp retained about 10 times more phosphorus from the flooding
river than it returned by surface and groundwater to the river for the remainder of the year
(Mitsch et al., 1979).
OLD WOMAN CREEK, OHIO - A NATURAL INSTREAM WETLAND IN A COASTLINE
ENVIRONMENT
Old Woman Creek State Nature Preserve and National Estuarine Research Reserve is a coastal
wetland located adjacent to Lake Erie in Erie County, Ohio (Figure 9). The wetland itself is 30
ha and extends about 1 km south of the Lake Erie shoreline (see Mitsch 1988, 1989; Klarer and
Millie, 1989; Mitsch et al., 1989; Reeder, 1990; Mitsch and Reeder, 1991a,b) and is a good
example of an instream natural wetland such as the one shown in Figure 1. It is approximately
0.34 km wide at its widest point. Wetland depths may reach up to 3.6 m in the inlet stream
channel but are usually less than 0.5 m. Klarer and Millie (1989) estimated that the retention
time of the wetland varies between 24 hours (at peak flow) and 114 hours (at average flow). The
wetland has an outlet to Lake Erie that is often open but that can be closed for extended periods
by shifting sands forming a barrier beach. Rare but dramatic seiches on Lake Erie can reverse
the flow, causing lake water to spill into the wetland. Most of the aquatic habitats within the
wetland are open-water planktonic systems with about 30 percent of the wetland as floating-
leaved beds of American lotus (Nelumbo lutea). The major land use within the watershed (68.6
km^) is agricultural, and hence the wetland is a receiving system for significant loadings of NPS
9
-------
pollution, primarily in the spring. Sedimentation in the wetland was estimated to have been 0.76
mm y"1 prior to agricultural development in the early 1800s and more than 10 times that (10 mm
y"1) at present (Buchanan, 1982).
Table 2 contrasts rates of phosphorus retention in Old Woman Creek from empirical model
calculations, from field data collected in 1988 (Reeder, 1990; Mitsch and Reeder 1991a), and
from a simulation model developed for the system (Mitsch and Reeder, 1991b). The empirical
phosphorus retention model of Richardson and Nichols (1985) predicted 8-19 mg-P m'^T1 being
permanently buried in the wetland sediments under standard phosphorus loading rates from Lake
Erie watersheds. This method of estimating phosphorus retention does not take into account any
of the interactions between Lake Erie and the wetland, nor does it consider any intrasystem
interactions.
A mass balance predicted from field data (Reeder, 1990; Mitsch and Reeder, 1991a) has a much
lower loading rate, primarily because the data reflect drought year conditions and do not include
an entire calendar year. Phosphorus retention from the field data was estimated to be 36 percent
of the inflow, nearly the same retention percentage as predicted by the empirical models. The
field data results indicate that the contribution of Lake Erie to phosphorus loading to the wetland
is minimal (0.4 percent of watershed inflow). Actual loading rates and retention may be higher
than either the empirical model or field data predict because data from one recent sediment core
show that an average of 22 mg-P m"2d_1 has been deposited over the past 180 years (Reeder,
1990). However, the sediment coring site is in the area of the marsh with the highest
sedimentation rate (Buchanan, 1982), so the sedimentation data probably represent a region of
maximum retention.
10
-------
A simulation model developed for Old Woman Creek takes into account the effects of hydrology
on loading rate, the amount of transformation occurring due to the biota, and the sedimentation
and resuspension estimated from model calibration. The simulation model predicts that 2.9
mg-P nT'd"1, or less than 10 percent, of the phosphorus inflow is permanently retained in the
wetland—more than the field estimates suggest but less than the empirical model and sediment
core predict. As expected, the primary producers transform some bioavailable phosphorus into
nonbioavailable forms, but this contribution to the sediments is minimal (2.2 percent) in the
model. Other simulations predicted that as much as 12.4 mg-P m^d"1 could be retained in the
marsh, especially under high water-level conditions (Reeder, 1990; Mitsch and Reeder, 1991b).
The model estimates are reasonable considering the variability of the data used for comparison,
and probably represent the best available estimate on phosphorus retention in this Great Lakes
coastal wetland.
THE DES PLAINES RIVER WETLANDS, ILLINOIS-EXPERIMENTING WITH WATERSHED
SIZE
The restoration of entire rivers has been shown to be an elusive goal in many parts of the world
because of years of drainage and channel "improvement," floodplain development, and significant
loads of sediments and other nonpoint pollutants. Even in persuing pollution control, we have
paid too much attention to the stream itself and not enough to its interactions with its floodplain.
One ecological engineering project on the Des Plaines River north of Chicago in Lake County,
Illinois, involves restoration of a length of a river floodplain and establishing experimental
wetland basins (Figure 10). This project, begun in 1982 as the "Des Plaines River Wetland
Demonstration Project," has as its goals "to demonstrate how wetlands can benefit society both
environmentally and economically, and to establish design procedures, construction techniques,
and management programs for restored wetlands" in a riverine setting (Hey et al., 1989). Four
11
-------
experimental wetland basins (1.9 to 3.4 ha) have been constructed and instrumented at the
northern half of the site for precise hydrologic control to achieve ecosystem experimentation
conditions for investigating hydrologic design of wetlands subjected to inputs of NPS pollutants
r
(Table 3). Two of the wetlands had inflows of 35 to 38 cm/wk (high flow) while the other two
wetlands had inflows of 10 to 16 cm/wk (low flow). In comparison, precipitation contributed less
than 2 cm/wk during this first year of experimentation. Water level was maintained at
approximately the same depth in each wetland basin.
Results from the experimental wetlands after 1 year of study point to substantial productivity in
all of the wetlands with most of the phosphorus and sediments from the river retained by the
wetlands. Tables 4 and 5 illustrate sediment and phosphorus budgets for the four experimental
wetlands. The low flow wetlands retained from 93 to 98 percent of the inflow of sediments while
sediment retention by the high flow wetlands was 88 to 90 percent (Table 4).
Table 4 also shows estimates of macrophyte and plankton production for the wetlands, and
sedimentation rates as measured by sediment traps. Sedimentation as measured by the traps was
much greater than that estimated by subtracting outflow sediment from inflow sediment. Some
of this difference is undoubtedly due to autochthonous productivity (macrophyte plus plankton
production), although these productivity estimates are small relative to sedimentation rates
measured by the traps. Overall, the results shown in Table 4 suggest that 1) the sedimentation
traps include a significant amount of resuspension from the wetland sediments and 2) the
autochthonous production of sediments (mostly organic) is generally greater than the
accumulation of sediments due to the pumped water.
The phosphorus budget for the wetlands in shown in Table 5. Resuspension of sediments is
obviously an important source of phosphorus to the sediment traps. Of greater interest, the
12
-------
calculated uptake of phosphorus by the macrophytes and water column algae and aquatic plants is
similar to net retention as estimated by input minus output calculations. This illustrates that these
wetlands are probably not being overloaded by phosphorus and may, in fact, be close to a
sustainable loading of phosphorus.
COMPARISON
A comparison of natural and constructed wetlands for the control of NPS loadings is enlightening
(Table 6). The natural wetlands discussed represent both a riparian and an instream wetland.
Heron Pond is flooded by approximately 10 percent of the floodwater from a 1,570 km2
watershed and had approximately 15,000 g m"2y"1 of sediments pass over it during flooding,
while an estimated 500 g m^y"1 were retained. By contrast, the Des Plaines River wetlands were
subjected to inflows of 200 to 900 g m^y"1 of sediments and retained from 200 to 800 g m^y"1.
In the natural wetland, 3 percent of the sediments are retained; in the constructed riparian
wetlands with pumps, about 90 percent of the sediments are retained.
The comparison of the wetland types for phosphorus retention is similar. In Heron Pond, the
riparian wetland retained about 4.5 percent of the phosphorus that passed over it during the flood
event. At Old Woman Creek, an instream wetland fed by a 69 km2 watershed retained
approximately 10 percent of the phosphorus that entered. Retention of phosphorus by the
constructed Des Plaines River wetlands was much higher, with 63-68 percent of the phosphorus
retained in the high-flow wetlands and 83-96 percent retained in the low flow wetlands.
Phosphorus retention as a function of phosphorus loading illustrates that wetlands built in
downstream locations will be generally less efficient in retaining nutrients than would smaller
upstream wetlands. But looking at efficiency alone would be misleading; wetlands built
downstream could retain more mass of nutrients. A tradeoff is obviously necessary.
13
-------
CONCLUSIONS
We must tie our rivers and wetlands back together in an ecologically sustainable way. Successful
ecological engineering requires that we take advantage of our increasing knowledge of ecology
and its principles (e.g., succession, energy flow, self-design, etc.) to construct and restore
wetlands as part of a natural landscape with minimum human maintenance. But we must resist
the ever-present temptation to over-engineer by channeling energies that cannot be channeled
and to over-biologize by introducing species that the design does not support. Boule (1988) has
recommended that the human contribution to the design of wetlands be kept simple without
reliance on complex technological approaches that invite failure. He states that "simple systems
ten'J to be self-regulating and self-maintaining." Self-design and self-organization are in fact the
very cornerstones of the ecological engineering approach and should be used whenever possible in
designing wetlands in the landscape.
14
-------
ACKNOWLEDGEMENTS
The author appreciates the thoughtful and complete reviews provided by Mark M. Brinson,
Robert L. Knight, and Richard Olson and the editing assistance of Ruthmarie H. Mitsch.
15
-------
REFERENCES
Boule, M. E., 1988. Wetland creation and enhancement in the Pacific Northwest, pp. 130-136.
In: J. Zelazny and J. S. Feierabend (eds.), Proceedings of the Conference on Wetlands: Increasing
Our Wetland Resources, Washington, DC. Corporate Conservation Council, National Wildlife
Federation, Washington, DC.
Brooks, R. P., 1989. Wetland and waterbody restoration and creation associated with mining, pp.
529-548. In: J. A. Kusler and M. E. Kentula (eds.), Wetland Creation and Restoration: The
Status of the Science. Island Press, Washington, DC.
Buchanan, D., 1982. Transport and deposition of sediment in Old Woman Creek Estuary of Lake
Erie. M.S. Thesis, The Ohio State University, Columbus, OH.
Dahl, T. E. 1990. Wetlands losses in the United States 1780's to 1980's. U.S. Department of
Interior, Fish and Wildlife Service, Washington, DC.
Davis, D. G., 1989. No net loss of the nation's wetlands: a goal and a challenge. Water
Environment and Technology, 4: 513-514,
Hey, D. L., M. A. Cardamone, J. H. Sather, and W. J. Mitsch, 1989. Restoration of riverine
wetlands: The Des Plaines River wetlands demonstration project, pp. 159-183. In: W. J. Mitsch
and S. E. Jorgensen (eds.), Ecological Engineering: An Introduction to Ecotechnology. John
Wiley & Sons, New York, NY.
16
-------
Klarer, D. M. and D. F. Millie, 1989. Amelioration of storm-water quality by a freshwater
estuary. Archiv fur Hydrobiologia, 116: 375-389.
Livingston, E. H., 1989. Use of wetlands for urban stormwater management, pp. 253-264. In:
D. A. Hammer (ed.), Constructed Wetlands for Wastewater Treatment. Lewis Publishers, Inc.,
Chelsea, MI.
Loucks, O. L., 1989. Restoration of the pulse control function of wetlands and its relationship to
water quality objectives, pp. 467-478. In: J. A. Kusler and M. E. Kentula (eds.), Wetland
Creation and Restoration: The Status of the Science. Island Press, Washington, DC.
Mitsch, W. J. (ed.), 1988. Ecological engineering and ecotechnology with wetlands: applications
of systems approaches, pp. 565-580. In: A. Marani (ed.), Advances in Environmental
Modelling. Elsevier, Amsterdam.
Mitsch, W. J. (ed.), 1989. Wetlands of Ohio's coastal Lake Erie: a hierarchy of systems. Final
Report, Ohio Sea Grant College Program, Columbus, OH.
Mitsch, W. J., 1990. Wetlands for the control of nonpoint source pollution: preliminary
feasibility study for Swan Creek watershed of northwestern Ohio. Final Report to Ohio
Environmental Protection Agency, The Ohio State University, Columbus, OH.
Mitsch, W. J., 1991. Ecological engineering: approaches to sustainability and biodiversity in the
U.S. and China, pp. 428-448. In: R. Costanza (ed.), Ecological Economics: The Science and
Management of Sustainability. Columbia University Press, New York, NY.
17
-------
Mitsch, W. J. and J. K. Cronk. Creation and restoration of wetlands: some design considerations
for ecological engineering. Advances in Soil Science: in press.
Mitsch, W. J., C. L. Dorge, and J. R. Wiemhoff, 1977. Forested Wetlands for Water Resource
Management in Southern Illinois. Research Report No. 132, Illinois Water Resources Center,
University of Illinois, Urbana, IL.
Mitsch, W. J., C. L. Dorge, and J. R. Wiemhoff, 1979. Ecosystem dynamics and a phosphorus
budget of an alluvial cypress swamp in southern Illinois. Ecology, 60: 1116-1124.
Mitsch, W. J. and S. E. Jorgensen, 1989. Introduction to ecological engineering, pp. 3-12. In:
W. J. Mitsch and S. E. Jorgensen (eds.), Ecological Engineering: An Introduction to
Ecotechnology. John Wiley & Sons, New York.
Mitsch, W. J., and B. C. Reeder, 1991a. Nutrient and hydrologic budgets of a Great Lakes
coastal freshwater wetland during a drought year. Wetlands Ecology and Management, 1(3): in
press.
Mitsch, W. J. and B. C. Reeder, 1991b. Modelling nutrient retention of a freshwater coastal
wetland: estimating the roles of primary productivity, sedimentation, resuspension and
hydrology. Ecological Modelling, 54: 151-187.
Mitsch, W. J., B. C. Reeder, and D. M. Klarer, 1989. The role of wetlands for the control of
nutrients with a case study of western Lake Erie. pp. 129-159. In: W. J. Mitsch and S. E.
Jorgensen (eds.), Ecological Engineering: An Introduction to Ecotechnology. John Wiley & Sons,
New York, NY.
18
-------
National Wetlands Policy Forum, 1988. Protecting America's Wetlands: an Action Agenda. The
Conservation Foundation, Washington, DC.
Newton, R. B., 1989. The effects of stormwater surface runoff on freshwater wetlands: a review
of the literature and annotated bibliography. University of Massachusetts, Amherst, MA.
Odum, H. T., 1989. Ecological engineering and self-organization, pp. 79-101. In: W. J. Mitsch
and S. E. Jorgensen (eds.), Ecological Engineering: An Introduction to Ecotechnology. John
Wiley & Sons, New York, NY.
Ogawa, H. and J. W. Male, 1983. The flood mitigation potential of inland wetlands. Water
Resources Research Center Publication No. 138, University of Massachusetts, Amherst, MA.
Reeder, B. C., 1990. Primary productivity, phosphorus cycling, and sedimentation in a Lake Erie
coastal wetland. Ph.D. dissertation, The Ohio State University, Columbus, OH.
Richardson, C. J. and D. S. Nichols, 1985. Ecological analysis of wastewater management criteria
in wetland ecosystems, pp. 351-391. In: P. J. Godfrey, E. R. Kaynor, S. Pelczarski, and J.
Benforado (eds.), Ecological Considerations in Wetlands Treatment of Municipal Wastewaters.
Van Nostrand Reinhold Company, New York, NY.
19
-------
Table 1. Sediment and Phosphorus Budgets for Heron Pond Riparian Cypress Swamp, Adjacent to
Cache River, Southern Illinois (From Mitsch et al., 1979).
River Passes Retained Percent
Load* over Wetland* by Wetland* Retained
~ 15,000 447 3
880 80.2 3.6 4.5
* . -2-1
in g m \
Sediments
Total Phosphorus
20
-------
Table 2. Comparison of Nutrient Retention Capabilities of Old Woman Creek Wetland by Different
Measurements (measurements are in mg-P m"2d_1).
Method
Inflow Outflow Retention Source
Empirical Model v. I 33-63
14-19 Mitsch et al., 1989
Empirical Model v. II 17-33
8-13 Mitsch and Reeder, 1991a
Field Data
2.2 1.4
0.8 Reeder, 1990; Mitsch
and Reeder, 1991a
Sediment Core
22 Reeder, 1990
Simulation Model 30.1 26.5 2.9 Mitsch and Reeder, 1991b
for 9-month period only
21
-------
Table 3. Hydrologic Experiment -- Des Plaines River Wetland Demonstration Project for Period
October 1989 through September 1990.
Wetland 3
Wetland 4
Wetland 5
Wetland 6
Size, acres
5.75
5.79
4.61
8.53
Size, hectares
2.33
2.34
1.87
3.45
Design Inflow, cm/wk
30
5
30
5
Inflow,cm/wk (n = 52)*
37.8 ± 2.2
9.8 ± 0.9
35.2 ± 2.2
16.1 ± 1.8
Outflow, cm/wk (n = 52)*
38.5 ± 1.9
9.0 ± 1.3
33.9 ± 2.0
4.1 ± 0.9
Precipitation, cm/wk
1.6
Evapotranspiration, cm/wk
1.9
* Numbers are average ± standard error
22
-------
Table 4. Budget of Suspended Sediments for Des Plaines River Experimental Wetlands, October 1989 through September 1990 (units are in g-dry
wt m*2 wk"1; inflows, outflows, sedimentation, and productivity are all estimated independently).
Sedimentation
Macrophyte
Est. Plankton
Experim.
Inflow
Outflow
Winter
Growing
Wt. Average
Increase
Production*
Production**
Wetland
Season
in Summer
(Growing
(Growing
(212 d)
(151 d)
(363 d)
Season)
Season)
3
17.4
2.1
117
381
227
264
18
13
4
4.1
0.3
13
353
154
340
24
11
5
16.6
1.7
25
199
97
174
26
18
6
4.2
0.1
15
178
83
163
6
16
* average of 1989 and 1990 estimated above-ground biomass accumulation divided by 22 weeks
** assumes aquatic efficiency for each wetland, average of 4,330 kcal m"2-d"1, and accumulation of 50 percent of gross primary productivity,
averaged over summer week
-------
Table 5. Budget of Total Phosphorus for Des Plaines River Experimental Wetlands, October 1989 through September 1990 (units are in
mg-p m"2 wk~1 ).
Experim. Inflow Outflow Sedimentation Traps Macrophyte Est. Algal
Wetjand Weighted Average* Uptake** Uptake***
3 45.1 16.5 186 22 4
4 10.3 1.7 126 32 3
5 43.7 14.0 91 23 5
6 13.7 0.5 85 5 5
* assumes approximately 1 mg-P g dry wt"1 in sediments in traps x annual sedimentation rate
** macrophyte biomass accumulation x weighted concentration, averaged over 52 weeks
*** assumes aquatic efficiency for each wetland, average 4,330 kcal m"2 d*1, 5-month growing season, accumulation of 50 percent of
gross primary productivity, and 0.7 mg-P g dry wt"1 (Cladophora), averaged over 52 weeks
-------
Table 6. Comparison of Nonpoint Source Sediment and Phosphorus Retention in Natural And Constructed Wetlands
sediments, g m"2 y"1 phosphorus, g-P m~2 y"1
watershed
size, km2 loading retention (%) loading retention (%)
Natural Wetlands
Heron Pond, Southern IL (riparian)
Old Woman Creek, OH (instream)
Constructed Wetlands
Des Plaines River Wetlands (riparian with pump; year 1)
Wetland 3 - high flow
1.4*
905
796
(88)
2.34
1.49
(63)
Wetland 4 - low flow
0.4*
213
198
(93)
0.54
0.45
(83)
Wetland 5 - high flow
1.0*
863
775
(90)
2.27
1.54
(68)
Wetland 6 - low flow
0.5*
218
213
(98)
0.71
0.69
(96)
•approximate equivalent watershed size for pumping rate
1,570
69
15,000
447 (3)
80.2
II
3.6 (4.5)
1.1 (10)
-------
LEGENDS FOR FIGURES
Figure I. Example of a constructed instream wetland with part of the flow bypassing the
wetland during high flow conditions.
Figure 2. Cross-section of a natural or constructed riparian wetland in dry and flooding seasons.
Figure 3. Example of riparian wetland fed by pump to maintain experimental conditions.
Figure 4. Example of riparian wetland fed by gravity.
Figure 5. Watershed design showing alternatives of many smaller upstream wetlands versus one
larger downstream wetland.
Figure 6. Details of multiple upstream wetlands showing wetlands intercepting small streams
(solid lines) and drainage tiles (dotted lines).
Figure 7. Example of a terraced wetland for steep terrain.
Figure 8. Position of Heron Pond riparian wetland adjacent to the Cache River in southern
Illinois (from Mitsch et al., 1977, 1979). In cross-section, the wetland is situated as shown in
Figure 2.
Figure 9. Location of Old Woman Creek wetland in northern Ohio as one large instream wetland
at downstream location of 69 km2 watershed (from Reeder, 1990).
26
-------
Figure 10. Des Plaines River Wetland Demonstration Project in northeastern Illinois, showing
experimental wetlands 3 through 6 at northern edge of site. Water is pumped to wetlands from
the Des Plaines River and it flows by gravity back to the river as shown in the schematic of
Figure 3.
27
-------
Figure 1
-------
Natural
Levee
Riparian Wetland
DRY SEASON
Floodplaln
FLOODING EVENTS
Figure 2
-------
Pump
Figure
3
-------
Figure 4
-------
Multiple
upstream
Ł> wetlands
Single
downstream
wetland
Figure 5
-------
Figure 6
-------
Figure 7
-------
OWater Sample Station
OassRvhTiON Well
® GftdlNG Stwtioki
O LiTTERTRftp/ftfliM Gage
V Sediment Trap
V
V ~ ~
HERON
POND
.5
ki tom»t® p
Figure 8
-------
Ohio
Old Woman
Creek
Wetland
Ohio Rout* 2
Darrow Rd
Approximate
¦*T. ¦
Kilometers
Figure 9
-------
Research
trailers
Meteorological
Station
73
g
S
•¦5
.Mill Creek
%
Inflow
pipeline
Outflow
'swale
. Des Plaines
River
Pump
station
N
500 1000
feet
200
400
meters
Experimental Wetlands
River, Quarries, and
Spawning Areas
Passive Wetland Area
Site Boundary
Site
Location
Figure 10
-------
DESIGNING CONSTRUCTED WETLANDS SYSTEMS TO TREAT
AGRICULTURAL NONPOINT SOURCE POLLUTION
Donald A. Hammer
Regional Waste Management Department
Tennessee Valley Authority
2F 67B Old City Hall Building
Knoxville, Tennessee 37902-1499
ABSTRACT
Increasingly concentrated animal husbandry practices and more intensive row crop farming have
expanded agricultural pollution problems. Implementing accepted best management practices
(BMPs) for erosion control and waste handling along with a combination of 1) onsite constructed
wetlands, 2) nutrient-sediment control systems in small watersheds, and 3) natural wetlands along
streams and at strategic locations in large watersheds may provide low-cost, efficient control.
Design recommendations and examples are included.
The views expressed herein are those of the author and do not necessarily represent those of the
Tennessee Valley Authority.
-------
INTRODUCTION
Public concern about water pollution during the last 30 years has resulted in State and Federal
legislation regulating discharges and provided financial assistance for municipal treatment
facilities. Substantial progress in treating point sources has and is continuing to occur, especially
in larger cities and with major industries. Widespread implementation of wetlands treatment
technology may accomplish similar objectives with small community and small industry sources.
However, anticipated improvements in the Nation's waters have not been realized, and recent
evaluations reflect a growing concern over nonpoint source (NPS) pollution, especially
agricultural wastewater and agricultural cropland runoff, and urban stormwater runoff. These
principal contributors to NPS pollution problems have been difficult to remedy with conventional
wastewater treatment and soil/water conservation methods.
NPS pollution from agricultural operations, urban areas, failed home septic tank drain fields,
mining, and other land disturbing activities continues to detrimentally impact 30 - 50 percent of
our nation's waterways. Increasing regulatory focus on agricultural NPS pollution probably
results not only from reduction or elimination of point sources but also from real changes in
waste loading in receiving streams because of changing animal husbandry practices. Previously,
free-ranging livestock (including poultry) at relatively low population densities caused little
aquatic pollution because wastes were widely dispersed and natural soil systems recycled nutrients
onsite. However, the historical and continuing tendency to confine livestock in ever smaller areas
to improve production efficiency has the effect of concentrating animal waste loading with
subsequent runoff to nearby streams. Concurrent removal of woody and non-woody riparian
vegetation to increase efficiency by using all available acreage, or incidental removal due to
livestock grazing and loafing has eliminated the buffer strip that formerly protected streams from
direct pollutant impacts (Hammer, 1989a).
2
-------
Farmers are unlikely to purchase and operate package treatment plants. Requiring a hog
producer to purchase a multimillion dollar treatment system to deal with the waste from 1,000
hogs, which has similar organic loading but is much more concentrated than waste from a city of
1,000 residents, is unrealistic since many farmers are heavily in debt with marginally profitable
operations. On the other hand, constructed wetlands waste treatment systems would seem more
amenable to the substantial range of hydraulic and pollutant loading, temporal fluctuations,
dispersed nature, and the need for low-cost, low-technology systems acceptable to farmers.
Furthermore, planting and maintenance requirements on the farm differ little from skills needed
in growing other crops, and land costs are relatively low.
NATURAL WETLANDS
Although natural wetlands systems provide many functional benefits to our society, virtually all
can be grouped into three broad categories—life support, hydrologic buffering, and water quality
improvement. Most of us are familiar with the many types and large numbers of animals,
especially birds, that are dependent upon wetlands. But how many of us realize that the crayfish
industry in Louisiana, the shellfish and much of the finfish industry along our coasts, and the
furrier industry that clothes our elegant women are also all dependent upon wetlands (Mitsch and
Gosselink, 1986). Wetlands also reduce flooding along rivers and streams by reducing and
desynchronizing peak runoff through slowing flood water velocities. At the other extreme,
delayed flows emanating from wetlands augment base flows in streams and rivers maintaining
levels essential for aquatic life. Finally, contaminated waters flowing through natural wetlands
are cleansed by a combination of physical, chemical, and biological activities, and emerge as
clean water (Hammer, 1990).
3
-------
Wetlands ecosystems have intrinsic abilities to modify or trap a wide spectrum of water-borne
substances commonly considered pollutants or contaminants. Doubtless, our ancestors perceived
and exploited these abilities. Only in more recent times, however, have casual observations
fostered renewed interest in wetlands. Such casual observations have led to investigations that
documented changes in concentrations of various materials after processing by natural wetlands
systems. In fact, much of the early work on constructed wetlands for wastewater treatment was
stimulated by observing this purification phenomenon in natural wetlands systems (Seidel, 1971;
Kadlec, et al., 1974; Odum and Brown, 1976; Small, 1977). For example, many observers have
noticed accelerated soil erosion after heavy rains wash across unvegetated soils, and some have
been fortunate enough to encounter situations were silt-laden waters transiting natural wetlands
systems were readily compared with unprocessed waters. The striking visual differences were
easily verified by sampling and analysis, and the information became an important component in
a communal body of knowledge on natural wetlands values. Most ecologists believed this
phenomenon was widespread, and a few even suggested that it might occur on a large scale,
though little documentation was available.
I recently had the opportunity to observe an example of water quality improvements in river
waters by a natural wetlands system on a very large scale in the Pantanal of western Brazil and
adjacent portions of Paraguay and Bolivia. This area is a large basin bordered by high plateaus
and a savannah on the east, a semi-deciduous forest on the north, and a moderate mountain range
and a semi-deciduous forest on the west. Runoff from these regions causes much of the
11,000,000 hectare area to be flooded from December to June, and a significant but unmeasured
proportion is permanently wet. Many rivers enter the Pantanal from the eastern highlands,
gradually disappearing and then reforming on the western and southern boundaries and draining
off to the south. Over geological time, alluvial deposits of highlands silt has gradually
4
-------
transformed a flat or concave basin floor into a convex, dome-like surface with higher elevations
in the center and lower on the margins.
Doubtless, this region provided important water improvement functions since tectonic forces
created the original basin. But accelerated erosion and pollution from clearing and agricultural
activities and other anthropogenic sources has tremendously increased the contaminant loading of
rivers draining the plateaus on the east and north. The Rio Taquiri alone carries more than
30,000 metric tons of silt per day plus a variety of agrochemicals from soybean fields on the
eastern plateau (Amaral, 1989). Other rivers transport lower silt loads, but most receive untreated
sewage and industrial and mining pollution before reaching the Pantanal.
Amazingly, alarmingly high concentrations of silt and pollutants in inflowing river waters are
reduced to innocuous levels in waters of rivers draining the region (Cadavid, 1989). Examination
of a topographic map (abstracted in Figure 1) provides insight into the general process, though it
does not reveal the complex of purification mechanisms. Notice the size, especially width, of the
Rio Taquiri as it drops off the plateau and enters the Pantanal on the east. A fairly wide, deep,
and fast flowing river courses out into the Pantanal, and rather quickly its width, depth, and
velocity are reduced. A third of the way into the Pantanal, numerous small braided streams arise
flowing perpendicularly out of the Rio Taquiri into the adjacent regions. Progressively
increasing water loss with penetration into the Pantanal drastically reduces the Rio Taquiri until
it almost disappears. In fact, the Pantanal functions as an 11,000,000 hectare sponge that absorbs
inflowing waters, cleanses them of impurities, and slowly releases clean water through minor
streams that aggregate into larger rivers along the southern and western boundaries. This large
natural wetlands complex transforms heavily polluted influent waters into clean waters, and the
slow release of waters collected during the rainy season augments base flow in the Rio Paraguay
5
-------
during the dry half of the year. On a large scale as well as in local areas, natural wetlands can
perform substantial improvements in water quality and quantity.
Constructed wetlands have recently received considerable attention as low-cost, efficient means
to clean up many types of wastewater. Though the concept of deliberately using wetlands for
water purification has only developed within the last 20 years, in reality, human societies have
indirectly used natural wetlands for waste management for thousands of years. We have always
dumped our wastes into nearby streams or wetland areas. And as they do for natural terrestrial
ecosystems, wetlands processed these wastes and discharged relatively clean water. However, as
human populations increased and concentrated in towns and later cities, the increased quantity of
wastes discharged into a small area soon overloaded natural wetlands and other aquatic systems
damaging the wetlands and destroying their function in removing water borne pollutants.
Without wetlands treatment buffering downstream areas, human wastes damaged aquatic life in
rivers, bays, and oceans and threatened drinking water supplies.
WETLANDS PURIFICATION FUNCTIONS
VEGETATION
Water purification functions of wetlands are dependent upon four principle components—
vegetation, water column, substrates, and microbial populations. The principle function of
vegetation in wetlands systems is to create additional environments for microbial populations
(Pullin and Hammer, 1989). Not only do stems and leaves in the water column obstruct flow and
facilitate sedimentation, they also provide substantial quantities of surface area for attachment of
microbes and constitute thin-film reactive surfaces. In addition to the microbial environments in
the water column of lagoons, wetlands have much additional surface area on portions of plants
within the water column. Plants also increase the amount of aerobic microbial environment in
6
-------
the substrate incidental to the unique adaptation that allows wetlands plants to thrive in saturated
soils. Most plants are unable to survive in water-logged soils because their roots cannot obtain
oxygen in the anaerobic conditions rapidly created after inundation. However, hydrophytic, or
wet-growing plants, have specialized structures somewhat analogous to a mass of breathing tubes
in their leaves, stems, and roots that conduct atmospheric gases, including oxygen, down into the
roots. Because the outer covering on the root hairs is not a perfect seal, oxygen leaks out creating
a thin film aerobic region—the rhizosphere—around each and every root hair. The larger region
outside the rhizosphere remains anaerobic but the juxtaposition of a large, in-aggregate, thin film
aerobic region surrounded by an anaerobic region is crucial to transformations of nitrogenous
compounds and other substances. Wetlands vegetation substantially increases the amount of
aerobic environment available for microbial populations, both above and below the surface.
Wetlands plants generally take up only very small quantities (<5 percent) of the nutrients or other
substances removed from the influent waters. However, some systems incorporating periodic
plant harvesting have slightly increased direct plant removals at considerable operating expense.
MICROBIAL ORGANISMS
Microbes—bacteria, fungi, algae, and protozoa—alter contaminant substances to obtain nutrients
or energy to carry out their life cycles. In addition, many naturally occurring microbial groups
are predatory and will forage on pathogenic organisms. The effectiveness of wetlands in water
purification is dependent on developing and maintaining optimal environments for desirable
microbial populations. Fortunately, these microbes are ubiquitous, naturally occurring in most
waters, and likely to have large populations in wetlands and contaminated waters with nutrient or
energy sources. Only rarely, with very unusual pollutants, will inoculation of a specific type or
strain of microbes be needed.
7
-------
SUBSTRATES
Substrates—various soils, sand, or gravel—provide physical support for plants; reactive surface
area for complexing ions, anions, and some compounds; and attachment surfaces for microbial
populations. The water column—surface and subsurface water—transports substances and gases to
microbial populations, carries off byproducts, and provides the environment and water for
biochemical processes of plants and microbes.
ANIMALS
Invertebrate and vertebrate animals harvest nutrients and energy by feeding on microbes and
macrophytic vegetation, recycling and in some cases transporting substances outside the wetlands
system. Functionally, these components have limited roles in pollutant transformations, but they
often provide substantial ancillary benefits (recreation/education) in successful systems. In
addition, vertebrate and invertebrate animals serve as highly visible indicators of the health and
well-being of a marsh ecosystem, providing the first signs of system malfunction to a trained
observer. Some invertebrates and many vertebrates occupy upper trophic levels within the system
that are dependent upon robust, healthy populations of micro and macroscopic organisms in the
critical lower levels. Declines in lower-level populations (including those involved in pollutant
transformations) are reflected in changes in more visible animals in the higher levels. However,
observations on types and numbers of indicator species must be carefully interpreted and/or
compared to conditions in natural, unimpacted wetlands by an experienced wetlands ecologist,
since certain species thrive in overloaded, poorly operating systems.
CONSTRUCTED WETLANDS
Constructed wetlands, in contrast to natural wetlands, are man-made systems that are designed,
built, and operated to emulate natural wetlands or functions of natural wetlands for human
8
-------
desires and needs. As used for wastewater treatment, constructed wetlands may include
swamps—wet regions dominated by trees, shrubs, and other woody vegetation—or bogs, which are
low-nutrient, acidic waters dominated by Sphagnum or other mosses. However, we most
commonly refer to them as marshes. Marshes are shallow-water regions dominated by emergent
herbaceous vegetation—cattails, bulrushes, rushes, and reeds—and are adapted to a tremendous
variety of soil and climatic conditions. Some marsh plants occur on every continent except
Antarctica. Marshes are also adapted to a wide range of water quality conditions as well as
substantial fluctuations in water flows and depths. Although bogs and swamps have been used
for wastewater treatment, both are difficult to establish or manage and require fairly stable water
quality and quantity conditions. Alterations in either are likely to cause undesirable changes in
the structure and function of bogs and swamps.
The vast majority of wetlands constructed for wastewater treatment are classified as surface-
flow or free-water surface systems (Reed, 1990), that is, influent waters that flow across and
largely above the surface of the substrate materials. Substrates are generally native clay or soil.
In the other major class, subsurface-flow systems, waters flowing through the system pass
entirely within the substrate, and free water is not visible. Substrates in subsurface-flow systems
are typically various sizes of gravel or crushed rock. Though subsurface-flow systems appeared
to have considerable potential only a few years ago, in practice virtually all subsurface-flow
systems with 2 or more years of operating history have experienced serious clogging problems. In
addition subsurface flow systems are unable to maintain adequate dissolved oxygen levels for
ammonia removal. Thus, only a few subsurface-flow systems that treat municipal waste are
operating in North America, but many of the European municipal systems are of this type. Only
surface-flow systems have been used for mine drainage, agricultural waste, urban stormwater,
industrial wastewaters, or other applications to date. Because a number of the operational
subsurface flow systems have experienced clogging problems, only surface flow systems can be
9
-------
recommended for anything less than tertiary polishing of effluents with low concentrations of
nutrients.
Constructed wetlands currently treat wastewaters from towns and small cities, mine drainage,
urban stormwater runoff, livestock production, failed septic tank fields, land fill leachate, paper
mills, tanneries, food processing plants, petroleum refineries, and many other small industrial
sources (Reddy and Smith, 1987). Operating systems are located from sea level to 5,000' and
from the tropics to subartic regions in Ontario and the Scandinavian countries (Miller, 1989; Brix
and Schierup, 1989). Since operation is dependent on chemical and biological processes, pollutant
removal efficiencies decline somewhat during low temperatures but discharge levels remain well
below permit limits.
NONPOINT SOURCE POLLUTION
The most efficient approach to controlling agriculturally related NPS pollution—and the most
acceptable to landowners—employs a combination of accepted BMP for waste handling and
erosion control along with constructed and natural or restored wetlands systems in a hierarchial
system. Normal BMP's include dry-stacking and disposal of solid wastes, roof guttering, lagoons,
land application, terraces, grassed waterways, nutrient management, conservation tillage, and crop
rotations. Following installation and/or use of BMPs, first-order control uses constructed
wetlands designed and operated specifically for treating wastewater emanating from concentrated
livestock areas, processing facilities, and in many cases, septic tanks serving the farm household.
Second-order control consists of nutrient/sediment treatment systems strategically located
downstream from the wetlands treatment systems, at the lower end of grassed waterways and
within intermittent stream courses throughout the individual farm (Figure 2). Third-order
control deploys nutrient/sediment treatment systems, constructed wetlands/pond complexes, and
10
-------
restored or created wetlands at specific sites within a watershed that may include many individual
farms. Fourth order systems consist of larger wetlands in the lower reaches of an individual
watershed that function primarily for hydrologic buffering, and life support values in addition to
limited water purification (Figure 3).
From a different perspective, first-order control may be considered simply wastewater treatment
at the source, second order is treatment with constructed systems of less concentrated, aggregate
wastewater from a variety of sources, third order is represented by buffer strips of riparian
wetlands along permanent streams, small restored or created wetlands at specific points in the
upper reaches of the watershed, and fourth order consists of larger areas of restored or created
wetlands at tributary stream intersections in the lower sections of the watershed. Furthermore,
first-order systems are principally designed and operated for wastewater treatment; second-
order systems provide treatment but also produce some ancillary benefits; while third- and
fourth-order systems function much the same as regional, natural wetlands accomplishing water
purification, hydrologic (flood) buffering, life support, and related beneficial values.
Unfortunately, a few of the wetlands descriptors have been used synonymously and need precise
definition to insure common understanding.
Natural wetlands are those areas wherein, at least periodically, the
land supports predominantly hydrophytes and the substrate is
predominantly undrained hydric soil or the substrate is non-soil
and is saturated with water or covered by shallow water at some
time during the growing season of each year. Natural wetlands
have and continue to support wetlands flora and fauna.
11
-------
Restored wetlands are areas that previously supported a natural
wetlands ecosystem but were modified or changed, eliminating
typical flora and fauna, and used for other purposes. These areas
have then subsequently been altered to return to poorly drained
soils and wetlands flora and fauna to enhance life support, flood
control, recreational, educational, or other functional values.
Created wetlands formerly had well-drained soils supporting
terrestrial flora and fauna but have been deliberately modified to
establish the requisite hydrological conditions producing poorly
drained soils and wetlands flora and fauna to enhance life support,
flood control, recreational, educational, or other functional values.
Constructed wetlands consist of former terrestrial 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 wastewater. Constructed wetlands are
essentially wastewater treatment systems and are designed and
operated as such, though many systems do support other functional
values.
Constructed and created wetlands occupy formerly dry sites; whereas, natural and restored
systems occupy locales that historically were poorly drained and supported wetlands flora and
fauna. In addition, wetlands built or managed for life support and other functional values would
be subject to protective provisions with implementation of Section 404 regulations; whereas,
wastewater treatment systems (constructed wetlands) requiring management practices that may be
12
-------
inimical to other functional values would be subject to NPDES provisions rather than 404
regulations.
In the holistic watershed approach discussed above, wetlands components of onsite wastewater
treatment systems and sediment/nutrient control systems are constructed wetlands; whereas,
riparian buffers, strategically placed marsh/pond complexes on permanent streams, and
downstream marshes or bottomland hardwoods are created or restored wetlands. The former
(first- and second-order) will require deliberate management/manipulation to maintain optimal
treatment performance, but the latter (third- and fourth-order) would provide substantial water
purification without active management and would support additional wetlands functions. First-
and second-order systems are located within the boundaries of an individual farm, and third-
order systems polish the runoff from a number of farms. In contrast, fourth-order systems
control NPS pollution from an entire watershed.
Since most erosion control practices, solid waste handling, land application, lagoon treatment,
riparian buffers, and large, fourth-order wetlands systems are fairly well understood and widely
employed, the following discussion will be limited to methodologies of constructed wetlands for
onsite treatment and the new nutrient/sediment control systems developed by Robert
Wengryznek, Soil Conservation Service, Orono, Maine.
LIVESTOCK WASTEWATERS
In cooperation with the Soil Conservation Service (SCS) and Auburn University, the Tennessee
Valley Authority (TVA), in 1988, initiated a constructed wetlands project to evaluate treatment
performance and to develop design/operating criteria at Auburn's Sand Mountain Agricultural
Experiment Station in northeast Alabama (Hammer, et al 1989). The system receives effluent
from a secondary lagoon treating waste from approximately S00 hogs (Figure 4). The design
13
-------
includes a small farm pond for flood protection and dilution water, a mixing pond, and two
replicates of five individual cells containing different vegetation. Different loading rates are
applied to different cells to test plant survival and removal efficiencies under fairly extreme
conditions (>100 mg/L NHa). A similar experimental system treating wastewater from a dairy
operation has recently been initiated by the SCS near Newton, Mississippi.
The farm pond was a less expensive alternative to excavating a flood water channel alongside the
wetlands cell complex that serendipidously provided us an opportunity to demonstrate simple
treatment for truly NPS runoff. Contoured terraces in the winter pasture above and alongside the
farm pond direct virtually all of the runoff from the pasture to the upper end of the pond. High
nutrient concentrations in pasture runoff supported an excessive algae bloom the first spring and
summer after construction, and pond overflow discharging into a ditch flowing alongside the
wetlands cell complex was of poor quality. However, merely installing a fence across the upper
end and west side of the pond and planting wetlands vegetation on the pond margin and in the
discharge channel has eliminated algae blooms and poor quality water in the ditch in subsequent
years. The shallow, upper reaches of the pond are densely covered with Fimbristylis, Carex,
Juncus, and Scirpus, which remove nutrients from the pasture runoff, eliminating impacts to the
farm pond and downstream waters.
We also cooperated with Mississippi State University, the Mississippi Bureau of Pollution Control,
and the SCS in constructing an operational constructed wetlands treatment system for a 500-hog
operation at Mississippi State's Pontotoc Experiment Station (Figure 5). Since nitrogen (NHj) is
typically the most significant component of lagoon discharges, this design reflects current
thinking on combining marsh-pond-marsh units within a single cell to improve nitrogen removal
(Figure 6). I also added typical overland flow strips to compare removal efficiencies with the
wetlands cells.
14
-------
ROWCROP RUNOFF
In northern Maine, runoff from potato fields jeopardizes cold, deep lakes with a lake trout and
land-locked salmon sport fishery that is economically significant to Aroostook County. In
cooperation with other organizations, the Orono office of the SCS designed and constructed two
demonstration nutrient/sediment control systems in watersheds with other BMPs for erosion
control already in place. The nutrient/sediment system consists of a sedimentation ditch—bermed
on the lower side—leading to an overland flow meadow, followed by a cattail marsh, a pond, and
a final polishing meadow. Results have been very good with more than 80 percent removal of
sediment, nitrogen, and phosphate from rowcrop runoff, and the treatment systems provide black
duck breeding habitat as well as bait-fish rearing sites (Anon., 1991; Higgins, 1991).
DESIGNING A CONSTRUCTED WETLANDS FOR LIVESTOCK
WASTEWATER TREATMENT
Constructed wetlands systems for control of agricultural wastewater can be designed on the basis
of information from the many municipal wastewater treatment systems in operation all over the
world and a few agricultural systems in North America and Australia.
SYSTEM COMPONENTS
Emergent Marsh: A shallow basin with densely growing marsh vegetation—typically cattail
(Typha), bulrush (Scirpus validus or cyperinus), reed (Phragmites), or rushes {Juncus,
Eleocharis)—in 10-20 cm of water. The marsh functions to remove organic load (BODg),
suspended solids (TSS), and pathogens as well as in ammonification.
IS
-------
Pond: A constructed pond with 0.5-1 m water depths similar to an aerobic lagoon. Duckweed
(Lemna) grows on the surface of the pond and various algae within the water column.
Submerged pondweeds with linear, filiform leaves (Polamogeton, Ceratophyllum, Elodea,
Vallisneria) may be planted in shallow portions of the pond to increase microbial attachment
surface area. The pond functions in further reduction of BODGand most significantly for
nitrification and denitrification.
Meadow: The meadow is constructed and may be operated similar to an overland flow system. It
is planted with reed canary grass (Phalaris arundinacea) or other water tolerant grasses and
sedges, and it receives the effluent from the pond as shallow sheet flow distributed across the
width of the meadow cell. When operated under continuous flows, water depths are maintained
at 1-5 cm, but if operated as an overland flow system, wastewater is batch applied to one cell
flowing across and down the cell to drain off the lower end while the alternate cell is allowed to
dry and oxidize. In either case, the meadow functions for removal of TSS (primarily algae)
generated in the pond and for nitrification and denitrification.
Comparatively, the marsh functions most efficiently for BOD& TSS, and pathogen removal, but
the pond and overland flow meadow, because of the greater amount of oxidized environment, are
more efficient at transforming ammonia to nitrogen gases. However, a lightly loaded marsh and
meadow will provide similar removal efficiencies, though the total required treatment area may
be similar to the combined area in a marsh/pond/meadow system. Primary treatment in a lagoon
and a marsh loaded at 100 Kg BOD^ha/day will provide treatment to secondary discharge
standards - <30 mg/L BOD6and TSS and <200 colonies/100 ml fecal coliforms - but limited
nutrient control.
16
-------
DESIGN CRITERIA
Primary treatment should be provided by a single- or multistage lagoon system designed to
achieve a 50 percent reduction of the BOD5 and TSS loading in the wastewater stream. If a
lagoon is not present or impractical, a settling basin designed to remove solids, grit, and debris
must be located upstream of the wetlands.
Wetlands site selection and delineation will be controlled by the requirement to provide gravity
flow for wastewater to the system, between system components, and within each component of
the system to eliminate costs and maintenance of pumping wastewaters. Similarly, water control
structures or devices should consist of simple "T" pipe along the length of inlet distribution piping
and swiveling "elbow" piping or flashboard/stoplog constructs for discharge control structures.
Neither clogs as easily or requires adjustment or other complex maintenance typical of ball- or
gate-valve flow control devices (Hammer, 1991).
WASTE LOADING COMPUTATIONS
Determination of effective treatment area required to achieve desired discharge standards is based
upon: 1) the quantity (mass) of organic wastes to be treated per day; and 2) the capacity for a
given area of wetlands to transform a fixed quantity (mass) per day. Consequently, calculations
on the required treatment area begin with determining the total organic load generated at the
feedlot, barn, or other facility or in some cases the entire production unit. Representative values
of BODfr nitrogen, and phosphate production per day from cattle, swine, and poultry are shown
in Table 1.
Estimation of waste generation by specific parameter for dairy cattle, swine, and poultry is
calculated by multiplying the number of stock by the value (in grams) for the parameter of
17
-------
interest and converting to kilograms (Kg). Since these are average values that do not include
waste hay or feed and other sources of organic waste, a prudent designer will include a factor of
10-20 percent to estimate the total organic load generated per day.
A less exact method uses typical concentrations of 2000-4000 mg/L BODg, 300-500 mg/L NHS
and 75-150 mg/L total P in raw livestock wastewater and measures the volume of daily flows to
estimate the total daily waste loading for each parameter. However, concentrations vary
substantially with husbandry practices, type and age of stock, and with seasons of the year.
The total organic load generated per day can then be used to examine the treatment capacity
available in an existing anaerobic or aerobic lagoon or to design a new lagoon to provide primary
treatment. In either case the lagoon should accomplish a minimum of 50 percent and preferably
60 percent reduction of BOD5 and suspended solids to reduce the amount of treatment area
required in the wetlands system.
Incorporating a lagoon or settling basin for primary treatment provides storage capacity for
seasonal application to the wetlands, reduces the treatment area needed in the wetlands, and
accomplishes pollutant reduction more efficiently than a stand-alone wetlands system. To
illustrate this, a plot depicting pollutant removal efficiency or concentration levels on the Y axis
and retention time (an indirect volumetric value) or capacity as volume or effective treatment
area on the X axis produces an exponentially decreasing curve with the line for lagoons nearest
the Y axis, the wetlands line intermediate, and the overland flow line farthest from the Y axis.
Almost irrespective of the treatment method used, the greatest reductions in organic loading and
solids occurs at the highest initial concentrations with substantially lower percentage reductions
occurring as pollutant concentrations decline. For example, reduction of BOD6 from 3000 mg/L
to 300 mg/L often requires relatively little treatment area or retention time as compared to the
-------
treatment needed to reduce 300 mg/L to 30 mg/L. The reduction from 30 mg/L to 10 mg/L
requires even more treatment area or retention time. Not only are lagoons more efficient (unit
area basis) at high pollutant concentrations, but levels above 300-500 mg/L would stress a
wetlands or overland flow systems, perhaps even to the point of failure. Conversely, wetlands
and overland flow systems are much more efficient (unit area basis) at reducing 300 mg/L to 30
mg/L and even more so in reducing 30 mg/L to 10 mg/L. Since most natural treatment methods,
including wetlands, produce small amounts of BODsand some solids, levels below 5-10 mg/L are
unlikely to be achieved.
Lagoons are also effective at reducing high levels of phosphate to moderate levels. However,
lagoons are ineffective at removing ammonia through nitrification and denitrification because
ammonification during organic matter decomposition creates ammonia. Consequently, anaerobic
lagoon discharges may have ammonia nitrogen levels of 400-500 mg/L, though an in-series
aerobic lagoon following an anaerobic lagoon will have lower nitrogen levels. Since the age of an
existing lagoon, past and current management practices, and loading influence removal
performances, it is imperative that the designer have available information on actual
concentrations of contaminant parameters in the lagoon discharge before estimating expected
loading on the wetlands system.
REQUIRED TREATMENT AREA
The treatment area needed to reduce organic and nutrient levels in lagoon discharge to secondary
discharge standards has been determined from wetlands use to treat municipal wastewaters and
from transformation/assimilation studies of natural wetlands systems. To achieve a discharge
level of 30 mg/L BODb the wastewater loading must not exceed 100 Kg/ha/day (90
pounds/acre/day) of effective treatment area. Similarly for NHj discharge levels below 10 mg/L,
the total nitrogen (TKN) loading rate must not exceed 10 Kg/ha/day (9 pounds/acre/day). Not
19
-------
surprisingly, phosphorus recommendations are one order of magnitude lower (e.g., 1-1.5
Kg/ha/day) since wastewater concentrations of these substances differ by a reverse order of
magnitude. Generally, livestock as do municipal wastewaters have 10 times as much nitrogen as
phosphorus and 10 times as much BODs as nitrogen so that the required treatment area for each
of these parameters tends to coincide for any specific wastewater stream. Note also that organic
loading of 10 times the nitrogen level will insure that minimally adequate carbon is available for
desired nitrogen transformations.
The effective treatment area required is calculated by dividing the organic load by 100 to derive
the answer in hectares. For example, a wastewater flow with 300 Kg/day would require 3
hectares of effective treatment area in the marshes to produce discharge flow concentrations of
<30 mg/L and, similarly, with the other parameters. However, this represents effective treatment
area, not necessarily everything within the dikes. The effective treatment area is determined by
measuring across the cell from the base of the internal side of the dike, not from the top of the
dike, since in large systems the dike network occupies a significant amount of surface area.
During periods of low water level operation (5-8 cm), adequate treatment area is still available if
this design specification method is used.
A minimum of two marsh cells should be included to allow for removing one from service for
maintenance, if necessary. A single pond will be adequate, and a single meadow may be used if
the meadow is to be operated under continuous flow. If the meadow is to be operated as an
overland flow system, then the system must have two meadow cells to permit alternating
application and drying.
20
-------
Proportionately, after computation of required area for each component—marsh, pond and
meadow—the effective treatment area tends to be approximately 50 percent in the marshes, 30
percent in the pond, and 20 percent in the meadow(s).
Computations of Required Treatment Areas-Tertiarv Treatment
For example, the following computations are for a dairy herd of ISO cows.
Waste load generated per day:
BOD5 produced per cow per day ¦ 773 grams X 150 cows - 115950 grams ¦ 115.95 kilograms per
day, or 116 Kg/day.
Factoring in waste hay or other feed, etc. ¦ 116 X 1.1 • 127.6 or 128 Kg/day.
Nitrogen produced per cow per day ¦ 186 g X 150 - 27900 g ¦ 27.9 Kg or 28 Kg/day.
Anaerobic lagoon: The lagoon may be loaded at 200 Kg BOD^/ha/day in regions with limited
periods of air temperatures below 0°C or 100 Kg BOD^/ha/day in colder regions or seasons.
Removal efficiency « 50% of BODj; 20% of nitrogen.
Effluents — BODb« 128 X 0.5 - 64 Kg/day; Nitrogen - 28 X 0.8 - 22.4 Kg/day.
Marshes: Waste load ¦ 64 Kg/day.
Application rate - 100 Kg BODg/ha/day.
Treatment area » 64 Kg/day / 100 Kg BOD^ha/day - 0.64 hectares in two marsh cells.
Marsh N removal - 30% reduction; effluent from marshes - 22.4 X 0.7 » 15.78 or 16 Kg/day in
effluent.
21
-------
Pond: Application rate « 40 Kg N/ha/day.
Required area = 16 Kg N/day / 40 Kg/ha/day - 0.4 ha.
Pond N removal « 60% reduction « 6.4 Kg/day in effluent.
Meadow -- continuous flow: Application rate « 20 Kg N/ha/day.
Required area - 6.4 Kg N/day / 20 Kg/ha/day « 0.32 ha.
Meadow N removal ¦ 90% reduction « 0.6 Kg/day.
Meadows operated as overland flow systems may be loaded at 30-40 Kg/ha/day.
Total Wetlands System Area
Marshes - 0.64 ha; pond - 0.4 ha; meadow ¦ 0.32 ha; Total « 1.36 hectares or 3.4 acres.
Required Treatment Area-Secondary Treatment
Reduction of organic loading, suspended solids, and fecal coliforms to secondary levels may be
obtained without the pond and meadow, but discharge levels of nitrogen will be high unless a
lower loading rate, e.g., 50 Kg BODj/ha/day, is used for the marshes.
Marshes: Waste load « 64 Kg/day.
Application rate - 75 Kg BOD$/ha/day.
Treatment area « 64 Kg/day / 75 Kg BOD^/ha/day - 0.85 hectares or 2.1 acres in two marsh
cells.
Marsh N removal » 60% reduction; effluent from marshes - 22.4 X 0.4 ¦ 8.96 or 9 Kg/day in
effluent
22
-------
CONFIGURATION
Though we tend to design rectangular cells due to ease of drawing and calculating, the wetlands
cells could as readily be trapezoidal or semi-circular. However, if irregular designs are used, the
widest portion should be located at the inlet end to facilitate equal flow distribution. Regardless
of the shape, marsh cells should have a 3-5:1 length to width ratio to reduce excessive loading at
the inlet end—as occurs in long, narrow cells—and to provide adequate length to initiate
nitrification after most of the BODs and TSS load has been removed. Wide, short cells perform
fairly well for BODBand TSS removal but poorly for ammonia removal.
Ponds may be rectangular, square, round, or broadly elliptical, but either extreme in length to
width ratios must be avoided unless complex inlet and collector distribution systems are used to
prevent short circuiting.
Meadows should be rectangular, with a 3-5:1 length to width ratio or greater if operated as
overland flow systems.
DIKES AND WATER CONTROL STRUCTURES
Each component of the wetlands system is basically a shallow pond or lagoon. Design and
construction techniques used for farm ponds or treatment lagoons are appropriate for general
features such as dikes, berms, and typical flashboard/stoplog water control devices. However,
dike freeboard must accommodate an organic matter (peat) accumulation rate of 2-3 cm/yr in the
marshes and should extend 75 cm above normal water level elevation for each 10 years of
projected operation. Adequate freeboard and water level control is also necessary to provide
capacity for flow beneath expected thickness of ice cover in colder climates.
Piping from the lagoon must be directed to a splitter box incorporating simple weir or flashboard
devices to adjust the inlet flow into each marsh cell. The inlet pipe for each marsh cell and the
23
-------
meadows should intercept the inlet distribution pipe at its midpoint in a "T" configuration, and
the distributor pipe must extend across 90 percent of the width of the cell. Inlet distributor pipes
must be level and supported by concrete stands 30-45 cm above the substrate. "T" pipe fittings
equidistant along the length of the inlet distribution pipe facilitate precise adjustment of flows
from each "T" ensuring even distribution of wastewaters across the width of the cell. Large
gravel or rock (10-15 cm) should be placed immediately below and in front of the inlet
distributor to reduce erosion.
Collector piping at the effluent end of each marsh cell and the meadow should be slightly below
the cell bottom in a gravel lined trench extending 90 percent of the width of the cell. The
discharge pipe may intercept the collector pipe at any convenient location but must terminate in a
weir/flashboard structure or an "elbow pipe" mounted with an "O" ring to permit swiveling of the
elbow up or down to maintain desired water depths in the marsh or meadow cells.
BOTTOM FORM AND LINERS
Bottom slopes for the marshes, pond, and meadow should be essentially flat. Width slope for the
marshes and meadow must be flat to ensure equal flow distribution of wastewaters. Length slope
may not be >0.05 % in the marshes and continuous flow application meadows. Meadows designed
for overland flow operation should have slopes of 2-3 percent.
Each component of the system should have an impermeable (hydraulic conductivity of
-------
CONSTRUCTION
Careful supervision of contractors is imperative to insure that grade and elevation specifications
are precisely met. Otherwise, considerable difficulty with short-circuiting and reduced treatment
capacity may occur and be impractical to correct during operation.
VEGETATION
Marshes should be planted with cattail or bulrush on 1 meter centers during the first half of the
growing season. Pondweeds in weighted cotton mesh bags and duckweed may be simply placed
in the pond. Perennial grasses suitable for the meadow include Reed Canary, Tall Fescue,
Redtop, Kentucky Bluegrass, Orchard Grass, Common Bermuda, Coastal Bermuda, Dallis Grass,
and Bahia. Marsh planting materials may be dug locally if suitable digging methods are used.
Within a natural stand of cattail or bulrush, remove the stems, rhizomes, and roots from a 0.5 m2
area, and then move over 1 meter before renewing digging. If this method is used before the
midpoint of the growing season, depleted areas will be recolonized by the end of the growing
season with little impact on natural wetlands.
Planting materials may also be obtained from commercial sources, but in either case, each root
stock must include 20-25 cm of stalk after removal of the tops. Failing to remove tops will result
in wind-throw before the roots have become established, and lack of a short stalk protruding
above the surface of the water will cause plant mortality if wastewater with little dissolved
oxygen is applied to the cell (Hammer, 1991).
Depending upon labor costs, digging local materials may be more costly than purchasing plant
stocks from a nursery or supplier. Generally, planting materials may be purchased for
$150/thousand from Wildlife Materials in Oshgosh, Wisconsin, and planting costs are $5,000-
6,000 per hectare. Digging locally and planting has cost up to $12,000 per hectare. Sago or other
25
-------
pondweeds (linear, filiform leaves) and tapegrass are usually planted as tubers and simply
weighted in mesh bags and dropped into the water or placed in soft, wet muds at desired
locations.
Local digging is facilitated by using a back hoe that removes the top 15-20 cm of substrate and
plant materials. Similarly, a trencher may be used to dig shallow trenches perpendicular to the
long axis of each cell with plants set into the trench and soil damped around them.
Planting bulrush or cattail is similar to planting any garden plant, but careful supervision is
important because plants with damaged roots or plants that are placed incorrectly will not
survive. Shallow flooding followed by draining, which leaves "soupy" substrates, creates ideal
growth conditions for new plant stocks and facilitates hand or mechanical planting. The area
should be flooded to 1 -5 cm after planting, but water levels must not overtop cut stalks from the
original plants or new shoot growth. After all planting is finished, water levels should be
gradually raised to normal operating elevations as the plantings grow higher, but water levels
must not overtop new growth during the first growing season. Emergent wetlands plants are not
as susceptible to drowning after the first growing season or in waters with relatively high
dissolved oxygen content.
Alternatively, cattail, and to some extent the other species, can be established by simply
manipulating the water levels at the appropriate time of the year. Or cattail and others may be
seeded, but germination rates are very low, e.g., 3-5 percent. Unfortunately, this method is
dependent on natural means of seed dispersal and germination and may require more than one
growing season to develop a dense stand.
26
-------
Grasses may be sown following standard recommendations. Full strength wastewater should not
be applied until grass height has exceeded 15-25 cm depending upon the type and variety.
Mosquito fish (Gambusia) and top minnows (Pimephales) or other bait fish may be introduced
into the marsh and pond after operating water depths have been stabilized. Bottom-feeding fish
(e.g., carp, catfish) should not be used, since low turbidity water is important to various
treatment processes.
OPERATION
Equal flow distribution across the width of the marsh and meadow cells is obtained by adjusting
the angle of each "T" outlet in the inlet distributor piping. Adjustment is done by inserting a
lever (short board) into each "T" and gently rotating the "T" to the proper elevation. Discharge
control structures (flashboard/"elbow pipe") must be adjusted to maintain 10-20 cm water depths
in the marsh cells, 0.8-1.3 m in the pond, and 1-5 cm in the meadow.
Routine weekly inspections are necessary to ensure equal flow from each "T" outlet on the inlet
distributor piping. If clogging occurs, rotate the "T" downward to drain the pipe, remove the
obstruction, and rotate it upward to the desired operating angle. Plant debris obstructing outlet
control structures may also need to be removed by similarly rotating the elbow piping or raking
from a flashboard device. Water levels in each component should be checked and adjusted as
necessary and all piping visually inspected for cracks or leaks. Dikes and flow control structures
should be inspected for leaks and corrective action implemented.
Flow distribution within cells should be occasionally inspected to detect channel formation and
short-circuiting and corrected by planting vegetation or filling soil in any channels. Vegetation
27
-------
should be visually inspected for signs of disease (yellowing/browning, spots, wilting, etc.), insect
infestations, or stress (stunted growth) during other inspection periods.
Shrubs or trees must be removed from the wetlands cells because they will shade out desirable
plant species. Weeding of herbaceous plants is unnecessary as is harvesting. Accumulated leaf
and stalk litter creates a desirable layer of humic materials on the surface of the cells within
which much of the wastewater treatment occurs. Foot traffic should be minimized and vehicular
traffic prevented within the cells because either will compress and damage the humic/compost
surface layer. Pesticides or other chemicals that may harm the vegetation should not be directly
applied or introduced into the wastewater stream.
NUTRIENT/SEDIMENT CONTROL SYSTEM
The nutrient/sediment control system combines marsh/pond components of constructed wetlands
with other erosion/sediment management elements to use physical, chemical, and biological
processes for removal of sediment and nutrients from runoff. This system was originally
designed for potato fields in northern Maine, but it could be easily adapted to pasture or crop
field runoff as well as to urban stormwater runoff in other regions. Though it may be used in
any small watershed, system performance and longevity will be considerably enhanced if standard
erosion control practices - grassed waterways, terraces, no-till cultivation, sediment basins, filter
strips, diversions, etc. - are put in place prior to locating the nutrient/sediment system. Land
area requirements range from 0.5-1.4 ha for 23-68 ha watersheds, respectively. In addition,
functioning is not dependent upon a minimum critical size, so a number of units can be
judiciously located in small or large watersheds to accomplish treatment of virtually all runoff.
28
-------
Components of the system include a sediment basin, level-lip spreader, primary grass filter,
wetland (marsh), deep pond, and a polishing filter (Figure 7). The sediment basin is a trapezoidal
trench designed to collect larger sediment and organic particles and to regulate flow to the
remainder of the system. The level-lip spreader consists of a trench filled with crushed rock or
sorted stone that provides sheet flow to the third component. The primary grassed filter is a
modified overland flow unit with lower gradient, broader width, and tile subsurface drains for
fine sediment and nutrient removal. The freshwater wetland consists of an emergent marsh
similar to the constructed wetlands described above, which along with the deep pond (2-5 m)
functions primarily for nutrient removal. The final component, a polishing filter, generally
consists of a natural wet meadow, shrub, or wooded area for removal of algae and some nutrients.
If excessive runoff is anticipated (i.e., an urban watershed), a small retention basin above a
smaller sediment ditch or in leu of the ditch may be added, but the basin/ditch and level-lip
spreader must provide sheet flow to the primary grass filter for proper operation.
These systems are typically situated at the downstream end of grassed waterways or other small
water courses prior to their junction with intermittent or permanent streams, rivers, and lakes.
Consequently, an individual farm may require a number of small, judiciously placed units to
accomplish complete control of nutrient and sediment removal. However, the minimal acreage
and flexible design requirements are amenable to a variety of topographic, land use, soil, and
meteorological conditions. Sizing is related to size of watershed and anticipated runoff (Table 2).
Plant species include grasses appropriate to the region that will provide good ground cover during
the period of highest expected runoff (i.e., cool season grasses in northern regions and fescue or
Bermuda strains in the southeast). Cattail, bulrush, or rushes may be used in the marsh and the
transition zone with the pond supports submergent wetlands species—pondweeds, tape grass, etc.
29
-------
Recommended maintenance consists of removal of accumulated sediment in the ditch and grass
from the primary grass filter. If extensive algal mats develop in the pond, their removal during
maximum growth will enhance nutrient removal. In addition, if the pond is used for bait fish
production, it will enhance nutrient transport out of the system and financial return to the
landowner. Obviously, larger fish, especially bottom feeders such as carp or catfish, would be
detrimental to sediment removal and biological processes and should not be used.
In practice, nutrient/sediment control systems have removed 90-100 percent of suspended solids,
85-100 percent of total phosphorus, 90-100 percent of BOD6, and 80-90 percent of total nitrogen
from potato field runoff in northern Maine. Construction costs have ranged from $13,500 to
$23,000 for 8 ha and 67 ha watersheds, respectively (Wengryznek, pers. comm.).
These costs, amortized over the expected life of the system, plus annual maintenance average only
$50-$60 per ha per year for the contributing watershed. This is a very favorable factor when
compared to the potential costs of less effective and non-productive conservation practice
alternatives.
Nutrient/sediment control systems should not be used to replace conventional BMPs but should
complement a sound land and water management program in each watershed.
OTHER CONSIDERATIONS IN CONSTRUCTED WETLANDS TREATMENT SYSTEMS
ADVANTAGES
Advantages of constructed wetlands include relatively low construction costs—essentially grading,
dike construction, and vegetation planting with little steel or concrete—and low operating costs.
Maintenance consists of monitoring water levels and plant vitality, collecting NPDES samples,
30
-------
and grounds maintenance (mowing dikes and roadways). Properly designed and constructed
systems do not require chemical additions, internal pumping, sludge handling, or other
procedures of conventional treatment systems. Neither do they require plant harvesting, except
in specialized applications using floating plants—water hyacinth (Eichhornia) or duckweed
(Lemna) for nutrient removal after conventional treatment. In these cases, maintenance costs
may be very high.
Typically construction costs range from 1/10 to 1/2 of the cost of comparable conventional
treatment systems. For example, a TV A designed system at Benton, Kentucky, which polishes
primary lagoon effluent, cost $260,000 in 1986 compared to a 1972 estimate of $2.5 million for a
comparable conventional treatment system. Two other systems designed for secondary and
tertiary treatment for communities of 500 (Hardin) and 1,000 (Pembroke) users varied from
$212,000 to $366,000. Operating costs for these systems are less than $10,000 per year. A TVA
wetlands that controls acid mine drainage cost $28,000 to construct and plant—about the same as
the costs for chemicals alone to provide comparable treatment for only 1 year (Brodie, et al,
1988). Operating costs, other than monitoring sample collection and analysis, have been less than
$500 per year.
Wastewater treatment efficiencies are very good, especially for BOD6, TSS, and fecal coliform
bacteria with common discharge values of 10-20 mg/L and 50-100 colonies per 100 ml. With
proper design and adequate treatment area, removal of nitrogen compounds and phosphorus are
readily accomplished. As can be expected, performance varies with different designs, wastewater
sources, amount, and type of pre-treatment- and treatment-area/retention times, with the most
variation related to the type of system and treatment-area/retention times. Constructed wetlands
are also amenable to substantial fluctuations in loading rates, adapting to weekly and annual
31
-------
fluctuations in flows. This adaptation of constructed wetlands to fluctuations in loading can be
observed, for example, from a high school in northwest Alabama.
Constructed wetlands can provide ancillary benefits in the form of wildlife habitat, recreational
and environmental space, or simply urban greenspace. Recreational activities derive from
vertebrates, larger invertebrates, and, to some extent, vegetational components. For example, the
Areata system in California has been described as a bird watching "hotspot" in a national "birding"
publication (Gearhart et al., 1989). During a Sunday visit to the Martinez, California, wetlands
system, although available police and janitorial employees were unable to direct me to the
wastewater treatment plant, a chance encounter with a local birdwatcher saved the day. Amateur
naturalists and environmental educators are able to quickly identify and exploit the educational
and recreational benefits of a simulated marsh nearby (James and Bogaert, 1989). Systems located
near urban areas may also provide greenspace benefits (Smardon, 1989) or simply open, natural
areas that attract a variety of low-intensity recreational users - walking, jogging, picnicking,
relaxing, etc. Treatment system operators, pleased with the attention and support received from
local citizens, usually welcome ancillary uses. More importantly, many realize that recreational
benefits and pleasing aesthetics of wetlands systems may reduce opposition to new or expanded
systems.
However, wetlands constructed for wastewater treatment, at least initially, are comparatively
simple, often monotypic species systems. A properly designed and constructed cell with adequate
treatment area that is covered in a dense stand of Typha or Scirpus will efficiently remove target
contaminants from influent waters while providing habitat for a few muskrats, blackbirds, and
some songbirds but little else (Hammer, 1989b). Even if operated at maximum efficiency,
however, it will not have adequate capacity to store flood waters or release substantial quantities
to amplify low stream flows in dry conditions. Wastewater treatment has been maximized
32
-------
through optimized design and operating criteria, and all other functional values have been
subordinated. But the water improvement function is still efficient and enduring even though
other wetlands functional values are substantially reduced or nonexistent.
DISADVANTAGES
Constructed wetlands are land intensive—they require much more land area than do package
treatment plants—and they require relatively level surfaces. However, nutrient/sediment systems
need relatively less land area and may be constructed in watershed locales with considerable
relief. Where land costs are high such as in larger cities or in very rugged terrain that would
need considerable cut and fill constructed wetlands are more expensive to construct than
conventional systems, although lower operating costs over a 20 or 30 year plant lifetime must be
factored into the decision process. Current mass-loading design recommendations require 6-20
ha of treatment area per 4,000 ms, depending upon the level of pre-treatment and the desired
discharge limits.
In addition, present design, construction, and operating criteria are imprecise—the reason for the
range of treatment area requirements in the above. And wetlands systems, either natural or
constructed, are complex, dynamic systems about which we have only limited understanding,
However, a number of experimental and operating systems throughout the world are beginning to
accumulate the data base from which precise design and operating criteria will be developed.
Another disadvantage is delayed operational status. Because peak removal efficiencies of
constructed wetlands are dependent upon vegetation growth and establishment, design
efficiencies are not likely to be attained until after two or perhaps three growing seasons.
Completing construction and planting does not immediately translate into full operational status,
and potential users must plan to gradually phase in wetlands operation concurrent with phase out
33
-------
of the existing conventional system. Proper operation of the existing plant during this phase can
be critical to avoid discharging activated sludge during high flow periods or simply "wasting"
sludge to the wetlands system before it is fully operational.
Treatment system longevity is poorly documented, since no successful operating-scale system has
been in operation for more than 20 years. Because these systems simulate natural wetlands
ecosystems that have functioned to purify water for thousands of years, I expect that system
efficiency is not likely to be detrimentally impacted by age, but artificial constraints may require
modifications or restarting after some period of time. Litter/detritus accumulation rates have
been measured at about 2 cm/year in municipal systems with no loss of treatment efficiencies
(Haberl and Perfler, 1989). Therefore, designs should incorporate this accumulation factor in
dike height specifications and dikes should have 50-55 cm of freeboard for a 20-year operating
lifetime, or greater for longer operational status. At that point, the system may need to be
cleaned out and restarted, and after testing to identify possible toxic substances, accumulated
litter may be composted or land applied similar to conventional sludge.
Finally, improperly designed or operated constructed wetlands could create pest problems-
mosquitoes or rodents-that may cause adverse impacts to local residents. Both are easily avoided
with appropriate designs and operating procedures (Martin and Eldridge, 1989)
TOXICS ACCUMULATION
Since wetlands treatment systems transform or remove pollutants from inflowing waters, the
ultimate fate of certain substances within the wetlands ecosystem is of more than academic
interest (Hammer. 1990). Depending upon the source, influents may contain various natural and
anthropogenic organic compounds, metals including heavy metals, path08e„ic organismJ) !alts,
etc. A few materials (e.g., selenium) are selectively taken up by plants, but most are precipitated
34
-------
or complexed within the substrate. Generally, only 4-5 percent of the nutrient loading on a
wetlands system is incorporated into plant or animal tissue. However, some metals may occur in
relatively high concentrations. For example, iron levels as high as 5000 mg/kg and manganese
levels up to 4100 mg/kg were present in cattail leaves and stems grown in experimental cells that
were heavily loaded with acid mine drainage. Only traces of other metals were present (TVA,
unpublished data). Copper was nonexistent in cattail from a natural marsh but averaged 6.1
mg/kg in cattail from two municipal wastewater treatment systems. Higher concentrations of
lead were found in a natural cattail stand (1.7 mg/kg) than in cattail from the municipal systems
(0.3 gm/kg).
Though only low levels of potential toxic metals occurred in these samples, long-term effects of
relatively high levels of iron and manganese are not known. In the short term, iron and
manganese did not appear to have detrimental effects on cattail growth and vitality in the
experimental cells. In fact, plants in the upstream portion of each cell were more robust than
plants in the lower sections. Upper portions of each cell received raw inflowing acid mine
drainage that probably contained small concentrations of micronutrients in addition to
substantially higher concentrations of iron and manganese. Differential robustness within each
cell was likely due to micronutrient uptake in the upstream portion and limited micronutrient
availability to plants in the lower sections.
GENERAL CONSIDERATIONS
Because constructed wetlands are open, outdoor systems, they receive inputs of animal and plant
life from adjacent areas and from distant sites. Over time are likely to become more and more
similar to other naturally occurring wetlands in a region (Hammer and Bastian, 1989). Though
we may design and build a system with a specific substrate and only one or two plant species
currently thought to be highly efficient, over time, many types of plants and animals will take up
35
-------
residence. Consequently, a constructed wetlands is likely to become more similar to a natural
wetlands as the system matures and ages. To prevent these invasions and attempt to maintain a
monoculture would be difficult and costly and may be self-defeating. Living organisms that
become established in an operating system are not likely to detrimentally impact treatment
efficiency and may very well improve system operation. In addition, maintaining a monoculture
is difficult, as any farmer knows, since the single species stand or monoculture is often
susceptible to disease, insects, or grazing animals. For example, cattail in a marsh that was
treating acid water seepage from coal ash storage ponds was devastated by an outbreak of
army worms during the first year of operation (Snoddy et al., 1989).
Much can be learned from both natural wetlands receiving wastewaters and from constructed
systems with a few years of operating history. Despite some attempts to reduce wetlands
treatment systems to minimal components and simplify treatment areas by using the most
efficient combinations of substrate, vegetation, and loading rates, most successful systems are
indistinguishable on casual examination from natural marshes. In fact, poorly performing systems
that I have visited did not appear to be viable marsh ecosystems. Generally, the absence of an
important component, attribute, or characteristic was obvious to anyone with experience in
natural marsh ecosystems. Conversely, successful systems are often quite similar to a natural
marsh, and it s beginning to appear that the basis for design of wetlands constructed for
wastewater treatment should be to simulate the structure and functions of a natural marsh
ecosystem (Hammer, 1991).
This is not to suggest that natural wetlands are casually useable for wastewater treatment. We
know too little of the complex interractions among a myriad of components in natural systems
and too little of optimal treatment area requirements or application rates to risk damaging natural
wetlands. These systems are too valuable to lose, since research on natural wetlands systems will
36
-------
continue to increase our understanding and ability to design and build constructed wetlands for
specific purposes. Intact natural wetlands also provide a host of other benefits to society.
Many are searching for an inexpensive, efficient wastewater treatment process, and constructed
wetlands are an attractive alternative. But constructed wetlands are not a panacea for all
wastewater treatment problems. Although experimental work has been underway for more than
30 years, the technology is still in its infancy and much remains to be learned on design,
construction, and operation (Hammer, 1989c). Most previous work has been directed towards
municipal wastewater treatment, and our information is adequate to conservatively design and
operate systems for that use. Within the last 5 years a number of experimental and operating
systems treating acid mine drainage have provided similar information. But the substantial
potential for treating NPS pollution especially urban stormwater runoff and agricultural
wastewaters, industrial wastes, and even failed septic tank drain fields at individual home sites
remains to be developed.
Wetlands accomplish water improvement through a variety of physical, chemical and biological
processes operating independently in some circumstances and interacting in others. Vegetation
obstructing the flow and reducing the velocity enhances sedimentation, and many substances of
concern are associated with the sediment because of clay particle adsorption phenomena.
Increased water surface area for gas exchange improves dissolved oxygen content for
decomposition of organic compounds and oxidation of metallic ions. But the most important
process is similar to decomposition occurring in most conventional treatment plants-only the
scale of the treatment area and composition of the microbial populations is likely to be different.
In both cases an optimal environment is created and maintained for micro-organisms that conduct
desirable transformations of water pollutants. Wetlands systems use larger treatment areas to
37
-------
establish self-maintaining systems that provide environments for similar microbes but also
support additional types of micro-organisms because of the diversity of microenvironments. The
latter, along with a larger treatment area, frequently provide more complete reduction and lower
discharge concentrations of water-borne contaminants. Regardless, most removal or
transformation of organic substances in municipal wastewaters or metallic ions in acid mine
drainage is accomplished by microbes—algae, fungi, protozoa, and bacteria.
Conventional wastewater treatment systems require large inputs of energy, complex operating
procedures, and subsequent costs to maintain optimal environmental conditions for microbial
populations in a small treatment area. The low capital and operating costs, efficiency, and self-
maintaining attributes of wetlands treatment systems result from a complex of plants, water, and
microbial populations in a large enough land area to be self-sustaining. It may be less costly to
construct a minimally sized, least-component wetlands treatment system, but operational costs to
maintain that system could easily negate initial cost savings. However, for small communities,
farms, mines, and some industries, a conservatively designed and biologically complex system
may provide more efficient treatment, greater longevity, and reduced operating requirements and
costs.
38
-------
REFERENCES
Amaral, M. R., 1989. The Impact of the Floods of the Pantanais Matogrossenses. Comissao de
Defesa do Pantanal, Rua Cuiaba No. 440, Corumba, MS, Brazil. 30pp.
Anon, 1991. Nutrient/Sediment Control System (Draft Interim Standard and Specification).
CODE 1-05. USDA-SCS. Orono, ME.
Brix, H. and H.-H. Schierup, 1989. Danish experience with sewage treatment in constructed
wetlands, pp. 565-573. In: D. A. Hammer (ed.), Constructed Wetlands for Wastewater
Treatment. Lewis Publishers, Inc., Chelsea, MI.
Brodie, G. A., D. A. Hammer, and D. A. Tomljanovich, 1988. Man-Made Wetlands for Acid
Drainage Control in the Tennessee Valley. Proc. Mine Drainage and Surface Mine Reclamation,
1:325-331. Bur. Mines Inf. Cir. 9183.
Cadavid, E. A., 1989. Conservation Plan for the Upper Paraguay Basin, Brazil. Secretaria
Estadual de Meio Ambiente, Governo do Estado de Mato Grosso do Sul, Campo Grande, MS,
Brazil. 69pp.
Gearheart, R. A., F. Klopp, and G. Allen, 1989. Constructed free surface wetlands to treat and
receive wastewater: Pilot Project to full scale, pp. 121-137. In: D. A. Hammer (ed.),
Constructed Wetlands for Wastewater Treatment. Lewis Publishers, Inc., Chelsea, MI.
39
-------
Haberl, R. and R. Perfler, 1989. Root-zone system: Mannersdorf-new results, pp. 606-621. In:
D. A. Hammer (ed.), Constructed Wetlands for Wastewater Treatment. Lewis Publishers, Inc.,
Chelsea, MI.
Hammer, D. A., 1989a. Constructed wetlands for treatment of agricultural waste and urban
stormwater. In: S. K. Majumdar et al. (eds.), Wetlands Ecology and Conservation: Emphasis in
Pennsylvania. The Pennsylvania Academy of Science.
Hammer, D. A., 1989b. Protecting Water Quality with Wetlands in River Corridors. Proceedings
of the International Wetland Symposium on Wetlands and River Corridor Management. July 5-
9, 1989, Charleston, SC, The Association of Wetlands Managers, Inc.
Hammer, D. A., (ed.) 1989c. Constructed Wetlands for Wastewater Treatment. Lewis Publishers,
Inc., Chelsea MI. 831pp.
Hammer, D. A. and R. K. Bastian, 1989. Wetlands ecosystems: Natural water purifiers? pp. 5-
19. In: D. A. Hammer (ed.), Constructed Wetlands for Wastewater Treatment. Lewis Publishers,
Inc., Chelsea, MI.
Hammer, D. A., B. P. Pullin, and 3. T. Watson. 1989. Constructed Wetlands for Livestock Waste
Treatment. Proceedings of the National Nonpoint Conference, St. Louis, MO, April 23-26, 1989.
Hammer, D. A., 1990. Water Improvement Functions of Natural and Constructed Wetlands.
Proceedings of the Newman Teleconference Seminar Series - Protection and Management Issues
for South Carolina Wetlands, pp. 129-157. Clemson University, March 28, 1990.
40
-------
Hammer, D. A., 1991. Creating Freshwater Wetlands. Lewis Publishers, Inc., Chelsea, MI. (in
press).
Higgins, M., 1991. The Use of Constructed Wetland Systems in Treating Agricultural Runoff:
1990 Data Summary. Report of the Department of Civil Engineering, Univ. of Maine, Orono,
ME.
James, B. B. and R. Bogaert, 1989. Wastewater Treatment/Disposal in a Combined Marsh and
Forest System Provides for Wildlife Habitat and Recreational Use. pp. 597-605. In: D. A.
Hammer (ed.), Constructed Wetlands for Wastewater Treatment. Lewis Publishers, Inc. Chelsea,
MI.
Kadlec, J. A., R. H. Kadlec and C. J. Richardson, 1974. The Effects of Sewage Effluent on
Wetland Ecosystems. Research Applied to National Needs. Grant GI-34812X, Univ. of
Michigan, Ann Arbor, MI.
Martin, C. V. and B. F. Eldridge, 1989. California's experience with mosquitoes in aquatic
wastewater treatment systems, pp. 393-398. In: D. A. Hammer (ed.), Constructed Wetlands for
Wastewater Treatment. Lewis Publishers, Inc., Chelsea, MI.
Miller, G., 1989. Use of Artificial Cattail Marshes to Treat Sewage in Northern Ontario, Canada,
pp. 636-642. In: D. A. Hammer (ed.), Constructed Wetlands for Wastewater Treatment. Lewis
Publishers, Inc., Chelsea, MI.
Mitsch, W. J. and J. G. Gosselink, 1986. Wetlands. Van Nostrand Reinhold Company, New
York, NY. 539pp.
41
-------
Odum, H. T. and S. Brown, 1976. Regional Implications of Sewage Effluent Application in
Cypress Domes. Freshwater Wetlands and Sewage Effluent Disposal Symposium. May 1976.
Univ. of Michigan, Ann Arbor.
Pullin, B. P. and D. A. Hammer, 1989. Comparison of Plant Density and Growth Forms Related
to Removal Efficiencies in Constructed Wetlands Treating Municipal Wastewaters. 62nd Ann.
Conf., Water Pollution Control. Federation, San Francisco, CA, Oct. 1989.
Reddy, K. R. and W. H. Smith (eds.), 1987. Aquatic Plants for Water Treatment and Resource
Recovery. Magnolia Publishing, Orlando, FL. 1032pp.
Reed, Sherwood C. (ed.), 1990. Natural Systems for Wastewater Treatment - Manual of Practice
FD-16. Water Pollution Control Federation, Alexandria, VA. 270pp.
Seidel, K., 1971. Macrophytes as functional elements in the environment of man. Hydrobiology,
20(1): 137-147.
Small, M. M., 1977. Natural Sewage Recycling Systems. Brookhaven National Laboratory Report
60530, Upton, NY.
Smardon, R. C., 1989. Human perception of utilization of wetlands for waste assimilation, or
how do you make a silk purse out of a sow's ear? pp. 287-295. In: D. A. Hammer (ed.),
Constructed Wetlands for Wastewater Treatment. Lewis Publishers, Inc., Chelsea, ML
Snoddy, E. L., G. A. Brodie, D. A. Hammer, and D. A. Tomljanovich, 1989. Control of the
armyworm, Simvra henrici (Lepidoptera: Noctuidae), on cattail plantings in acid drainage
-------
treatment wetlands at Widows Creek Steam-Electric Plant, pp. 808-811. In: D. A. Hammer
(ed.), Constructed Wetlands for Wastewater Treatment. Lewis Publishers, Inc., Chelsea, MI.
43
-------
Table 1. Livestock Waste Production
Production
per Day
(grams)
BOD5 N P
Total Volume
(meter8)
Dairy Cows 773 186 33 0.05
(455 Kg
Swine 180 204 68 0.03
(91 Kg)
Poultry-layers 6 ].3 0.5 0.0001
(1.8 Kg)
Poultry-broilers 3.3 0.7 0.3 0.00005
(1 Kg)
44
-------
Table 2. Minimum Sizing Criteria for a Nutrient/Sediment Control System for Runoff
from Typical Agricultural Rowcrop Fields
Watershed Sediment Grassed M2- h Deep Polishing
Size* Ditch Filter Pond Filter
(ha) (m2) (m2) (m2) (m2) (m2)
<10 70 700 925 925 1000
20 90 925 1200 1500 1400
30 115 1200 1400 2000 1700
40 140 1400 1600 2600 2100
60 185 1900 2100 3700 2800
•For larger watersheds, size of the components is increased proportionally to the drainage area.
45
-------
FIGURE LEGENDS
Figure 1. A schematic map of the Pantanal, Mato Grosso do Sul, Brazil.
Figure 2. First order treatment wetlands are located onsite in proximity to the wastewater source
while second order control is remotely situated to provide treatment of combined runoff before it
reaches a permanent stream.
Figure 3. Each farm has first-order and second-order control systems, and permanent streams
have third-order systems strategically located. Final polishing is provided by fourth-order
systems consisting of large natural wetlands in the lower reaches of the watershed.
Figure 4. An experimental wetlands wastewater treatment system for a 500-hog operation at
Auburn University's Sand Mountain Experiment Station.
Figure 5. An operational wetlands wastewater treatment system for a 500-hog operation at
Mississippi State University's Pontotoc Experiment Station.
Figure 6. The marsh-pond-marsh design concept to enhance ammonia removal in constructed
wetlands treatment systems.
Figure 7. The nutrient/sediment control systems developed by the Soil Conservation Service in
Maine.
46
-------
no
¦ fAZ MNU COVIMOlCt
*hl RtNCNO NOVO
»U »Al* «MCIHA
COXIM
BOLIVIA
com; mba
AQUIOAUANA
CAMPO GRANDE
BRAZIL
MATO GROSSO DO SUL
t • N m M MIC
_n__j—i
Figure 1
-------
Crop Field
Crop Field
Grassed Waterways
House
. Dairybarn
Septic Tank
Lagoons
Dry Stack
Swine Barn
Wetlands*
Cells
Pasture
Pasture
Crop Field
Crop Field
Nutrient/Sediment Treatment
Nutrient/
Sediment
Treatment
- - —
Woodlot
-------
\
******
y ««¦******
**********
**********
*****j
***j
ylW,\
Key
Permanent
stream
Riparian
buffer
tTTTTTT
waterways
Nutrient/
sediment
system
Farm pond
Constructed
wetland
treatment
system
Livestock
facilities and
lagoon
Natural
wetlands
Figure 3
-------
Stormwater
Pond
Barn
Barn
Lagoon
Lagoon
Mixing
Pond
Control
Juncus
Mixed
Scirpus
Typha
Figure 4
Wet
Meadow
-------
Mars
_Cell$
*" * ^ *&3L
\ *>.
^ ^ s\i
f *j\\^ ^ms|s Y^ » i
Overland
Flow
Cells
Lagoons
Swine
Barn
Figure 5
-------
Figure 6
-------
Inflow
)>/)/>>>>>>>>)>}>»>>>}>>>)>)
Sediment Level-Lip
Ditch Spreader
Grass
Filter
Marsh
Pond
Grass
Filter
(Side
View of
System)
V////////////////////A
Figure 7
-------
DEVELOPING DESIGN GUIDELINES FOR CONSTRUCTED WETLANDS TO REMOVE
PESTICIDES FROM AGRICULTURAL RUNOFF
John H. Rodgers, Jr.
and
Arthur Dunn
Department of Biology
Biological Field Station
University of Mississippi
University, MS 38677
ABSTRACT
This paper presents a research strategy for evaluating the capability of constructed, restored, and
natural wetlands to assimilate and process pesticides associated with agricultural runoff from
croplands. A modeling approach that is central to this research strategy is presented and the
mathematical foundation is explicitly stated. This approach generates predictions that can be
experimentally and rigorously tested. Criteria for selection of "model" pesticides for
experimentation include factors such as use patterns and amounts as well as intrinsic
characteristics of the pesticide. The design of the experimental constructed wetland cells for this
research includes water flow and depth control, clay liners to prevent infiltration, and wetland
vegetation as a variable. The experimental strategy should permit optimal transfer of study
results from site to site and ultimately provide recommendations for pesticides that are
compatible with wetlands as well as design characteristics for constructed wetlands to be used
with specific crop-pesticide combinations.
1
-------
COMPARISON OF WATER QUALITY FUNCTIONS OF CREATED, RESTORED AND
NATURAL WETLANDS
Wetlands have numerous functions and values including water quality improvement (Greeson et
al., 1979; Mitsch and Gosselink, 1986; Hammer and Bastian, 1989). Protection, creation, and
restoration of wetlands may be of great value in integrated strategies for controlling nonpoint
source (NPS) runoff. As interfaces between cropland and water, some wetlands are subject to
episodic NPS agricultural runoff (Wauchope, 1978). These wetlands can retain and process many
NPS contaminants (Cooper, 1989). When contaminants such as pesticides are processed in
wetlands, their impacts are not subsequently realized in downstream lakes, rivers, streams, and
reservoirs.
NPS agricultural contaminants entering wetlands may include physical, chemical, or biological
contaminants (Wauchope, 1978; Gersberg et al., 1987). Physical contaminants include materials
such as particulates or detritus. Chemical contaminants include pesticides, organics, and
inorganics such as nutrient salts. Biological contaminants include bacteria from animals such as
dairy cattle and chickens, as well as protozoa and other pathogens (Portier and Palmer, 1989).
Probably the most important NPS agricultural contaminants causing problems in receiving systems
are particulates, nutrients, and pesticides. (Wauchope, 1978; Wallach, 1991). Frequently,
pesticides that are washed from croplands by rainfall and cause adverse impacts on adjacent
waters come from fields where a wetland buffer strip was not maintained. In these instances,
water quality problems and even fish kills may be a consequence of the loss of the wetland
(Wolverton and Harrison, 1975).
The objective of this paper is to present, by example, a strategy for developing design criteria for
wetlands to control NPS runoff from agricultural systems with emphasis on pesticides. This
2
-------
research strategy examines factors affecting the limits of wetlands for assimilating agricultural
pesticides. A long-term goal is to couple the appropriate pesticides with correctly designed
constructed or restored wetlands for maximal retention and processing of NPS pesticides.
RESEARCH FOUNDATION AND APPROACH
Farmers in the southeastern United States have developed for agricultural use most of the
available land bordering rivers, streams, reservoirs, and lakes. The notion of "plowing with one
wheel of the tractor in water" became popular in order to maximally utilize the rich flood plain
soils. These former wetland areas have been lost, particularly in the Mississippi Valley, at an
alarming rate (Horwitz, 1978). The research described in this paper is based on the premise that
wetlands' functions and values ranging from their chemical and material processing ability to
their values for wildlife propagation can be coupled to mitigate the effects of NPS agricultural
runoff, particularly pesticide components. Since wetlands have a unique ability to retain and
process materials, it seems reasonable that constructed or restored wetlands as buffer strips
between agricultural activities and receiving aquatic systems could mitigate impacts of pesticides
in this runoff.
The factors influencing pesticide fate and effects in agricultural systems and wetlands have been
a matter of fairly recent investigation (Wauchope, 1978; Reinert and Rodgers, 1984). We
currently have a relatively good understanding of the characteristics of aquatic systems as well as
the characteristics of pesticides that regulate their fate and effects. Modeling of pesticide fate in
agricultural systems and adjacent wetlands yields predictions of pesticide runoff in aqueous and
particulate form, biotransformation rates, wetland retention, photolysis rates, hydrolysis rates,
sorption, and other factors that may influence pesticide fate (Reinert and Rodgers, 1984). This
knowledge can be used to enhance or select for factors in wetlands that promote processing of
3
-------
pesticides (Table 1). The research described in this paper consists of four parts. First, a
modeling effort was undertaken, focusing on characteristics of pesticides and wetlands that would
be important in agricultural pesticide retention and transformation. Next, pesticides were
selected for testing and evaluation. Third, experimental wetlands were designed and constructed
to test hypotheses regarding the fate of pesticides in wetlands. And fourth, hypotheses are being
tested and conclusions drawn.
MODELING EFFORT
The pesticide transfer and transformation model used to guide this research is based on wetland
physical, chemical and biological characteristics and processes that regulate the fate and
persistence of a pesticide (Table 1). The rate of removal of a pesticide in a wetland is compared
with the pesticide's residence time (PRT) in the wetland to determine the propensity of a wetland
to reduce the concentration of a pesticide in downstream aquatic systems and thereby to mitigate
subsequent nontarget effects. For this model, the PRT is defined as the time available in the
wetland for pesticide transfer and occurrence of transformation processes such as volatilization,
hydrolysis, photolysis, and biotransformation (Reinert and Rodgers, 1984). The PRT is a
function of: (1) runoff events that add more pesticide to the wetland, or (2) processes that
transfer the pesticide out of the wetland. Although this is a simplified view of pesticide
transport (Wauchope, 1978), it will suffice until we have sufficient data from wetlands to permit
more sophisticated or complicated analysis.
The pesticide transfer and transformation model used for these wetland studies is a first-order or
pseudo first-order mass balance model in which the pesticide transfer and transformation rate is
directly proportional to the pesticide concentration. In this model, the overall transformation and
transfer rate coefficient (KT) is an aggregate of the individual rate coefficients for the processes
4
-------
noted above (e.g., hydrolysis, biotransformation, etc.). The implication of the model is that
pesticide removal processes in wetlands can be adequately described by simple exponential decay,
and it perhaps implies a simpler mechanism than actually occurs. As fate and persistence data are
accumulated for pesticides in wetlands, the frequency of deviations from simple exponential
behavior can be evaluated. However, it should be noted that similar modeling efforts have been
very useful and provided predictive capabilities for wastewater processing in wetland situations
(Bavor et al., 1989; Kadlec, 1989; Steiner and Freeman, 1989). To date, the mass balance
approach with exponential modeling has been accurate for a variety of pesticides and wetlands
(Reinert and Rodgers, 1984; Reinert et al., 1988; Cassidy and Rodgers, 1988).
The first-order or pseudo first-order model of pesticide fate (transfer and transformation) in a
wetland is the following:
where Ct and Ct are pesticide concentration at time t and the initial concentration, respectively,
and where 6 is the specific base of the decay function. For this exponential model, Kx is the
overall transfer and transformation rate coefficient with units of reciprocal time. A useful
expression of KT is the pesticide transformation and transfer half-life (rp which is the time
required to reduce the mass of a pesticide to 50 percent of its original value. Mathematically:
For experimental and design purposes relative to this study, we can express the model for
pesticide transfer and transformation in wetlands as a basic equation:
Ct« Q rKTl
(1)
(2)
5
-------
Łt = e "°-693 (PRT/Tj)
Ci
(3)
CT/ C; is the removal ratio for a pesticide at the end of a given PRT.
Explicitly stated, the assumptions behind the pesticide transfer and transformation model are: (1)
pesticide transfer and transformation in wetlands follows first-order or pseudo first-order
kinetics, and (2) transport of a pesticide in a wetland can be reasonably approximated by a single
number, the PRT. The PRT in a given wetland will be determined by the characteristics of that
wetland such as plant density, porosity, wetland dimensions, water flow, pesticide retention
factors such as sorption, etc., as well as runoff events that influence pesticide inputs to the
wetlands. The residence time of a pesticide in a wetland is also a function of the intrinsic
chemical character of the pesticide. As noted above, these assumptions will probably permit
accurate predictions for most situations until more data are available to indicate more complicated
or appropriate approaches.
Predictions from the pesticide transfer and transformation model are summarized in Figures 1
and 2. In Figure 1, the T± is related to the PRT and pesticide removal in wetlands is estimated.
For example, to achieve at least 95 percent removal of a pesticide in a wetland with a PRT of 1
month, the critical T± is approximately 7 d. On the other hand, if the PRT is 6 months, the
required for 95 percent removal may be considerably longer, on the order of 50 d. If the of a
pesticide and the PRT in a wetland are 15 and 50 d, respectively, then one would expect removal
of at least 90 percent. When these pesticide removal rates are compared to environmental
toxicology information, the mitigation capabilities of wetlands for reducing effects of pesticides
on downstream aquatic systems can be predicted.
6
-------
The modeled relationship of pesticide removal with respect to T± and PRT is illustrated in Figure
2. This figure shows the amount of pesticide remaining as a function of the 7^/PRT ratio. From
this illustration, it is clear that significant pesticide removal (< 50 percent remaining) will occur
when the equals the PRT (7^/PRT ¦ 1). Pesticide removal increases still further (or the
amount of pesticide remaining decreases) as the becomes a smaller fraction of the PRT
(7yPRT < 1). At 7yPRT ratios of 3.0 or less, removal is essentially linear. A further
prediction from the model that can be derived from Figure 2 is that the utility of wetlands for
pesticide removal diminishes rapidly when the exceeds the PRT. At 7^/PRT ratios greater
than 1, the change in required to significantly reduce the amount of pesticide remaining in a
wetland is large relative to that required when the 7^/PRT is less than 1. For example, at a
7"^/PRT ratio of 8, less than 10 percent of the pesticide will be removed in the wetland.
Pesticides with these 7^/PRT ratios will not be practically removed by wetlands. Another key
practical observation from this modeling effort is that pesticide removal efficiency in different
types of wetlands may vary strongly with factors that influence residence times in differing
wetlands and physical, chemical and biological factors that regulate pesticide half-lives. In
addition, it was apparent from the modeling effort that one could predict the physical dimensions
of a wetland necessary to obtain a required pesticide removal fraction if the relationships between
PRT and wetland characteristics such as vegetation density and hydrosoil characters are known.
Clearly, this would be a profitable area for further study.
The mass balance approach for pesticide fate is used in this research strategy to evaluate the role
of wetlands in mitigating the effects of pesticides in NPS runoff. During the initial sensitivity
analysis of the model, it was apparent that processing of the pesticide would be relative to the
retention of the pesticide in the wetland. Retention of the pesticide in the wetlands was a
function of sorption and contact time in the wetland. Sorption is largely a function of the
available mass of plants or vegetation in the wetland as well as the hydrosoil character (Reinert
7
-------
and Rodgers, 1987). Contact time that influenced retention of the pesticide in the wetland was
largely a function of volume and flow. Processing of the pesticide in the wetland was a function
of biotransformation, photolysis, hydrolysis, and other biological and chemical processes. The
wetland dimensions required to optimize retention and processing could be discerned from the
relationship between the half-life of the pesticide and the wetland physical/chemical character.
This modeling effort yielded a series of hypotheses that could be experimentally tested in
replicated constructed wetlands cells using some "model" or example pesticides.
SELECTION OF PESTICIDES FOR TESTING
Pesticide selection for experimental purposes was based on three factors. First, the use pattern
and volume of pesticide used in Mississippi were selection criteria. It was considered important
to investigate pesticides that are heavily used in large volumes and widely used for a number of
crops. Secondly, the toxicology and fate characteristics of the pesticide were important. Initially,
we wanted to choose a pesticide that would theoretically be very efficiently removed and
processed in a wetland, and then choose a pesticide that we expected would be largely
incompatible with wetlands in order to illustrate the bounds of wetland processing abilities. As
an additional criterion, we wanted to choose a pesticide for which we had analytical and
toxicological experience. We need to have efficacious analytical techniques for the pesticide in
water, sediment, and plant and animal material. If we have a toxicological data base, animals and
plants may be used to evaluate the bioavailability of the pesticide as it is retained in the wetland.
Based on available data (Wauchope, 1978; Reinert and Rodgers, 1987; State of Mississippi, 1990),
a number of candidate pesticides were evaluated (Table 2). In Mississippi, both insecticides and
herbicides are widely used in relatively large volumes on soybeans, cotton, wheat, and rice (State
of Mississippi, 1990). Pyrethroid insecticides are the predominant materials used and are viable
8
-------
candidates for use with wetland buffer strips. Pyrethroids are not particularly water soluble
(solubility in water 10-20 ppb) and readily sorb to plants and soil. They are generally subject to
hydrolysis, photolysis, and biotransformation. Herbicides that are widely used include triazines,
fluridone, and copper-based materials (State of Mississippi, 1990). These herbicides may not be
compatible with wetland buffer strips if they cause direct impacts on the wetland vegetation.
It is apparent that the timing and magnitude of the rainfall event after pesticide application is
crucial in determining pesticide loss in runoff. Measured losses of pesticides in runoff from
agricultural fields ranges from 0.5 to 5 percent, depending on weather and slope of the field. In
terms of ecological impacts on receiving aquatic systems, the mass of pesticide lost from an
agricultural field may not be the primary issue. Of primary concern may be the intermediate
intensity rainfall events that yield maximal concentrations of pesticide in receiving aquatic
systems. Small rainfall events may not yield sufficient mass of pesticide residue when diluted by
receiving systems to be a problem. Large or catastrophic rainfall events (which produce runoff
loses of 2 percent or more of the applied amount of pesticide; Wauchope, 1978) may actually
result in lower concentrations of pesticides in receiving aquatic systems than intermediate events
(Bailey et al., 1974). Thus, impacts of pesticides in runoff from agricultural fields may be most
severe from rainfall events that are large enough to mobilize pesticides but small enough to avoid
excessive dilution resulting in maximum concentrations in the receiving system (Trichell et al.,
1968; Wauchope, 1978; Wallach, 1991). All of the pesticide may be mobilized if the rainfall
occurs soon after application. It is important to recall that pesticide concentration in runoff may
vary more than an order of magnitude during a single event. The most precipitous impacts on
non-target species in receiving aquatic systems are usually related to the maximum exposure
concentration and not the average of a runoff event. The fraction of pesticide mass sorbed to
particulates and the dissolved or aqueous fraction that is transported depends on the character of
the pesticide and the sorbents or the agricultural soil (Rodgers et al., 1987). Still at issue is the
9
-------
bioavailability or bioactivity of particulate-bound and dissolved pesticides in agricultural runoff.
This is an issue that can be, at least partially, resolved by the research strategy.
EXPERIMENTAL DESIGN
The experimental design used for testing model-derived hypotheses incorporates two major
variables. First, the major biological component, the vegetation, of the wetland design was
crucial. Wetland plants have differing growth habits and physiology that affect their ability to
perform in buffer strips to process agricultural pesticide runoff (Wolverton and Harrison, 1975;
Gersberg et al., 1985; Bowmer, 1987; Good and Patrick, 1987; Stengel and Schultz-Hock, 1989).
For example, wetland plants may contribute to or promote formation of anaerobic, low-redox
hydrosoils or they may aerate their root zone forming a largely aerobic, high-redox potential
hydrosoil (MacMannon and Crawford, 1971; Whitlow and Harris, 1979; Mendelssohn et al., 1981;
Faulkner and Richardson, 1989; Guntenspergen et al., 1989).
For this research, three wetland plants were chosen with widely varying physiology and growth
habits. Typha latifolia (cattail) will produce aerobic zones and considerable biomass in the
wetland hydrosoil (Rodgers et al., 1983; Faulkner and Richardson, 1989). Scirpus cyperinus
(bulrush) and Zizania aquatica (wild rice) should permit formation of anaerobic zones in the
wetland hydrosoils (Whigham and Simpson, 1977; Whitlow and Harris, 1979). The influence of a
wetland plant on the hydrosoil character as well as providing sorption surface for pesticide
retention were predicted by the modeling effort to be important factors in pesticide removal
efficiency in the experimental wetland cells.
Secondly, the physical and chemical character of the wetland was also very important for this
research strategy. The dimensions of the wetland are important in retention capacity for runoff
-------
from rainfall events (Figure 3). These experimental wetland cells are constructed with water
recirculation capability so a variety of runoff scenarios can be simulated (Kadlec, 1989; Steiner
and Freeman, 1989). Water depth in each wetland cell can be readily controlled to provide a
suitable habitat for the chosen vegetation. Each wetland is lined below the hydrosoil with 10 cm
of compressed bentonite clay with permeability less than 10"6cm/sec to prevent infiltration.
Ground water monitoring wells are located at the periphery of the constructed wetland
experimental area to ensure that the pesticides do not penetrate the clay liners. The length is
extended on some of the constructed wetland cells to permit longer retention times at a given
flow relative to adjacent, shorter wetland cells (Figure 3). The wetland experimental area is
surrounded by a subsurface drain field and diversion ditch to ensure that no unwanted surface or
ground water enters. A catch basin is located downstream for emergency capacity to accept
water from the experimental wetland cells. The Wetland Research area is located on the 300-ha
Biological Field Station of the University of Mississippi near Oxford (Lafayette County).
TESTABLE HYPOTHESES-DISCUSSION
Using the research strategy presented, a series of experiments are underway that we have
designed to test important hypotheses regarding the ability of wetlands to mitigate the impacts of
pesticides in agricultural runoff. These hypotheses are presented as questions below.
• Does the type of wetland vegetation play an important role in the efficiency of pesticide
retention and processing in wetlands?
• Can the relatively simple transfer and transformation model used to predict pesticide
fate in wetlands be validated or modified to obtain satisfactory estimates?
• Are insecticides processed by wetlands more efficiently than herbicides?
11
-------
• Do constructed, essentially monospecific wetlands perform better than successional or
multi-species wetlands?
• What are the maintenance costs for wetland buffer strips? Can the wetlands function
efficiently for a period of time without considerable maintenance?
• Does sediment or particulate loading influence wetland functions? Are sorbed pesticides
bioavailable or bioactive?
• Do the wetland sediments or hydrosoils regulate the physiology of the vegetation or are
the sediment or hydrosoil characteristics regulated by the extant vegetation?
• Can wetland buffer strips and pesticides be compatibly linked to practical
recommendations that can be incorporated into routine agricultural practices?
• Are wetland functions or values compromised when wetlands are used as buffer strips
for nonpoint source pesticide runoff from agricultural croplands?
These and other ancillary hypotheses will require several years of experimental efforts to resolve.
It is hoped that the experimental strategy will permit an optimal effort to gather information that
will be readily transferable from site to site and of maximum utility.
12
-------
REFERENCES
Bailey, G. W., A. R. Savank, Jr., and H. P. Nicholson, 1974. Predicting pesticide runoff from
agricultural land: a conceptual model. Journal of Environmental Quality, 3: 95-102.
Bavor, H. J., D. J. Roser, P. J. Fisher, and I. C. Smells, 1989. Performance of solid-matrix
wetland systems viewed as fixed film bioreactors. pp. 646-657. In: D. A. Hammer (ed.),
Constructed Wetlands for Wastewater Treatment. Lewis Publishers, Inc., Chelsea, MI.
Bowmer, K. H, 1987. Nutrient removal from effluents by an artificial wetland: influence of
rhizosphere aeration and preferential flow studied using bromide and dye tracers. Water
Research, 21: 591-599.
Cassidy, and J. H. Rodgers, Jr., 1988. Response of Hydrilla (Hydrilla veriicillata (L.f.) Royle) to
Diquat and a model of uptake under nonequilibrium conditions. Environmental Toxicology and
Chemistry, 8: 133-140.
Cooper, C. M., 1989. Status of current technology on constructed wetlands. Submitted to the
DEC Task Force, National Sedimentation Laboratory, USDA-ARS, Oxford, MS.
Faulkner, S. P. and C. J. Richardson, 1989. Physical and chemical characteristics of freshwater
wetland soils, pp. 41-72. In: D. A. Hammer (ed.), Constructed Wetlands for Wastewater
Treatment. Lewis Publishers, Inc., Chelsea, ML
13
-------
Gersberg, R. M., R. Bermer, S. R. Lyon, and B. V. Elkins, 1987. Survival of bacteria and viruses
in municipal wastewater applied to artificial wetlands, pp. 237-245. In: K. R. Reddy and W. H.
Smith (eds.), Aquatic Plants for Water Treatment and Resource Recovery. Magnolia Publishing,
Orlando, FL.
Gersberg, R. M., B. V. Elkins, S. R. Lyons, and C. R. Goldman, 1985. Role of aquatic plants in
wastewater treatment by artificial wetlands. Water Research, 20: 363-367.
Good, B. J. and W. H. Patrick, Jr., 1987. Root-water-sediment interface process, pp. 359-371.
In: K. R. Reddy and W. H. Smith (eds.), Aquatic Plants for Water Treatment and Resource
Recovery. Magnolia Publishing, Orlando, FL.
Greeson, P. E., J. R. Clark, and J. E. Clark (eds.), 1979. Wetland Functions and Values: The
State of our Understanding. American Water Resources Association, Minneapolis, MN.
Guntenspergen, G. R., F. Sterns, and J. A. Kadlec, 1989. Wetland vegetation, pp. 89-103. In:
D. A. Hammer (ed.), Constructed Wetlands for Wastewater Treatment. Lewis Publishers, Inc.,
Chelsea, MI.
Hammer, D. A. and R. A. Bastian, 1989. Wetlands ecosystems: natural water purifiers, pp. 5-
19. In: D. A. Hammer (ed.), Constructed Wetlands for Wastewater Treatment. Lewis Publishers,
Inc., Chelsea, MI.
Horwitz, E. L., 1978. Our Nation's Wetlands—An Interagency Task Force Report. Council on
Environmental Quality, U.S. Government Printing Office, Washington, DC.
14
-------
Kadlec, R. H., 1989. Hydrologic factors in wetland water treatment, pp. 21-40. In: D. A.
Hammer (ed.), Constructed Wetlands for Wastewater Treatment. Lewis Publishers, Inc., Chelsea,
MI.
MacMannon, M. and R. M. M. Crawford, 1971. A metabolic theory of flooding tolerance: the
significance of enzyme distribution and behavior. New Phytologist, 10: 299-306.
Mendelssohn, I. A., K. L. McKee, and W. H. Patrick, Jr., 1981. Oxygen deficiency in Spartina
altemiflora roots: metabolic adaptation to anoxia. Science, 214: 439-441.
Mitsch, W. J. and J. G. Gosselink, 1986. Wetlands. Van Nostrand Reinhold, New York, NY.
Portier, R. J. and S. J. Palmer, 1989. Wetlands microbiology: form, function, process, pp. 21-
40. In: D. A. Hammer (ed.), Constructed Wetlands for Wastewater Treatment. Lewis Publishers,
Inc., Chelsea, MI.
Reinert, K. H., M. L. Hinman, and J. H. Rodgers, Jr., 1988. Fate of Endothall during the Pat
Mayse Lake (Texas) aquatic plant management program. Archives of Environmental
Contamination and Toxicology, 17: 195-199.
Reinert, K. H. and J. H. Rodgers, Jr., 1987. Fate and persistence of aquatic herbicides. Reviews
of Environmental Contamination and Toxicology, 98: 61-98.
Reinert, K. H. and J. H. Rodgers, Jr., 1984. Validation trial of predictive fate models using an
aquatic herbicide (Endothall). Environmental Toxicology and Chemistry, 5: 449-461.
15
-------
Rodgers, J. H., Jr., K. L. Dickson, F. Y. Salem, and C. A. Staples, 1987. Bioavailability of
sediment-bound chemicals to aquatic organisms—some theory, evidence and research needs, pp.
245-266. In: K. L. Dickson, A. W. Maki, and W. A. Brungs (eds.), Fate and Effects of Sediment
Bound Chemicals in Aquatic Systems. Pergamon Press, Elmsford, NY.
Rodgers, J. H., Jr., M. E. McKevitt, D. O. Hammerland, K. L. Dickson, and J. Cairns, Jr., 1983.
Primary production and decomposition of submergent and emergent aquatic plants of two
Appalachian rivers, pp. 298-301. In: T. D. Fontaine III and S. M. Bartell (eds.), Dynamics of
Lotic Ecosystems. Ann Arbor Science Publishers, Ann Arbor, MI.
State of Mississippi, 1990. Pesticide Use. Department of Agriculture and Commerce, Division of
Plant Industry, Mississippi State, MS.
Steiner, G. R. and R. J. Freeman, Jr., 1989. Configuration and substrate design considerations
for constructed wetlands wastewater treatment, pp. 363-377. In: D. A. Hammer (ed.),
Constructed Wetlands for Wastewater Treatment. Lewis Publishers, Inc., Chelsea, MI.
Stengel, E. and R. Schultz-Hock, 1989. Denitrification in artificial wetlands, pp. 484-491. In:
D. A. Hammer (ed.), Constructed Wetlands for Wastewater Treatment. Lewis Publishers, Inc.,
Chelsea, MI.
Trichell, D. W., H. L. Morton, and M. G. Merkle, 1968. Loss of herbicides in runoff water.
Weed Science, 16: 447-449.
Wallach, R., 1991. Runoff contamination by soil chemicals: time scale approach. Water Resources
Research, 27: 215-223.
16
-------
Wauchope, R. D., 1978. The pesticide content of surface water draining from agricultural
fields—a review. Journal of Environmental Quality, 7: 459-472.
Whigham, D. F. and R. L. Simpson, 1977. Growth,, mortality, and biomass partitioning in
freshwater tidal wetland populations of wild rice (Zizania aquatic a var. aqualica). Torry
Botanical Club Bulletin, 104: 347-351.
Whitlow, T. H. and R. W. Harris, 1979. Flood tolerance in plants: a state-of-the-art review.
Technical Report E 79-2. U.S. Army Engineers Waterways Experiment Station, Vicksburg, MS.
Wolverton, B. C. and D. D. Harrison, 1975. Aquatic plants for removal of mevinphos from the
aquatic environment. Journal of the Mississippi Academy of Sciences, 19: 84-88.
17
-------
Table 1. Transfer and removal processes in wetlands that are important in mitigation of nonpoint
source pesticide runoff.
TRANSFER PROCESSES REMOVAL PROCESSES
flow volatilization
sorption photolysis
solubility hydrolysis
retention biotransformation
infiltration
18
-------
Table 2. Pesticides commonly used in Mississippi (State of Mississippi, 1990).
Tralomethrin (Scout)
Permethrin (Ambush/Pounce)
Cypermethrin (Ammo/Cymbush)
Flucythrinate (Payoff)
Fenvalerate (Pydrin)
Carbaryl (Sevin)
Malathion (Malathion 5 EC)
Trifluralin (Treflan)
Dimilin (Scepter)
Metrubuzine (Sencor)
Benomyl (Dupont Benlate)
Mancozeb (Manzate 200)
19
-------
LIST OF FIGURES
Figure 1 - Transformation and transfer half-lives as a function of pesticide residence times
(PRT) in wetlands for various degrees of removal.
Figure 2 - Predictions from the pesticide transformation/transfer model (the fraction remaining)
as a function of the transformation to pesticide residence time ratio.
Figure 3 - Experimental constructed wetland cells. Design permits comparison of wetland plant
species and successional wetlands.
20
-------
Q
LLI
O
LU
LU
UJ
LL
CO
$
Q
$
O
2
UJ
cc
LU
9
o
i—
CO
UJ
OL
a
LU
<
X
•
UJ
X
o
h-
1000
100
10
O 25 % REMOVAL
• 50 % REMOVAL
~ 90 % REMOVAL
¦ 95 % REMOVAL
A 10
1 WEEK 1 MONTH
1000
6 MONTHS 1 YEAR
PESTICIDE RESIDENCE TIME (DAYS)
Figure 1
-------
LU
cc
111
Q
O
h-
CO
LU
CL
0
123456789 10
HALF-LIFE TO PESTICIDE RESIDENCE TIME RATIO O^/PRT)
-------
SUBSURFACE DRAIN
TYPHA
TYPHA
WLD RtCE
WILD RICE
o Monitoring Well
Figure 3
WILD RICE
p-1
I
L-,
TYPHA
TYPHA
DRAIN
-------
ANCILLARY BENEFITS AND POTENTIAL PROBLEMS
WITH THE USE OF WETLANDS
FOR NONPOINT SOURCE POLLUTION CONTROL
Robert L. Knight
CH2M HILL
7201 NW 11th Place
Gainesville, Florida 32602-1647
ABSTRACT
Wetlands utilized for the control of nonpoint source (NPS) pollution provide a variety of
secondary benefits in addition to their primary roles of flood attenuation and water quality
enhancement. Ancillary benefits provided by these wetlands typically include photosynthetic
production, secondary production of fauna, food chain and habitat diversity, export to adjacent
systems, and services to human society such as aesthetics, hunting, recreation, and research.
While most NPS wetlands provide these ancillary benefits, the quantitative magnitude of these
functions may vary greatly from one system to another. Similarly, some functions provided by
NPS control wetlands may be detrimental to the wetland flora and fauna or to society. This
paper provides a brief review of the ecological knowledge available about these functions and
provides guidance on optimizing the appropriate ancillary benefits and avoiding undesirable side
effects while achieving primary NPS control goals.
1
-------
INTRODUCTION
Wetlands perform a variety of functions including physical, chemical, and biological processes
that create economic or aesthetic values to society or life-support for plant and animal
populations (Gosselink and Turner, 1978; Sather and Smith, 1984; Mitsch and Gosselink, 1986;
Erwin, 1990; and Kusler and Kentula, 1990). A partial list of wetland functions includes water
storage/flood attenuation, nutrient assimilation/transformation, sediment storage, photosynthetic
production, secondary production, food chain and habitat diversity, export of carbon and
organisms to adjacent ecosystems, and aesthetic/recreational/educational human uses.
While a specific wetland may perform some or all of these functions, the relative magnitude of
each function exhibited by a wetland, as well as the number of functions exhibited, is highly
variable. The observation that the term "wetland" includes a diverse array of different
ecosystems, with a diversity of abiotic and biotic forcing functions, leads to the deduction that
"all wetlands are not created equal." Published summaries of wetland function display a broad
range of quantitative functional attributes, even for a single wetland plant community type
(Mitsch and Gosselink, 1986). For example, net primary productivity (NPP) varies from as low
as 50 g/m2/y in arctic tundra to 3,500 g/m2/y in southeastern marshes, and litter production
varies from 460 to 2,000 g/m2/y in freshwater marshes (Nixon and Lee, 1986). Total wetland
nitrogen assimilation rate varies from less than 0.08 to over 90 g/m2/yr (Nixon and Lee, 1986;
Knight, 1990). While the functional attributes of wetlands are variable in quality and quantity,
approximate functional levels for individual wetlands can be estimated based on a knowledge of
the structure of the wetland and the array of forcing functions affecting the wetland (Erwin,
1990). In the example above, freshwater marshes display a wide range of NPP rates in North
America, yet a knowledge of a specific marsh's latitude, hydrology, vegetation, and soil type can
greatly increase the accuracy of an estimate of annual NPP.
2
-------
The rapidly expanding understanding of the relationships between a wetland's structure and
function can be used to predict and enhance the functions of wetlands designed for control of
nonpoint source (NPS) pollutants. While wetlands can be used to accomplish the primary
objectives of NPS control (reduction of peak stormwater flows and control of suspended and
dissolved pollutants), ancillary benefits also may be achieed through thoughtful site selection
and design. This paper provides a summary of the ancillary benefits and potential problems
associated with purposely including wetlands in NPS control strategies, and recommends specific
design features that can be used to optimize this technology. Other papers in this volume
(Mitsch; Hammer) and elsewhere (Hammer, 1989) discuss the primary objectives of NPS control
with wetlands. Research concerning wetland evaluation techniques (Golet, 1978; Greeson et al.,
1979; Richardson, 1981; Kusler and Riexinger, 1986) provides a yardstick for comparing the
ancillary benefits of wetlands constructed for NPS pollution control with natural wetland
functions.
ANCILLARY BENEFITS OF NPS CONTROL WETLANDS
The primary objectives of most NPS control projects utilizing wetlands include (1) water
storage/flood attenuation and (2) water quality enhancement through assimilation/ transformation
of sediments, nutrients, and toxic chemicals. The ancillary or secondary benefits that may be
gained from NPS control wetlands include (1) photosynthetic production, (2) secondary
production of fauna, (3) food chain and habitat diversity, (4) export to adjacent ecosystems, and
(5) aesthetic/recreational/educational human uses. The potential for inclusion of each of these
ancillary benefits in NPS control wetlands as well as their potential quantitative functional levels
are described below.
3
-------
PHOTOSYNTHETIC PRODUCTION
One function shared by all wetlands is photosynthetic production by vascular and non-vascular
plants. The magnitude of this primary productivity, however, varies greatly between wetland
types and even within a single wetland habitat type, due to varying environmental forcing
functions at work on individual wetland areas. Wetlands generally have higher NPP than
adjacent uplands (900 to 2,700 g/m2/y in wetlands compared with approximately 500 g/m2/y in
grasslands) due to the subsidies of water and nutrients from these adjacent systems (Richardson,
1978).
Enhancing the primary production of wetlands constructed for NPS control may or may not be
desirable depending on the goals of the project. If a goal is to contribute organic matter as the
basis of a food chain leading to domesticated or wild animal populations, then factors that may
limit primary productivity can be supplied to some extent through system design and operational
control. Forcing functions most frequently limiting primary productivity of constructed wetlands
include light, water, macronutrients, and micronutrients. In emergent marshes, light is the most
likely factor to limit algal production. If a project goal is reduction of algal suspended solids, a
typically included design feature is a densely vegetated emergent zone at the downstream end of
the wetland treatment system. If algal productivity is desired to enhance an aquatic food chain
(for example, fish or shellfish culture), open water and deeper areas should be included in the
design.
In constructing wetlands dependent upon stormwater, water itself may be the most limiting
environmental factor during extended dry periods. High primary production is not realized in
dry wetlands. To maintain a high level of primary productivity in a constructed wetland during
all seasons, an alternate water source should be provided. Highest NPP is generally measured in
shallow (less than 0.3 m), regularly flooded emergent marshes (Brown et al., 1979). This may be
4
-------
due to the availability of water combined with higher sediment dissolved oxygen levels in shallow
or flowing systems (Gosselink and Turner, 1978). More natural, fluctuating water levels will
generally result in lower NPP. Constructed wetlands designed to simulate natural wetland
hydroperiods need to include water-level control structures to prevent damaging flood flows and
to replicate the slow bleed-down of water levels following storms (Livingston, 1989). Prolonged
high water in a constructed wetland will result in a rapid successional change from emergent,
shallow water vegetation to an aquatic system dominated by phytoplankton, filamentous algae, or
floating aquatics (Guntenspergen et al., 1989). If these prolonged flood events occur on several
occasions each year, with dry spells interspersed, a highly stressed and potentially unproductive
plant community will result.
FAUNAL PRODUCTION
Secondary production of herbivorous and/or carnivorous animals in wetlands constructed for NPS
control may be directed to aquaculture of "domesticated" species or to enhancement of wildlife
populations. Generally, greater operational control is required to direct secondary production to
\
"domesticated" species such as crayfish, fish, or other forms of aquaculture. Wengrzynek and
Terrell (1990) have developed a generic constructed wetland design for NPS control that may
incorporate baitfish or freshwater mussel production as an ancillary benefit. Wetlands dominated
by grasses also may be highly productive areas for livestock grazing.
To many conservationists, the most exciting ancillary benefit of the construction of wetlands for
NPS control is enhancement of wildlife populations. Wetland and impoundment design for
wildlife enhancement is relatively well known (Weller, 1978; Smith et al„ 1989; Weller, 1990). A
diversity of plant species and growth forms with a variety of plant and seed maturity dates will
provide a wider range of wildlife niches within the wetland. Total wildlife production also may
be most closely correlated with water quantity and quality. Wildlife are attracted to wetlands that
5
-------
have perennial water, and areas that are flooded less frequently generally will have smaller
populations of wetland-dependent wildlife species. However, fluctuating water levels create
additional niches and result in higher wildlife diversity. Wetlands created for water quality
enhancement also may provide expanded habitat for threatened or endangered wildlife species.
Water quality is important to wildlife production through its control on primary production.
Constructed wetlands receiving waters with higher nutrient content generally have larger wildlife
populations. Increased NPP is transferred primarily through the detritus food chain to
invertebrates and small fish, reptiles, amphibians, and birds. As long as these intermediate
consumer populations are not restricted from colonizing and increasing their numbers within a
wetland with high plant productivity, they will develop a food base for the more highly visible
avifauna typical of constructed wetlands.
If inflow waters have lower nutrient levels, typical of nonpoint sources from less developed
watersheds, then food chain support will be less than the maximum possible level. In a lower
productivity NPS pollution constructed wetland, wildlife species diversity may be higher than in
a wetland with higher primary productivity. Typically, a wetland's value for wildlife increases
with the association of neighboring undeveloped upland habitat.
The physical design features of a wetland may have a greater influence than nutrient levels on
faunal diversity and abundance. For example, waterfowl populations are enhanced by the
provision of open water areas interspersed with deep emergent marsh and upland islands. An
approximate ratio of wetland area devoted to marsh and open water that will provide maximum
habitat for a variety of waterfowl is about 1:1 (Weller, 1978).
6
-------
Wading birds require a different habitat mix than waterfowl. These species require shallow,
sparsely vegetated littoral areas or perching substrates adjacent to open water areas. NPS
constructed wetlands can be designed to provide a broad shelf of emergent marsh with water
depths of less than 20 to 30 cm to benefit populations of wading birds. Deep, open water areas
adjacent to a shallow marsh provide additional foraging habitat for wading birds. Open water
areas and the transitional ecotones between marsh, open water areas, and adjacent uplands help
promote higher populations of herptiles and fish which, in turn, provide a forage base for wading
and diving birds. Inclusion of a diverse fish population during system startup or through natural
immigration will lengthen the consumer food chain and provide potential support for raptors such
as ospreys, hawks, eagles, and kites.
Many other varieties of birds will colonize constructed wetlands, depending on regional
occurrence and available habitats. If living or dead trees are included in the wetland, they will
serve as perching and possible nesting sites for numerous bird species (Hair et al., 1978). Nesting
boxes may be provided to encourage use of the site by wood ducks and owls. Upland islands
surrounded by open water provide protection for ground-nesting bird species.
A variety of small or large mammals may utilize constructed wetlands, depending on available
food and habitat resources. Small mammals will develop large populations on upland areas
adjacent to and within the constructed wetlands. These populations provide a forage base for
raptors and large wading birds. Larger mammals also may be included, which adds to overall
system diversity and creates the possibility of a byproduct of animal skins. Nutria, musk rats, and
beaver generally will not colonize a constructed wetland unless perennial water is available.
Large mammals may provide important feedback control of NPP in wetlands due to their ability
to rapidly reduce biomass for lodges and food and to maintain patches of high net productivity,
early successional marsh (Weller, 1978; Hair et al., 1978).
7
-------
HUMAN USES
A major functional value of wetlands is their importance for consumptive (plant harvesting,
livestock grazing, hunting, aquaculture, etc.) and nonconsumptive (aesthetics, recreation, and
research) human uses (Nash, 1978; Reimold and Hardisky, 1978; Sather and Smith, 1984; and
Smardon, 1988).
Consumptive uses such as waterfowl hunting and fur trapping are more easily quantified than
nonconsumptive functions (Chabreck, 1979). Nonconsumptive human uses of wetlands
constructed primarily for water quality treatment include recreation, nature study, aesthetics, and
education. An increasing number of treatment wetlands are being designed as attractive and
informative park-like areas. Due to their urban setting, storm water treatment wetlands such as
Greenwood Park in Orlando, Florida, and Coyote Hills located east of San Francisco Bay,
California, are heavily visited and utilized for field trips and other educational purposes. Two
wetlands constructed for wastewater treatment (Areata, California, and Ironbridge, Florida) are
important recreational areas offering jogging and bird watching opportunities as major ancillary
benefits. The human uses listed above (including the desirable benefit of just knowing that the
wetland and its wildness are still there at the edge of town) are perhaps the most important
factors in the popular support for protecting and enhancing the existing wetland resource base.
DESIGN FOR ANCILLARY BENEFITS
General wetland design features affecting the secondary functions of NPS wetlands were
presented above. This section focuses on specific design considerations to provide those wetland
features and their resulting functions, while simultaneously optimizing the primary functions of
flood attenuation and pollutant removal.
8
-------
Wetland design includes decisions concerning siting, cell size and configuration, water flow and
depth control, planting, and stocking with wildlife species. Because the primary function of the
wetland a priori is NPS control, the water source is not a design decision; however, the level of
pretreatment is an important item for consideration. Each of these topics is discussed briefly
below.
WETLAND SITING
NPS control wetlands can be sited close to individual stormwater sources or further downstream
in a watershed, intercepting a tributary. One effect of wetland siting on the resulting wetland
functions is the quantity and timing of water in the system. Wetlands sited in headwater areas
generally will receive more irregular and less dependable inflows, potentially resulting in
prolonged dry conditions (unless soils are very impermeable or groundwater levels are normally
high). This relative lack of flooding will reduce the quantitative magnitude of ancillary wetland
values such as primary and secondary production. Maintenance of a healthy stand of wetland-
dependent vegetation may be difficult and upland or transitional species may eventually
predominate. This type of system certainly will have some wetland values and may support
differing faunal assemblages seasonally; however, the overall production of wetland-dependent
species will likely be lower than in a perennially flooded wetland.
Siting of the constructed wetland further downstream in the watershed may result in a different
constraint, namely too much water during stormwater periods. Design of a downstream system
may be "offline," allowing capture of only a portion of flood flows to prevent the washout of
vegetation and berms. A series of offline constructed wetlands, each capturing a portion of the
storm flows, can be used to deal with high storm volume. Wetlands located downstream in a
watershed have a higher potential to have perennial water and higher ancillary food-chain
benefits due to more constant base flows and generally higher groundwater levels.
9
-------
Wetland siting also may be very important because of other concerns such as the benefits of
having adjacent donor wetlands for plant seeds and wildlife; adjacent undeveloped uplands to
provide greater habitat diversity; or the importance of human contact and aesthetics. These siting
issues are dependent upon project-specific goals.
CELL SIZE AND CONFIGURATION
Wetland cell size is based primarily on water quality treatment and cost considerations. Larger
cells require less berm construction per unit area and fewer inlet/outlet structures, resulting in
reduced project costs per area. For example, larger constructed wetlands (greater than 100 ha)
may cost about $10,000/ha to construct, while smaller constructed wetlands may cost about
$50,000/ha. Cell size may affect usage by some larger wildlife species but it has minimal effects
on plant productivity or secondary production of most wetland animals (Sather and Smith, 1984).
A greater berm to cell area ratio is typical of smaller wetland cells and may result in increased
edge effect and increased nesting and feeding habitat for many mammal and bird species, as long
as berms are infrequently mowed or visited.
Islands surrounded by marsh or open water provide excellent habitat for nesting waterfowl.
Islands with trees are the preferred nesting habitat for wading bird rookeries in many wetlands.
Nesting islands for waterfowl should be only about 0.6 m above normal high water, while higher
and lower islands also may be valuable for other species for feeding, resting, or nesting.
Inclusion of open water areas not only improves the water quality treatment potential of
constructed wetlands (Knight and Iverson, 1990), but also greatly enhances their ancillary
benefits for wildlife. Mallard duck production is maximum in wetlands with approximately equal
areas of marsh and open water (Ball and Nudds, 1989). Open water areas can be created by
excavating a minimum of about 1.5 m below normal water level, and deeper excavations can
10
-------
provide greater hydraulic residence times and fisheries habitat. To prevent hydraulic short-
circuiting, open water areas should not be connected along the flow path, but rather interspersed
with densely vegetated shallow marsh habitat (about 0.3 m average depth or less).
Cell number and configuration in series or in parallel is a major consideration for treatment
capability and operational flexibility. These design considerations affect ancillary wetland
benefits primarily through their importance for water flow and depth control.
WATER FLOW AND DEPTH CONTROL
Water depth and flow rate are important factors affecting dissolved oxygen in wetlands. Higher
flow rates resulting from shallow water conditions tend to provide higher dissolved oxygen
concentrations in marsh areas due to the increased influence of atmospheric reaeration. Higher
dissolved oxygen levels generally result in higher secondary production of aquatic invertebrates
and vertebrates, increasing these ancillary wetland benefits. While deeper water in a marsh area
may increase hydraulic residence time, this longer reaction time in many cases does not result in
enhanced water quality treatment (oxidation of organic matter and ammonia) because of the
resulting reduction of dissolved oxygen.
Water depth is one of the main determinants affecting wetland plant growth. High water levels
will stress growth of emergent macrophytes and encourage dominance by floating or submerged
plants or algae. The hydrological tolerance range and optimum hydroperiod should be known for
any desired vegetation type and closely adhered to in design of water-level control structures.
Ideal design for water-level control allows water levels to be varied from zero (drained) to the
maximum depth tolerance of desired wetland plant communities. Stop logs or weir plates should
be of a type that effectively seal against leaks to help maintain water levels during periods of
11
-------
limited inflows. Multiple inlet and outlet weirs between adjacent cells allow greatest hydroperiod
control flexibility.
VEGETATION PLANTING
The selection of appropriate plant species for inclusion in a wetland constructed for NPS
control will greatly influence ancillary benefits such as primary and secondary productivity.
Improper plant species selection will result in low productivity, and a lengthy adaptive period
may be necessary until available plant species, either planted or occurring naturally, rearrange
themselves according to hydrologic factors. High plant diversity frequently can be achieved by
natural colonization from existing soil seed banks in an area graded and shallowly flooded, or by
using the technique of spreading muck and associated propagules from a donor wetland area
(Gilbert et al., 1981). Wetland vegetation establishment is most rapid with closely spaced plants
(less than 1 m on centers), planted during the growing season (Lewis and Bunce, 1980; and
Broome, 1990). No marsh species is totally unutilized by wildlife, either directly for food or
shelter or indirectly through the detritus food chain; therefore, expensive management to exclude
"noxious species" or to select for favored species may lower overall wildlife utilization in favor of
optimizing specific wildlife species. Burning marshes may be good management for waterfowl
species but may be poor management for other bird species, fish, or small mammals.
WILDLIFE STOCKING
Stocking constructed wetlands with mosquito fish (Cambusia affinis) has repeatedly been found
to provide effective mosquito control as long as deeper water refuge areas, periodically free of
floating vegetation, are available to provide perennial habitat. Mosquito fish, in turn, are an
important forage fish for wildlife. Other forage fish that can be easily stocked (such as shiners,
minnows, shad, and sunfish) contribute to a potentially long food chain of sport fish, reptiles,
wading birds, waterfowl, and raptors.
12
-------
Other food chain components may also be stocked or allowed to naturally immigrate to the
constructed wetland. Significant fur bearer populations (otter, mink, muskrat, or nutria) can be
supported in highly productive constructed wetlands. As with many wetland-dependent birds, a
mixture of open water and marsh habitat is also essential for enhancement of these mammal
species. Stocking of constructed wetlands located some distance from existing wetland habitat
may be very important in quickly establishing ancillary benefits for wildlife and aesthetic uses.
INFLOW PRETREATMENT
In conjunction with wetland hydroperiod, water quality is one of the key determinants of a
wetland's form and function. Primary water quality characteristics affecting wetland plant
communities are nutrients (especially nitrogen and phosphorus), suspended sediments, salts, pH,
and temperature. Other than nutrient concentrations, these same water quality characteristics also
greatly influence faunal populations. Inflow concentrations of biodegradable solids and the
ammonia form of nitrogen may have indirect effects on wetland flora and fauna through their
control of dissolved oxygen concentration in wetlands.
Within the normal range of fresh surface waters that might be captured by wetlands constructed
for NPS control, only suspended solids, nutrients, and salts in a few areas are likely to occur at
levels that might be detrimental to specific ancillary wetland goals. High suspended solids loads,
if released within the constructed wetland, may smother plant growth in inflow areas (Kuenzler,
1990). This potential problem generally can be controlled by the use of a pretreatment grassed
swale, a high-maintenance pretreatment wetland cell, or a pond prior to the habitat wetland
(Livingston, 1989). If mineral suspended solids (clays, silt, and sand) are trapped in a
pretreatment area, disruptive maintenance of the NPS wetland may be much less frequent.
13
-------
Nutrient levels also will be reduced somewhat by passage though swales and ponds. Nutrient
reduction may not be desired for a constructed wetland designed for the ancillary benefit of
wildlife enhancement. As noted earlier, higher nutrients generally result in higher primary
productivity of wetland plants and higher resulting wildlife utilization and production.
HUMAN ACCESS
Inclusion of boardwalks and blinds may greatly enhance the ancillary benefits of recreation and
scientific research in a constructed wetland. While public access to a created wetland may disturb
wildlife populations, these functions can be compatible if control of access to certain areas is
maintained and islands for roosting and nesting are provided.
DESIGN TO MINIMIZE POTENTIAL PROBLEMS
Wetlands utilized for NPS control may create nuisance conditions that potentially negate their
positive ancillary benefits. These nuisance conditions fall into two broad categories:
(1) conditions that are a nuisance or hazardous to humans and (2) conditions that are hazardous to
plants and wildlife. Each of these types of potential problems and their potential corrective
measures are described below.
NUISANCES TO SOCIETY
Historically, wetlands were considered to be nuisance areas, harboring disease, poisonous reptiles,
and noxious conditions. Wetlands drainage certainly saved many thousands of lives prior to the
age of miracle drugs for prevention of malaria and yellow fever. The irradication of roost insect-
transmitted diseases as well as effective biting-insect control through biological and chemical
agents has softened society's adaptive wetland loathing. However, periodic outbreaks of
encephalitis continue to result in warnings from public health officials concerning creation of
14
-------
wetlands for water quality treatment (Wellings, 1986). Mosquito control districts continue to get
dozens of calls following rainy spells that result in the synchronized emergence of biting adult
mosquitoes. Golfers and homeowners continue to insist on mowed margins along stormwater
ponds in the southern United States due to concerns about the occurrence of poisonous snakes.
Mosquito control using mosquito fish is relatively easy in constructed wetlands as long as
perennial water areas exist and strongly anoxic conditions are avoided (Martin and Eldridge,
1989). Many wetlands receiving only NPS loadings may periodically go dry, resulting in total loss
of mosquito fish populations. Without natural or intentional restocking with mosquito fish, these
constructed wetlands are likely to result in significant nuisance conditions if located near
populated areas.
Wetland-dependent venomous snakes such as the water moccasin (Ankistrodon pisivorous) and
alligators (Alligator mississippiensis) are attracted to created wetlands in the southeastern United
States that have high vertebrate and fish productivity. Warning signs, boardwalks, and mowed
hiking trails are generally adequate to prevent the potential loss of recreational values due to
poisonous snakes and other dangerous reptiles.
Because wetlands are infrequently used for water contact recreation, direct disease transmission
by water-borne pathogens is unlikely (Shiaris, 1985). Wetlands used for NPS control are not
generally utilized for human water supply. Due to potential toxic metals and pesticides
sometimes found in NPS waters, potable use must be carefully monitored. One potential toxin
pathway to humans is consumption of contaminated fish or wildlife from a treatment wetland.
As discussed more fully in the next section, toxic metals and organic compounds must be
prevented from accumulating to toxic concentrations in NPS wetlands to protect wildlife and
humans. Forethought in wetland design and water pretreatment, as well as periodic monitoring
15
-------
of wetland water quality, is necessary to prevent any potential detrimental food chain effects in
humans who consume wetland plants or animals.
ENVIRONMENTAL HAZARDS
Environmental hazards that may occur when wetlands are used for NPS control include effects
due to high loadings of pollutants that are normally subsidies (for example, too much water,
organic matter, or nutrients) and effects resulting from metals, pesticides, and other potentially
toxic chemicals.
Typical environmental changes resulting from too much water, biodegradable organic matter, or
hypereutrophication due to high nutrient inputs are generally attributable to the direct or indirect
action of all of these compounds to lower dissolved oxygen concentrations. Drastically lowered
dissolved oxygen can result in significant losses of wetlands vegetation and fauna. Limiting
hydrologic and oxygen-demanding loadings to wetlands is relatively easy to accomplish. While
fairly exotic wildlife impacts such as avian cholera and botulism (Friend, 1985) and parasitic
nematodes (Spalding, 1990) are possible due to depressed oxygen concentrations, these have not
been shown to be a widespread threat to the use of wetlands for NPS control.
Other pollutants removed by NPS treatment wetlands are conservative from the standpoint of
accumulation and storage in the wetland sediments, plants, and wildlife. Metals and
organochlorine compounds are among the most likely conservative pollutants that invariably
accumulate in wetlands receiving stormwater or agricultural drainage waters. While both direct
and indirect environmental effects of toxins are possible in NPS wetlands, there is little evidence
that these potential issues represent a real limitation on the use of wetlands for flood control and
water quality management (Chan et al., 1982). However, where inflow concentrations of toxins
16
-------
are a concern, or in specific environments where indirect toxic or lethal condition, may develop,
wetland planning and design must seek to minimize wildlife impacts.
Heavy metals generally follow one of two behaviors in wetlands (Gardner, 1980; Rudd 1987).
Metals such as arsenic, cadmium, chromium, nickel, and zinc are quickly concentrated in soils
and plants compared with water concentrations, primarily through direct adsorption and
absorption. Rooted plants also acquire some metals via uptake from soils. Bioconcentration
factors between the water and plant tissues are in the range from 100 to 1,000 times. In spite of
these increased concentrations in plants, concentrations of these metals are not magnified through
the food chain; dry weight concentrations decrease at higher food chain levels so that
bioconcentration factors between water concentrations and fish tissues are less than 100 times
(USEPA, 1986). These metals essentially reach saturation levels in tissue based on water
concentrations, and additional uptake is matched by tissue metal losses, resulting in a relatively
constant body burden. As long as source control or pretreatment prevents consistently high
concentrations of these metals in the wetland influent, levels toxic to biota are unlikely to occur.
Microbially methylated forms of mercury and lead bioaccumulate in plants and also become
concentrated through food-chain biomagnification. This concentration occurs because these
metal-organic complexes have an affinity for lipids and are accumulated in tissues during the
lifetime of an organism. Organochlorines such as DDT and dioxins biomagnify in the wetland
food chain because of the same affinity for fats.
As with other metals, excretion and release mechanisms also exist for methylated mercury, lead,
and organochlorines. Steady-state levels can be reached that do not result in toxic effects as long
as input concentrations to the wetland are low. Safe input concentrations are not clearly known;
therefore, compliance with published water quality standards is perhaps the best recommendation
17
-------
for these compounds. The existing water quality criteria for metals are intended to protect the
most sensitive organisms within the waters of the United States. In wetlands, these
concentrations are protective of invertebrates or fish that may reside in the vicinity of the inflow.
Since metal concentrations in NPS waters are frequently above these protective criteria, a tradeoff
is necessary if wetlands are to be used for this purpose and the potential ancillary benefits are to
be realized. Perhaps the most difficult issue to address is whether the creation of habitat for
hundreds or thousands of ducks and wading birds is ample justification to exceed wetland surface
water metal concentrations that are potentially chronically toxic to invertebrates or larval fish but
that will not result in chronically toxic conditions for adult fish or birds. Development of
biological criteria for wetlands to replace existing water quality criteria developed for streams
and lakes may provide an answer to this regulatory dilemma (USEPA, 1990),
To date, there are no generally known incidences of conditions in treatment wetlands (municipal
wastewater and stormwater) that have resulted in lethality to fish or other wildlife. The only
documented cases of toxicity to wetlands wildlife known to this author are releases from
hazardous waste sites (for example, USEPA, 1989), and discharges of agricultural irrigation
return flows in the western United States (Willard and Willis, 1988; Deason, 1989). Research with
agricultural drainage water at the Kesterson National Wildlife Refuge has recorded effects
including vegetation changes, losses of species, fish die-offs, and acute and chronic effects on
birds, primarily from highly concentrated levels of selenium. While sites such as Kesterson,
Imperial Valley, Stillwater, and other wildlife refuges dependent on agricultural waters were not
designed for treatment but rather for habitat, they serve as poignant examples of what must be
avoided in the design of new wetland treatment systems.
18
-------
SUMMARY AND RECOMMENDATIONS
The potential ancillary benefits derived from the wise use of wetlands for NPS control are great.
While protecting the Nation's surface water resources, NPS treatment wetlands can provide a
significant increase in the Nation's wetland resource base, provide a high level of additional
food-chain support for wildlife and humans, maintain genetic diversity, provide export to
neighboring ecosystems, and enhance society through provision of aesthetics, recreation, and
education. Maximizing these potential benefits through the planning and design of new wetland
treatment systems, or through modified operations and management of existing systems, is a
worthy goal. Minimizing the potential negative effects of this emerging technology is an equally
important requirement.
The state of our knowledge concerning wetland functions and what environmental factors control
these functions is rapidly advancing due to wetlands research efforts worldwide. The practical
application of this diffuse knowledge base is difficult because of the lack of an organized
synthesis concerning wetlands management. The sole recommendation of this author is that the
existing knowledge concerning wetlands management for ancillary benefits be organized into
practical guidelines, to be periodically updated, for the planning, design, and operation of
wetland treatment systems.
19
-------
REFERENCES
Ball, J.P. and T.D. Nudds, 1989. Mallard habitat selection: An experiment and implications for
management, pp. 659-671. In: R.R. Sharitz and J.W. Gibbons (eds.), Freshwater Wetlands and
Wildlife, Proceedings of a Symposium Held March 24-27, 1986, Charleston, S.C. US Department
of Energy, Oak Ridge, TN.
Broome, S.W., 1990. Creation and restoration of tidal wetlands of the southeastern United States,
pp. 37-72. In: J.A. Kusler, and M.E. Kentula, Wetland Creation and Restoration. The Status of
the Science. Island Press, Washington, DC.
Brown, S., M.M. Brinson, and A.E. Lugo, 1979. Strategies for protection and management of
floodplain wetlands and other riparian ecosystems, pp. 17-31. In: R.R. Johnson and J.F.
McCormick (eds.), Proceedings of the Symposium, Structure and Function of Riparian Wetlands,
December 11-13, 1978, Callaway Gardens, GA. USDA, Washington, DC.
Chabreck, R.H, 1979. Wildlife harvest in wetlands of the United States, pp. 618-631. In: P.E.
Greeson, J.R. Clark, J.E. Clark (eds.), Wetland Functions and Values: The State of Our
Understanding, Proceedings of the National Symposium on Wetlands, Lake Buena Vista, FL,
November 7-10, 1978. American Water Resources Association, Bethesda, MD.
Chan, E., T.A. Bursztynsky, N. Hantzsche, and Y.J. Litwin, 1982. The Use of Wetlands for
Water Pollution Control. EPA-600/600/2/12-82-086. 261 pp.
20
-------
Deason, J.P., 1989. Impacts of irrigation drainwater on wetlands, pp. 127-138. In: D.W. Fisk
(ed.), Proceedings of the Symposium on Wetlands: Concerns and Successes. September 17-22,
1989, Tampa, FL, American Water Resources Association, Bethesda, MD.
Erwin, K.L., 1990. Wetland evaluation for restoration and creation, pp. 429-458. In: J.A.
Kusler and M.E. Kentula, Wetland Creation and Restoration: The Status of the Science. Island
Press, Washington, DC.
Friend, M., 1985. Wildlife health implications of sewage disposal in wetlands, pp. 262-269. In:
P.J. Godfrey, E.R. Kaynor, S. Pelczanski and J. Benforado (eds.), Ecological Considerations in
Wetlands Treatment of Municipal Wastewaters. Van Nostrand Reinhold Co., New York, NY.
Gardner, W.S., 1980. Salt marsh creation: Impact of heavy metals, pp. 126-131. In: J.C.Lewis
and E.W. Bunce (eds.), Rehabilitation and Creation of Selected Coastal Habitats: Proceedings of a
Workshop. U.S. Fish and Wildlife Service, FWS/0BS-80/27.
Gilbert, T., T. King, and B. Barnett, 1981. An Assessment of Wetland Habitat Establishment at a
Central Florida Phosphate Mine Site. U.S. Fish and Wildlife Service, FWS/OBS-81/38. 95 pp.
Golet, F.C., 1979. Rating the wildlife value of northeastern freshwater wetlands, pp. 63-73. In:
P.E. Greeson, J.R. Clark, and J.E. Clark (eds.), Wetland Functions and Values: The State of Our
Understanding, Proceedings of the National Symposium on Wetlands, Lake Buena Vista, FL,
November 7-10, 1978. American Water Resources Association, Bethesda, MD.
21
-------
Gosselink, J.G. and R.E. Turner, 1978. The role of hydrology in freshwater wetland ecosystems,
pp. 63-78. In: R.E. Good, D.F. Whigham, and R.L. Simpson (eds.). Freshwater Wetlands,
Ecological Processes and Management Potential. Academic Press, New York, NY.
Greeson, P.E., J.R. Clark, and J.E. Clark (eds.), 1979. Wetland Functions and Values: The State
of Our Understanding, Proceedings of the National Symposium on Wetlands, Lake Buena Vista,
FL, November 7-10, 1978. American Water Resources Association, Bethesda, MD, 674 pp.
Guntenspergen, G.R., F. Stearns, and J.A. Kadlec, 1989. Wetland Vegetation, pp. 73-88. In:
D.A. Hammer (ed.), Constructed Wetlands for Wastewater Treatment: Municipal, Industrial, and
Agricultural. Lewis Publishers, Chelsea, MI.
Hair, J.D., G.T. Hepp, L.M. Luckett, K.P. Reese, and D.K. Woodward, 1978. Beaver pond
ecosystems and their relationship to multi-use natural resource management, pp. 80-92. In:
R.R. Johnson and J.F. McCormick (eds.), Strategies for Protection and Management of Floodplain
Wetlands and Other Riparian Ecosystems, Proceedings of the Symposium, December 11-13, 1978,
Callaway Gardens, GA. USDA, Washington, DC.
Hammer, D.A. (ed.), 1989. Constructed Wetlands for Wastewater Treatment: Municipal,
Industrial, and Agricultural. Lewis Publishers, Chelsea, MI.
Knight, R.L., 1990. Wetland systems. In: S. Reed (ed.), Natural Systems for Wastewater
Treatment, Water Pollution Control Federation, Alexandria, VA, MOPFD-16, 270 pp.
22
-------
Knight, R.L. and M.E. Iverson, 1990. Design of the Fort Deposit, Alabama, constructed wetlands
treatment system, pp. 521-524. In: P.F. Cooper and B.C. Findlater (eds.), Constructed Wetlands
in Water Pollution Control. IAWPRC, Pergamon Press, Oxford, UK.
Kuenzler, E.J., 1990. Wetlands as sediment and nutrient traps for lakes, pp. 105-112. In:
Proceedings of a National Conference on Enhancing the States* Lake and Wetland Management
Programs, May 18-19, 1989, Chicago, IL.
Kusler, J.A. and M.E. Kentula, 1990. Wetland Creation and Restoration: The Status of the
Science, Island Press, Washington, DC. 591 pp.
Kusler, J.A. and P. Riexinger (eds.), 1986. Proceedings of the National Wetland Assessment
Symposium, Portland, ME, June 17-20, 1985. Association of State Wetland Managers, Chester,
VT. 331 pp.
Lewis, J.C. and E.W. Bunce (eds.), 1980. Rehabilitation and Creation of Selected Coastal
Habitats: Proceedings of a Workshop, U.S. Fish and Wildlife Service, FWS/OBS-80/27.
Livingston, E.H., 1989. Use of wetlands for urban stormwater management, pp. 253-262. In:
D.A. Hammer (ed.), Constructed Wetlands for Wastewater Treatment Municipal, Industrial, and
Agricultural. Lewis Publishers, Chelsea, MI.
Martin, C.V. and B.F. Eldridge, 1989. California's experience with mosquitoes in aquatic
wastewater treatment systems, pp. 393-398. In: D.A. Hammer (ed.), Constructed Wetlands for
Wastewater Treatment: Municipal, Industrial, and Agricultural. Lewis Publishers, Inc., Chelsea,
MI.
23
-------
Mitsch, W.J. and J.G. Gosselink, 1986. Wetlands. Van Nostrand Reinhold Co., New York, NY.
539 pp.
Nash, R., 1978. Who loves a swamp? pp. 149-156. In: R.R. Johnson and J.F. McCormick (eds.)
Strategies for Protection and Management of Floodplain Wetlands and Other Riparian
Ecosystems, Proceedings of the Symposium, December 11-13, 1978, Callaway Gardens, GA.
USDA, Washington, DC.
Nixon, S.W. and V. Lee, 1986. Wetlands and Water Quality. A Regional Review of Recent
Research in the U.S. on the Role of Freshwater and Saltwater Wetlands as Sources, Sinks, and
Transformers of Nitrogen, Phosphorus, and Various Heavy Metals, U.S. Army Corps of
Engineers, Wetlands Research Program, Technical Report Y-86-2.
Reimold, R.J. and M.A. Hardisky, 1979. Nonconsumptive use values of wetlands. In: P.E.
Greeson, J.R. Clark, and J.E. Clark (eds.), Wetland Functions and Values: The State of Our
Understanding, Proceedings of the National Symposium on Wetlands, Lake Buena Vista, FL,
November 7-10, 1978. American Water Resources Association, Bethesda, MD.
Richardson, B. (ed.), 1981. Selected Proceedings of the Midwest Conference on Wetland Values
and Management, St. Paul, MN, June 17-19, 1981, Freshwater Society, Navarre, MN. 660 pp.
Richardson, C.J., 1979. Primary productivity values in freshwater wetlands, pp. 131-145. In:
P.E. Greeson, J.R. Clark, and J.E. Clark (eds.). Wetland Functions and Values: The State of Our
Understanding, Proceedings of the National Symposium on Wetlands, Lake Buena Vista, FL,
November 7-10, 1978. American Water Resources Association, Bethesda, MD.
24
-------
Rudd, T., 1987. Scope of the problem, pp. 1-29. In: J.N. Lester (ed.), Heavy Metals in
Wastewater and Sludge " eatment Processes. Volume 1. Sources, Analysis, and Legislation.
CRC Press, Boca Rate,., FL.
Sather, J.H. and R.D. Smith, 1984. An Overview of Major Wetland Functions and Values. U.S.
Fish and Wildlife Service, FWS/OBS-84/18.
Shiaris, M.P., 1985. Public health implications of sewage applications on wetlands:
microbiological aspects, pp. 243-261. In: P.J. Godfrey, E.R. Kaynor, S. Pelczanski, and J.
Benforado (eds.), Ecological Considerations in Wetlands Treatment of Municipal Wastewaters.
Van Nostrand Reinhold Co., New York, NY.
Smardon, R.C., 1988. Aesthetic, recreational, landscape values of urban wetlands, pp. 92-95. In:
J.A. Kusler, S. Daly, and G. Brooks (eds.), Proceedings of the National Wetland Symposium on
Urban Wetlands, June 26-29, 1988, Oakland, CA. Association of Wetland Managers, Berne, NY.
Smith, L.M., R.L. Pederson, and R.M. Kaminski, 1989. Habitat Management for Migrating and
Wintering Waterfowl in North America. Texas Tech University Press, Lubbock, TX. 560 pp.
Spalding, M.G., 1990. Antemortem diagnosis of Eustrongylidosis in wading birds
(Ciconiiformes). Colonial Waterbirds, 13: 75-77.
U.S. Environmental Protection Agency, 1986. Superfund Public Health Evaluation Manual. EPA
540/1-86/060. 145 pp.
25
-------
U.S. Environmental Protection Agency, 1989. Water Quality and Toxic Assessment Study,
Mangrove Preserve, Munisport Landfill Site, North Miami, Florida. Environmental Services
Division, Athens, GA.
U.S. Environmental Protection Agency, 1990. Biological Criteria. National Program Guidance
for Surface Waters. Criteria and Standards Division. EPA-440/5-90-004. 57 pp.
Weller, M.W., 1978. Management of freshwater marshes for wildlife, pp. 267-284. In: R.E.
Good, D.F. Whigham, and R.L. Simpson (eds.), Freshwater Wetlands, Ecological Processes and
Management Potential. Academic Press, New York, NY.
Weller, M.W., 1990. Waterfowl management techniques for wetland enhancement, restoration and
creation useful in mitigation procedures, pp. 517-528. In: J.A. Kusler, and M.E. Kentula (eds.),
Wetland Creation and Restoration: The Status of the Science. Island Press, Washington, DC.
Wellings, F.M., 1986. Letter to the Editor. Florida Water Resources Journal, 38: 38-39.
Wengrzynek, R.J. and C.R. Terrell, 1990. Using constructed wetlands to control agricultural
nonpoint source pollution. In: Proceedings of the International Conference on the Use of
Constructed Wetlands in Water Pollution Control, September 24-28, 1990, Churchill College,
Cambridge, UK..
Willard, D.E., and J.A. Willis, 1988. Lessons from Kesterson. pp. 116-121. In: J.A. Kusler, S.
Daly, and G. Brooks (eds.). Proceedings of the National Wetland Symposium on Urban Wetlands,
June 26-29, 1988, Oakland, CA. Association of Wetland Managers, Berne, NY.
26
-------
REGULATIONS AND POLICIES RELATING TO THE USE OF
WETLANDS FOR NONPOINT SOURCE POLLUTION CONTROL
Sherri Fields
Wetlands Division
U.S. Environmental Protection Agency
401 M Street, SW (A-104F)
Wash gton, D.C. 20460
ABSTRACT
The ability of wetlands to transform or trap nutrients and sediments has led to increasinn
attention to how wetlands can be used or their functions replicated to treat nonpoint source, nf
pollution. Protection, restoration, and creation of wetlands can be incorporated into nonooint
source pollution management strategies. There are, however. Federal regulations that prohibit the
indiscriminate use of wetlands for water treatment. The Clean Water Act regulates all diuhnr.!.
into "waters of the United States" including wetlands. Restored wetlands are subject to the sam*
protection and restrictions as natural wetlands. Created wetlands, on the other hand are
generally not considered "waters of the United States" if constructed solely for purposes of water
treatment. Protection, restoration, and creation of wetlands provide opportunities to realize a
number of functional benefits including water quality improvement.
1
-------
INTRODUCTION
The many functions and values of wetlands are widely recognized, particularly with regard to
their abilities to improve water quality. As a result, there has been increasing attention to how
these natural systems can be used or their functions replicated to help treat nonpoint source (NPS)
pollution. In considering this relatively new application of wetland functions, it is important to
be aware of the current policies and regulations that affect the use of wetlands and wetland
treatment systems. This paper addresses several topics. First, the relationship between wetlands
and NPS control is discussed. Second, definitions of natural, restored, and created wetlands are
provided. Third, the regulations and policies that govern the consideration of wetlands to treat
NPS pollution are discussed. And fourth is some discussion of how protection and restoration can
provide the incidental benefit of controlling NPS pollution.
WETLANDS AND NPS POLLUTION
In their natural orientation in the landscape, wetlands receive sediment and nutrients via runoff
from uplands, and may retain or transform a portion of these inputs. Recognition of this
capability by water quality engineers and others led to efforts to replicate these functions. Much
of the impetus for using wetlands to treat NPS pollution stems from the successful use of wetland
treatment systems to treat point source discharges, particularly wastewater effluent. Constructed
wetlands have been used to treat primary or secondary treated wastewater effluent. Secondary
wastewater effluent has been used to restore degraded wetlands. In addition, natural wetlands
have been used to "polish" secondary treated wastewater effluent.
In the last couple of years, the Environmental Protection Agency (EPA) has recognized the
important relationship between wetlands and NPS pollution control. In EPA's Nonpoint Sources
2
-------
Agenda for the Future (U.S. EPA, 1989), there is recognition of the ability of wetlands to help
achieve NPS control. In 1990, EPA's Office of Wetlands Protection and Office of Water
Regulations and Standards issued national guidance to encourage coordination between wetland
and NPS programs (U.S. EPA, 1990a). In this guidance, EPA identified how State NPS control
programs provide an opportunity to create, restore, and enhance wetland resources for water
quality benefits. Although there is still a great deal to be learned about the relationship between
wetland functions and NPS pollutants, the coordination of State and National programs can be a
useful step toward the goal of maintaining the chemical, physical, and biological integrity of the
Nation's waters.
DEFINITIONS
The use of wetlands for treatment of point source and NPS pollution has resulted in the
introduction of a number of terms. Though the definitions of these terms have not been
standardized across Federal agencies, EPA has attempted to be consistent in recent publications
(i.e., National Guidance: Wetlands and Nonpoint Source (U.S. EPA, 1990a) and Water Quality
Standards for Wetlands: National Guidance (U.S. EPA, 1990b)).
"Natural" Wetlands are defined by EPA regulations as those areas that are inundated or saturated
by surface or groundwater 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 (40 Code of Federal Regulations [CFR] Parts 122.2, 230.3, and 232.2).
These systems are typically found as interfaces between open water and land or as isolated
systems, such as prairie potholes. Wetlands exhibit a wide array of values and functions
including wildlife habitat, flood water attenuation, nutrient retention or transformation, sediment
retention, fish spawning habitat, and fish and shellfish habitat.
3
-------
Restoration of Wetlands refers to returning a wetland from a disturbed or altered condition with
lesser acreage or function, to a previous condition with greater acreage or function (U.S. EPA,
1990a). Restoration may involve reestablishing the original hydrology, vegetation, or other
parameters that will restore the original appearance and/or functions of a wetland. Or restoration
may involve removing a source of impact to a wetland. Related to restoration is enhancement,
which refers to increasing one or more natural or created wetland functions. An example of
enhancement is adding water to a drought impacted western wetland to increase its area and
usefulness as habitat for migratory birds.
Creation of Wetlands refers to bringing a wetland into existence at a site where one did not
formerly occur (U.S. EPA, 1990a). Wetlands are generally created from upland areas due to the
intentional or nonintentional introduction of water or changes in surface elevation. Wetlands may
be created intentionally, such as for waterfowl or wildlife habitat, as part of the Clean Water Act
(CWA) Section 404 permit mitigation requirements, or incidentally, as a result of activities such
as constructing highways.
Constructed Wetlands are a subset of created wetlands that are designed and developed
specifically for water treatment. Constructed wetland cells or systems are characterized by
combinations of controlled water or effluent flow, specific vegetation (such as cattails or
bulrushes), a waterproof liner, and other material resulting in the chemical improvement of the
water flowing through. Constructed wetlands have been widely used as part of municipal
wastewater treatment processes.
4
-------
PRINCIPAL POLICIES
"NATURAL" WETLANDS
Almost all natural wetlands are considered "waters of the United States" and, as such, are
protected under several provisions of the CWA. Section 101(a) of the Act states that it is the
objective of the CWA "to restore and maintain the chemical, physical, and biological integrity of
the Nation's waters." Furthermore, it is stated in subsection (7) of this paragraph that it is the
national policy that programs for the control of nonpoint sources of pollution be developed and
implemented so that the goals of the Act can be met through the control of both point and NPS
pollution. These sections provide the base justification for protecting natural wetlands from NPS
pollution.
Section 319 of the CWA was enacted to address impacts to the Nation's waters from nonpoint
sources. Under this funding incentive program, States are to develop NPS assessment reports and
implement management programs. The purpose of the assessments is to identify "waters of the
United States" that are impaired or threatened by NPS pollution, as well as the activities causing
the impacts. Under this program, some States are developing and implementing enforceable
provisions to address NPS pollution and to protect wetlands and other waters. In 1990,
amendments to the Coastal Zone Management Act were passed that included requirements for
development of enforceable management measures to address problems of NPS pollution in the
coastal zone. These requirements will increase the enforcement leverage States need to address
detrimental impacts.
Another important provision of the CWA that affects the use of natural wetlands for NPS control
is Section 303. This Section requires States to adopt water quality standards that include
designating uses for wetlands and other waters and to assign water quality criteria that will meet
5
-------
those uses. Although NPS discharges to wetlands are not directly regulated, States must prescribe
use of best management practices (BMPs) to ensure compliance with the applicable State water
quality standards. The refinement of BMPs is the mechanism in Section 303 to gradually reduce
NPS impacts to all State waters.
Section 402 of the CWA has provisions for protecting "waters of the United States" from
stormwater impacts. These impacts are typically a result of water quantity as well as water
quality, as water is collected, transported, and discharged more efficiently through a stormwater
system resulting in detrimental impacts to receiving waters, such as wetlands. The recent
stormwater provisions regulate some of the formerly unregulated urban storm runoff, thus
providing some regulation of impacts to wetlands and other waters.
Although nonpoint sources are not generally subject to Federal permits, under certain
circumstances, Section 404 regulations may apply. Discharges of dredged or fill material into
"waters of the United States" must be authorized by a permit under Section 404 of the CWA. For
example, if a proposed NPS treatment activity involves discharging fill material into a wetland, a
Section 404 permit may be necessary. Review of individual Section 404 permit applications
includes review under the National Environmental Policy Act, and consideration of other
applicable Federal laws and executive orders (such as the Endangered Species Act).
RESTORATION OF WETLANDS
Restoration of wetlands is addressed in several sections of the CWA. First, restoration of
degraded wetlands is consistent with Section 101 of the CWA (see discussion of Section 101
above). Secondly, such wetlands are afforded protection for preservation of previous (before
degradation) uses or functions under Section 303. Thirdly, if the source of degradation is a
nonpoint source, the wetland should be identified in the State NPS assessment and included in the
6
-------
State's management program. And lastly, Section 404 regulations may apply if restoration efforts
involve discharging fill material into a wetland.
Restoring degraded wetlands is a unique opportunity not only to improve the quality of our
Nation's waters, but also to recapture other wetland functions that may have been lost. Restored
wetlands provide some inherent water filtration functions that benefit adjacent waters. However,
it should be remembered that these systems are considered to be "waters of the United States"
and, therefore, NPS pollutants cannot be directed to them for treatment.
CREATION OF WETLANDS
As defined above, created wetlands are a result of human activity, usually for a specific purpose.
Wetlands can be created through diversions of water, discharges of treated effluent, or as a result
of other activities, such as grading to lower soil surface elevation and allowing natural flows to
inundate areas. Created wetlands often provide multiple benefits which may include water
quality improvement. However, wetlands created for purposes other than wastewater treatment
(e.g., mitigation of wetland losses under Section 404 or development of waterfowl habitat)
generally receive the same protections under the CWA as restored or natural wetlands. Any
contributions of these created wetlands to NPS control cannot lead to degradation of the wetlands.
If wetlands are designed, constructed, and maintained for the sole purpose of water treatment,
they are generally not considered "waters of the United States;" therefore, there are no applicable
Federal regulations that govern their use (40 CFR, part 122.2). However, in these cases, water
leaving the created wetland cannot significantly degrade or alter the water quality or other
designated or existing uses of the adjacent waterbody.
7
-------
Created wetlands can be used as part of a landscape-based approach to controlling NPS runoff.
However, adequate consideration in design, maintenance, and monitoring must be given to
prevent detrimental impacts to groundwater, birds, animals, and other biota.
HOW PROTECTION AND RESTORATION CAN PROVIDE
INCIDENTAL NPS BENEFITS
NPS best management practices (BMPs) have for years been limited to structural measures that
function as sediment traps or reduce erosion. More recently, nonstructural measures and
processes are being developed and implemented under the auspices of pollution prevention.
Pollution prevention is the use of processes, practices, or products that reduce or eliminate the
generation of pollutants and wastes or that protect natural resources through conservation or more
efficient utilization. Foremost, wetlands should be protected due to the many values and
functions they provide. But, in addition, protection and restoration of wetlands are also
acceptable management measures for preventing the impacts to water quality that result when
wetlands are destroyed or degraded.
One principle of protection is avoiding impacts to wetland and riparian areas, when practicable,
to maintain existing beneficial uses (functions) and to meet existing water quality standards. A
similar principle applies to restoration: when conditions are appropriate, restoration of wetland
and riparian areas is preferred over structural BMPs, or restoration can be used in conjunction
with BMPs to gain not only water quality benefits but also additional benefits for "waters of the
United States." The basic premise behind these approaches is that the benefit of improved water
quality will be realized if wetlands and riparian areas are maintained (or restored) in the
landscape to perform their natural functions. When this approach is used, additional BMPs, such
8
-------
as buffer zones, must be utilized to ensure that there is no adverse impact to wildlife using the
wetlands and that the integrity of the wetlands will be maintained over time.
CONCLUSIONS
The ability of wetlands and riparian areas to filter and convert sediment and nutrients is widely
accepted. Our increasing understanding of these wetland values has resulted in efforts to utilize
or replicate natural systems. Wetlands, including degraded wetlands, are protected by provisions
of the CWA from discharges of pollutants. On the other hand, the stipulations under Federal
regulations are not as distinct for restored, created, or constructed wetlands used for NPS control.
It is important to note that the goals of NPS control programs and the goals of wetlands
protection programs are not mutually exclusive. Protection and restoration of natural wetlands
will result in the realization of NPS control benefits. Just as natural wetlands maintain water
quality in a natural setting, created and constructed wetlands developed specifically for purposes
of water treatment can be used to improve or maintain water quality of "waters of the United
States."
The possibilities for helping restore nature's original capabilities or for replicating nature's
processes are still evolving. Research on the functions and environmental effects of wetland
treatment systems is critical to understanding the differences between natural and created
systems, particularly with respect to how they can contribute to improvement of the physical,
chemical, and biological integrity of our Nation's waters. Additional research is also needed to
evaluate the potential for adverse impacts to wildlife and other biota utilizing wetlands receiving
NPS pollution. Until we are able to know with more certainty the long term effects and other
9
-------
impacts of wetland treatment systems on the environment, we should proceed cautiously and
refrain from the indiscriminate use of these systems for NPS pollution control.
10
-------
REFERENCES
U.S. Environmental Protection Agency, 1987. Report on the Use of Wetlands for Municipal
Wastewater Treatment and Disposal, Office of Water, Washington, DC.
U.S. Environmental Protection Agency, 1989. Nonpoint Sources Agenda for the Future, Office
of Water, Washington, DC.
U.S. Environmental Protection Agency, 1990a. National Guidance: Wetlands and Nonpoint
Source, Office of Water, Washington, DC.
U.S. Environmental Protection Agency, 1990b. Water Quality Standards for Wetlands: National
Guidance, Office of Water, Washington, DC.
11
-------
THE ROLE OF WETLAND WATER QUALITY STANDARDS IN
NONPOINT SOURCE POLLUTION CONTROL STRATEGIES
Doreen M. Robb
U.S. Environmental Protection Agency
Office of Wetlands, Oceans, and Watersheds
401 M Street, SW (A-104F)
Washington, D.C. 20460
ABSTRACT
States are required to develop water quality standards for their wetlands by the end of Fiscal Year
1993 Standards are vital to the protection of wetlands from a broad array of perturbations
including nonpoint source (NPS) pollution. The natural water quality functions of wetlands make
them potential components of NPS control strategies, but protection of wetland structure and
functions takes precedence over their use in NPS control. Narrative biological criteria are one
part of standards and can serve as a mechanism to address NPS pollution impacts. Criteria can
ako h* used as a baseline to determine the effectiveness of best management practices. Numeric
biocriteria are under development and will require additional research.
1
-------
INTRODUCTION
How much of a sediment load can a wetland receive without being degraded? How much
phosphorus can an inland marsh assimilate before it eutrophies or changes vegetation type? The
answers to questions such as these are vital to States seeking to protect their wetlands from
nonpoint source (NPS) pollution. Through the proper development and implementation of water
quality standards (WQS) for wetlands, States can protect their wetlands and associated water
quality functions. This paper will describe the components of WQS, then discuss the transition
from narrative biological criteria to numeric biocriteria, and will finish by identifying U.S.
Environmental Protection Agency (EPA) and State information needs for further development of
WQS.
WATER QUALITY STANDARDS FOR WETLANDS
In July of 1990, EPA published national guidance requiring States to develop WQS for their
wetlands by the end of Fiscal Year 1993 (USEPA, 1990). This will afford wetlands the same
level of protection currently provided to other surface waters. Water quality standards consist of
three parts; first, wetlands must have designated uses (e.g., aquatic life support, recreation,
flood water attenuation, groundwater recharge) consistent with the goals of the Clean Water Act.
Next, narrative and numeric criteria are assigned to protect those uses. Narrative criteria are
statements of attained or attainable conditions of a waterbody, and numeric criteria are numeric,
usually chemical, endpoints that specify the maximum contaminant level that can be present
without impairing the use of that waterbody. Third, each waterbody must have an
antidegradation policy and implementation methods that protect previously existing uses of the
waterbody as well as providing additional protection to higher quality and outstanding waters
where the quality exceeds that necessary to maintain the uses.
2
-------
Wetland standards are an important tool for States wishing to broaden the protection of their
wetlands beyond minimizing the disposal of dredged and fill material as provided for by Section
404 of the Clean Water Act. Comprehensive narrative criteria, if implemented aggressively, can
be used to protect wetlands from physical and hydrological modifications, including increased
water flow, sedimentation, and nutrient overenrichment. By establishing criteria for a healthy
wetland, a baseline exists against which changes in floral or faunal composition may be detected
and evaluated. These baselines provide a basis for monitoring and assessing whether NPS
pollution has detrimentally impacted a wetland. These changes can also be indications that best
management practices (BMPs) are not achieving the desired result. BMPs can then be
strengthened and refined until the waterbody in question is brought under compliance with water
quality standards. The information needed to develop wetland standards and criteria will also be
useful to designers of constructed wetland treatment systems in defining maximum loading rates
for pollutants, and in monitoring system performance.
Water quality standards have evolved over the years. Originally, the 1965 Water Quality Act took
a water quality-based approach by requiring States to develop WQS that specified levels of
cleanliness for waters. Similar in concept to what we have today, numeric chemical criteria were
developed to protect waters on a chemical-by-chemical basis. At the time, this water quality
approach was not very effective since the necessary progam infrastructure to enforce standards
was lacking. Subsequently, the 1972 Clean Water Act established a technology-based approach
that regulated individual point source discharges through National Pollutant Discharge
Elimination System (NPDES) permits. These permits set guidelines for effluent limits and are
basically "end-of-pipe" controls. Once the NPDES program was established and enforcement
mechanisms were in place, the 1987 Clean Water Act (CWA) amendments re-established a water
quality-based approach to supplement technology-based controls.
3
-------
BIOLOGICAL CRITERIA
The 1987 Clean Water Act identified remaining serious pollution problems including toxic
pollutants and NPS pollution. NPS impacts include sedimentation, eutrophication, hydrologic
modification, bioaccumulation of toxics, increased turbidity and, subsequently, decreased light
penetration. These impacts cannot be fully addressed on a chemical-by-chemical basis, and can
lead to secondary impacts such as changes in vegetation type and associated biota. In an effort to
address the primary impacts, EPA and the States have utilized narrative criteria. All States have
adopted variations of aesthetic narrative criteria—the "free froms." "Free froms" are general
statements, such as "free from debris, noxious odors, and taste." The development of narrative
biological criteria (and numeric biocriteria in future years), however, is a new area of emphasis
for EPA that will more effectively address secondary impacts.
NARRATIVE BIOCRITERIA
The development of narrative biological criteria and their application to NPS issues has important
implications for wetlands. Narrative biological criteria are new requirements for all surface
waters and are statements of attained or attainable condition necessary to protect the biological
integrity of the waterbody. These criteria are flexible and can be written as very general or very
specific statements. They can take the form of a "free from" statement, such as "free from
activities that would substantially impair the biological community as it naturally occurs due to
physical, chemical, and hydrologic changes." In their broadest sense, biological criteria protect
the physical and structural components necessary for healthy aquatic habitat as well as the biota.
For example, one State used more specific language to protect the natural hydrologic conditions
of a wetland:
4
-------
"Natural hydrological conditions necessary to support the biological and physical
characteristics naturally present in wetlands shall be protected to prevent significant adverse
impacts on: (1) Water currents, erosion or sedimentation patterns; ... (3) The chemical,
nutrient and dissolved oxygen regime of the wetland; (4) The normal movement of aquatic
fauna; . . . and (6) Normal water levels or elevations [emphasis added].
Narrative biocriteria can be effective in protecting wetlands from adverse impacts of NPS
pollution if implemented effectively through BMPs. The examples described above are good first
steps, and provide greater protection to wetlands than existed through numeric chemical criteria,
but they have limitations and are difficult to enforce. A key point, however, is that the
development of narrative biological criteria by States is based only on existing (and defensible)
scientific information. For this reason, States should not have difficulty in developing narrative
biocriteria immediately.
NUMERIC BIOCRITERIA
The next step after the development of narrative biocriteria, however, is the development of
numeric biocriteria. This is a future emphasis for EPA and the States and is based on the
development of new scientific information. Numeric biocriteria have the potential to be more
protective than narrative criteria because they are "hard" numbers and less subject to
interpretation; therefore, they should be easier to enforce consistently. Currently, EPA is
working on national guidance for rivers and streams; development of guidance for wetlands is
slated for the future. An example of a numeric biocriterion for a coastal State is "vegetative
diversity no greater than 2 species for salt marshes and no less than 25 species for a freshwater
inland marsh" or "percent vegetative species change shall be no greater than 10 percent." Similar
numbers could be derived for other components of the biota such as benthic invertebrates,
breeding birds, and amphibians.
5
-------
NPS CONTROLS AND WETLAND WATER QUALITY STANDARDS
Wetlands have an important role in the landscape through their ability to improve water quality
by filtering, transforming, and accumulating pollutants and thereby protecting adjacent rivers,
lakes, and streams. This "buffering" function, however, also encourages overuse, and this overuse
can compromise these and other wetland functions, such as wildlife habitat and aesthetic and
recreational values. While wetlands may be useful components of NPS pollution control
strategies, the first goal must be protection of wetlands from pollution. EPA does not allow
surface waters to be used as disposal sites for wastewater, and State water quality standards exist
to ensure the protection of State waters, including wetlands. Consider the following examples: a
State restores a degraded wetland for the purpose of slowing water that will flow off a new
parking lot or highway, or a private landowner restores a degraded riparian area for the purpose
of filtering sediment and nitrogen-enriched water from a nearby feedlot. At first glance, the
"use" of a restored wetland in both of these examples protects the water quality of a nearby
waterbody such as a lake or stream and that waterbody meets State water quality standards.
However, in the case of the highway, toxics accumulate in the wetland in amounts that exceed
toxics criteria, and in the feedlot example, the riparian area retains sediment that eventually
modifies the flow of water through that area, changing vegetation and runoff patterns. In both
of these examples, although the action benefitted the adjacent waterbody, it did so at the expense
of the wetland and, therefore, those actions violated the water quality standards. Additional
management practices may need to be put in place, such as vegetated grass filter strips to buffer
the wetlands. Regardless of the solution used, the integrity of both the wetland and the adjacent
waterbody must be protected.
6
-------
MONITORING AND ASSESSMENT
State standards, however, are only one mechanism to control and minimize the degradation of
wetlands by NPS pollution. The ability to detect impacts through biological monitoring is another
important tool to prevent degradation and is critical to the effective use of water Quality
standards. Biological monitoring and assessment enables States to compile baseline information
on wetland condition. This information can then be used in developing biological criteria. Once
a State has established biological criteria for its wetlands, the State then has a regulatory
mechanism to deal with impacts that violate State water quality standards.
Before States can establish a monitoring program for their wetlands, however, they need to know
what to measure (i.e., indicators) and how to recognize an impact. For example, if a State knows
to sample a particular benthic invertebrate, they also need to know what should be the expected
population dynamics of that invertebrate in a particular type of wetland under "natural" and
"perturbed" conditions. They need to know whether a change in vegetation is a natural
community succession or an indicator of increased phosphorus loading causing an extremely
diverse plant community to shift to a monotypic community. Such information will enable States
to establish a wetlands monitoring program as well as aid them in future numeric biocriteria
development.
FUTURE RESEARCH NEEDS
As States work to establish narrative biological criteria, EPA will be developing guidance for
developing numeric biological criteria. Increasing the technical science base is necessary before
numeric biocriteria can be developed. Examples of important research questions include:
7
-------
• How do altered hydrology and sedimentation patterns impact wetlands and how does the
biological community react to these changes?
• What should States measure to discern these changes?
• How much change is too much?
Such information is needed for all wetland types and regions so States can monitor for and
recognize these impacts when they occur.
CONCLUSIONS
Wetlands have an important function in landscape water quality. As a result, they are often
included in strategies for controlling NPS pollution. Water quality standards, however, apply to
wetlands as well as to other waterbodies. Therefore, wetlands must be protected from NPS
pollution through, for example, the use of BMPs such as upland buffers. State development of
effective water quality standards for wetlands requires further research on indicators of wetland
health, impacts and indicators of physical and hydrological alterations, and thresholds for
sediment, nutrients, and toxics loading. These types of information will enable States to protect
their wetlands through technically defensible water quality standards.
8
-------
REFERENCE
U.S. Environmental Protection Agency, 1990. Water Quality Standards for Wetlands: National
Guidance. EPA 440/S-90-011, Office of Water Regulations and Standards, USEPA, Washington,
DC.
9
-------
RECOMMENDATIONS FOR RESEARCH TO DEVELOP GUIDELINES
FOR THE USE OF WETLANDS TO CONTROL RURAL NONPOINT SOURCE POLLUTION
Arnold G. van der Valk
Department of Botany
Iowa State University
Ames, Iowa 50011-1020
and
Robert W. Jolly
Department of Economics
Iowa State University
Ames, Iowa 50011-1020
ABSTRACT
Natural wetlands should not be used to reduce rural nonpoint source (NPS) problems. Properly
designed restored or created wetlands, however, can be used for this purpose in many agricultural
landscapes. Agricultural landscapes in which wetlands can be easily restored are the most suitable
areas. Major technical issues that need to be resolved before effective and realistic guidelines can
be developed for using restored wetlands to reduce NPS pollution include (1) the effects of
cr :3minants, particularly sediments and pesticides, on restored wetlands; (2) the fate of organic
( aminants in restored wetlands; (3) the development of site selection criteria; and (4) the
elopment of design criteria. There also are many social, economic and political barriers to
. .ing re
-------
INTRODUCTION
Rural nonpoint source (NPS) pollution is a landscape-level problem. Ultimately, eliminating,
minimizing, or redirecting the movement of materials within agricultural landscapes is the only
effective means of reducing NPS pollution to acceptable levels. Such a landscape-level solution
should include a combination of in-field and off-field approaches. In-field approaches should
include reduced inputs of nutrients and pesticides, best management practices to reduce soil
erosion, and improved cropping systems. Off-field practices should include more appropriate
land use, establishment of vegetated buffers or filter strips between farm fields and aquatic
systems, and the creation of sinks for NPS contaminants near their points of origin. Off-field
modifications, such as restoring or creating wetlands that act as nutrient sinks, offer potential as
part of a comprehensive landscape-level strategy for NPS pollution reduction.
Wetlands may be the most cost-effective sinks for contaminants in many agricultural landscapes.
Other papers in this volume review available data on wetlands as nutrient traps (Mitsch, Rodgers)
and the construction of wetlands for wastewater treatment (Hammer). We will take it as a given
that properly designed and constructed wetlands will intercept, store, and/or break down the
various contaminants normally found in agricultural runoff (sediments, nutrients, pesticides). We
do not wish to imply, however, that the fate of all contaminants in wetlands is fully understood
or that the impact of most contaminants on wetlands is understood at all. In fact, very little
about the latter is known. We also will assume that wetlands can be restored and created in
appropriate places in the landscape as needed. In recent years, there has been enough work done
on wetland restoration and creation that these per se are not issues. Much of the available
information on wetland restoration and construction is summarized in Kusler and Kentula (1990)
and Hammer (1989). Ancillary ecological and wildlife benefits from restoring wetlands in
agricultural landscapes are reviewed by Knight (this volume). Although these ancillary benefits
2
-------
are important, and the creation of wildlife habitat has been the primary motivation for most
wetland restorations so far, we will ignore them.
We do not recommend the use of natural wetlands as sinks for NPS contaminants. Natural
wetlands in agricultural landscapes are usually rare and often are already at risk because their
hydrology has been altered by regional drainage and because they may already receive significant
inputs of agricultural runoff (Davis et al., 1981; Stuber, 1988; Neely and Baker, 1989). These few
remaining wetlands are usually important habitat for many plant and animal species and are also
important recreational areas, particularly for waterfowl hunters. Anyhow, the size and location
of natural wetlands may not make them effective as sinks. If rivers, lakes, and reservoirs are to
be protected from degradation by contaminants in agricultural runoff, natural wetlands should
receive comparable protection. Wetlands have no magical properties that make them immune
from degradation. Guidelines are currently being developed for water quality standards for
wetlands (USEPA, 1990), and these should be applied to natural wetlands in agricultural
landscapes. Restoring or creating wetlands as sinks in agricultural landscapes, however, makes
sense, and all our comments assume that it is restored and created wetlands that will be developed
as sinks, not natural wetlands.
We are defining restored wetlands as wetlands established in natural basins whose natural
wetlands had been artificially drained. Restoring a wetland is usually done by blocking or
removing the basin's man-made drainage system. A created wetland is a wetland established in
an area that historically was not a wetland. Creating a wetland is done by excavating a suitable
basin or constructing dikes, and it often requires redirecting water to the new basin. Creating a
wetland is usually more expensive than restoring one. Constructed wetlands are a subset of
created wetlands that were established specifically for wastewater treatment. To date most
constructed wetlands have been built to treat point sources of human or animal waste or to treat
3
-------
urban storm runoff (Hammer, 1989). Although both restored and created wetlands should be
equally effective for reducing rural NPS problems, we believe that the additional cost of created
wetlands will make them, under many circumstances, uneconomical. So, we recommend that, at
least initially, research focus on the use of restored wetlands in agricultural landscapes where
wetland restorations can be done easily and cheaply.
Wetland restoration is not appropriate for all kinds of agricultural landscapes. Regions whose
agricultural landscapes contain drained wetlands are the best candidates for this approach. As a
first cut, suitable regions can be identified by inspection of Figure 1, which is taken from Pavelis
(1987) and reproduced in Dahl (1990). Figure 1 is a map of the United States that summarizes
data on wetland drainage. As a second cut, Figure 1 can be superimposed on the surface water
quality maps of the United States in Omernik (1977) or on fertilizer and atrazine application
maps in Moody (1990). This will delimit regions that have lost most of their wetlands and that
have the most polluted surface water. These areas are frequently congruent. The restoration of
wetlands in these regions should result in significant improvements in water quality. Based on
both wetland drainage and surface water quality criteria, the most suitable regions are the
Midwestern corn belt (Iowa, Illinois, Indiana, and western Ohio), the lower Mississippi River
valley (parts of Arkansas, Tennessee, Mississippi, and Louisiana), and southern Florida. There
are three reasons why these areas are suitable: (1) they have a climate and topography suitable
for wetlands; (2) land use in these regions is predominantly row-crop agriculture, the major
source of NPS pollution; and (3) surface or subsurface drainage networks connecting drained
wetlands can be used to channel runoff into restored wetlands. In other words, it is in these
drained landscapes that restoring wetlands will be easiest and will do the most to improve water
quality.
4
-------
To determine what research needs to be done to ensure that effective guidelines are developed
for using wetlands as sinks, we will first consider what are the major unresolved technical
questions and then what are the major social and economic barriers to the implementation of this
approach. Finally, we will recommend a number of research projects that we believe need to be
completed before final guidelines for the implementation of this approach to NPS pollution
abatement can be developed.
*
TECHNICAL ISSUES
Four general technical questions need to be resolved before the approach of restoring wetlands in
watersheds to reduce rural NPS pollution can be implemented: (1) what effects will agricultural
contaminants have on restored wetlands? (2) what is the fate of contaminants in restored
wetlands? (3) where should restored wetlands be sited? and (4) what are appropriate design
criteria for restored wetlands?
WHAT ARE THE EFFECTS OF AGRICULTURAL CONTAMINANTS ON WETLANDS?
There will be many constraints on the restoration of wetlands in agricultural landscapes that will
largely decide how these restorations will be done. The most significant of these is cost. Most
wetlands will need to be constructed at the least cost per wetland, farm, watershed, and region.
There also will be strong economic and social pressure to take the minimum amount of land out
of production. Consequently, most restored wetlands will have no, or only minimal, water
control structures; will have little, if any, basin excavation; will not normally be planted or seeded
to reestablish the vegetation; and will tend to be small. Many of these wetlands may be little
more than small, shallow depressions that hold water seasonally. Initially, their vegetation
composition, primary production, secondary production, and nutrient cycles will not resemble
those of natural wetlands.
5
-------
Because the primary source of water for restored wetlands will be agricultural runoff, often
containing high amounts of nutrients, sediments, and pesticides, these wetlands may never
become similar in composition, structure, or function to natural wetlands, which developed
without such inputs. The impact of contaminants on restored and natural wetlands has been little
studied (see Stuber, 1988), but it is not inconsequential. Reports of changes in the composition of
natural wetlands due to increased inputs of nutrients from surrounding agricultural systems (e.g.,
the ongoing invasion of cattails into the northern Everglades in Florida) or inputs of pesticides on
invertebrates (e.g., in prairie potholes; see Grue et al., 1986, 1988) suggest that there will be
impacts. There are two recent compilations of information on the impacts of contaminants in
agricultural runoff on wetlands, both emphasizing impacts on waterfowl (Sheehan et al., 1987;
Facemire, no date). The information reviewed in these two publications shows unequivocally that
contaminants impact both the flora and fauna of wetlands. What is not known is the extent to
which contaminants, particularly pesticides and sediments, will alter the development of restored
wetlands. Ongoing studies of the effect of sediment on litter decomposition and seed bank
recruitment (van der Valk and Jurik, unpublished) indicate that contaminant effects deserve
careful and detailed study. How closely restored wetlands will come to resemble natural wetlands
is unknown, but it is unlikely that they will ever very closely resemble pristine natural wetlands.
WHAT IS THE FATE OF AGRICULTURAL CONTAMINANTS IN WETLANDS?
Agricultural runoff contains a complex and highly variable mix of dissolved and suspended
contaminants (Neely and Baker, 1989). Its composition is a function of precipitation, topography,
regional land use patterns, soil characteristics, fertilizer and pesticide application rates, tillage
practices, etc. Of the three major classes of contaminants (nutrients, pesticides, and sediments)
usually found in agricultural runoff, only the fate of nutrients in wetlands has ever been
adequately studied (Howard-Williams, 1985; Bowden, 1987; Neely and Baker, 1989). The results
6
-------
of these studies indicate that denitrification is the major mechanism for the removal of nitrogen
and that sedimentation is ultimately the major mechanism for the removal of phosphorus.
Restored wetlands, as do natural wetlands (Martin and Hartman, 1987; Phillips, 1989), will act as
settling basins for sediment in agricultural runoff. What is a sustainable loading of sediment?
What criteria should be used to decide what is a sustainable loading? Because very little research
on sediment impacts on freshwater wetlands has been done (van der Valk et a!., 1983), it is
impossible to answer these questions.
There also is little known about the fate of most pesticides in any kind of wetland.
Consequently, there is no way to determine what sustainable loading rates of different pesticides
should be for restored wetlands. The effects of nutrient levels and sediment loads on pesticide
degradation rates in restored wetlands deserves particular attention.
In short, the lack of information on the fate and sustainable loading rates of sediments, pesticides
and, to a much lesser extent, nutrients in restored wetlands makes it impossible to develop
guidelines for the use of these wetlands to control rural NPS pollution.
WHERE SHOULD RESTORED WETLANDS BE SITED?
Figures 2 and 3 illustrate the two basic scenarios for the placement of wetlands in an agricultural
landscape. In Figure 2, the wetland is placed at the base of the watershed, and in this position all
water leaving the watershed will pass through it. The major advantage of this placement is that
only one wetland must be established per watershed. In contrast, Figure 3 shows the wetlands
distributed around the watershed so that each subwatershed has its own wetland. Individual
wetlands would be much smaller and possibly easier to establish. The advantage of the
distributed-siting pattern is that less runoff and erosion might occur in the whole watershed as a
7
-------
result of storing water and sediments high in the watershed (Novotny and Chesters, 1989). This
could reduce the total area of wetland needed in a watershed.
Because so little work on this topic has been done, the optimal placing of wetlands in agricultural
watersheds needs to be thoroughly examined. Optimal placement likely will vary from region to
region because of different topographies, land use patterns, and layouts of surface and subsurface
drainage networks.
WHAT ARE APPROPRIATE DESIGN CRITERIA?
Ideally the use of wetlands in a watershed to improve water quality will be part of a more
comprehensive plan to reduce NPS problems, including other off-field and improved in-field
practices. Improved in-field practices, e.g., lower fertilizer application rates, could significantly
reduce the area of wetlands needed in a watershed. Consequently, it makes sense to restore
wetlands as part of a comprehensive watershed plan so that the area of wetland needed can be
more realistically assessed.
In a water quality context, the single most important feature of a restored wetland is its size. The
appropriate size of a restored wetland will depend on (1) the contaminant of greatest local
concern that requires the longest residence time for its degradation, and (2) the percent reduction
of this contaminant that is required seasonally, annually, or interannually. In operational terms,
the size of a wetland should be determined by the expected total mass of various contaminants in
the runoff entering it during some period and the tolerable or sustainable loading of these various
contaminants per unit area of wetland. Both contaminant delivery and sustainable loading rates
are difficult to quantify, the former because precipitation events are highly variable seasonally
and interannually, and the latter because the sustainable loading for each contaminant is likely to
be different. Turnover time, in turn, is a function of both precipitation patterns, wetland size,
8
-------
location of inflows and outflows, flow patterns within the wetland, etc. As noted previously,
little is known about sustainable loading rates for restored wetlands for most agricultural
contaminants.
Although many models have been developed to estimate the delivery of sediments and
contaminants from nonpoint sources, they generally decrease in accuracy with increasing
watershed size (Novotny and Chesters, 1989). The most reliable method for calculating deliveries
of dissolved and sediment-bound contaminants to wetlands still needs to be determined.
Eventually, from available models and more detailed delivery studies, some simple rule of thumb
will need to be developed (e.g., one hectare of wetland is needed for each 100 hectares of
watershed) or a simple delivery model will have to be chosen that can be used on a daily basis in
the field.
One important design feature of wetlands that needs particular attention is maximizing residence
times of runoff in wetlands. Currently, most wetland restorations are done to create waterfowl
habitat. In landscapes where wetlands were drained with drainage tile networks, as in parts of
the Midwest, wetlands are often created by interrupting drainage tiles. The standard way this is
done is illustrated in Figure 4. Because inputs and outputs are physically adjacent, this design,
although suitable for creating waterfowl habitat, is inappropriate for water quality purposes
because residence time of water is effectively zero when the basin is full.
SOCIAL AND ECONOMIC ISSUES
We believe that social and economic considerations ultimately will decide whether landscape-level
approaches, such as the restoration of wetlands in watersheds to reduce rural NPS problems, can
and will be implemented. We have reached this conclusion because of our experiences with
9
-------
several agricultural landscape reconfiguration projects. Technical issues, although they are far
from insignificant, are not nearly as daunting as the organizational, social, political, and economic
issues that quickly arise when such projects are attempted. Agricultural landscapes will need to
be managed to meet not only environmental goals but also economic and social goals. The
cooperation of farmers and local community leaders is essential for the success of any landscape
management program. Existing institutions and organizations at the county, State, and Federal
levels will have a significant influence on the adoption and implementation of landscape
management programs. Consequently, economic and social analyses are essential for
understanding local attitudes, institutions and organizations, and economic constraints that must
be addressed to ensure the success of this approach.
Among the major social and economic issues that need to be addressed are: (1) what is the most
appropriate landscape unit for implementation of a wetland restoration program? (2) where
should wetlands be sited? (3) who will decide where wetlands will be sited? (4) how can
landowners' cooperation be obtained? (5) who will pay for the restoration and creation of these
wetlands? and (6) how cost effective is this approach, from both a private and public
perspective? The answers to these questions will undoubtedly require changes or adjustments in
a variety of public policies and regulations. Relevant public policy and regulatory issues are
discussed by Fields (this volume). Here, we will only mention a few economic and social issues
that we feel need to be raised, researched, and debated prior to the implementation of any large-
scale, landscape-level program to restore wetlands.
WHAT IS THE MOST APPROPRIATE LANDSCAPE UNIT?
That the watershed is the natural unit for dealing with water-quality issues has long been
recognized by researchers, many policy makers, and many administrators, but rarely by
landowners and even more rarely by public officials. Each watershed is a unique natural
-------
geomorphological unit in which uplands and lowlands are linked hydrologically. The impact of
land use and land-use changes on water quality can best be examined within a watershed, because
each watershed is an isolated entity. In other words, human impacts on the environment can be
most easily quantified within watersheds, and landscape modifications, such as restoring wetlands
to improve water quality, can best be evaluated within watersheds. Economic externalities that
arise from altering land use and management on individual farms can be more easily internalized
if the basic unit is the watershed rather than the farm. It is not possible, however, to internalize
all environmental costs at the watershed level. On the whole, the watershed is the most nearly
ideal unit for integrating environmental, agricultural, economic development, and social programs
in rural areas.
Can landscape-level programs be successfully organized on a watershed basis? To answer this
question, other questions must be considered: What is the best way to do this? What kinds of
economic and social incentives will be needed to induce farmers within a watershed to make
collective decisions? Who will write guidelines for developing wetland restoration programs and
who will approve programs for each watershed? How much of a problem will absentee landlords
be?
Because property, administrative, and political boundaries normally do not coincide with
watershed boundaries, landowners and public officials normally do not think in terms of
watersheds, and implementing soil erosion and water quality programs in a watershed framework
remains politically and socially challenging. For example, the 1990 Food, Agriculture,
Conservation and Trade Act retains, in effect, individual farms as the basic landscape unit.
There are, however, already laws and programs that use watersheds as a landscape unit, e.g., the
Watershed Protection and Flood Prevention Act, PL 566. PL 566 can provide Federal funding to
local sponsorship groups for flood control, water management and supply, grade stabilization and
11
-------
soil erosion prevention, and water quality projects. Determining how watersheds can be
incorporated into public laws and regulations dealing with water quality issues requires careful
consideration and investigation of existing and potential legal, economic, and regulatory options.
As noted, the landscape unit in agricultural programs, including most soil conservation programs,
has been the farm or the individual tract. Because the placement of wetlands in a watershed is
crucial if wetlands are to be effective as sinks for contaminants, it will be necessary to establish
them in specific places. Historically, most soil conservation programs have been voluntary, i.e.,
there has been little targeting of resources to treat areas with the most severe erosion problems.
Economic analyses of past programs suggest that this lack of targeting of resources has resulted in
soil erosion programs that often were not cost effective in many regions of the country, including
the Midwestern corn belt (Ribaudo et al., 1989). If the use of wetlands for water-quality
programs is going to be cost effective, it is essential that it be done as part of a targeted,
landscape-level, and comprehensive program.
Watersheds come in a variety of sizes. A determination must be made as to the appropriate size
or stream-order for watersheds to be used for controlling NPS pollution. If too large a watershed
is used, planning and implementation will become difficult because too many political and
administrative units will be involved. If too small, the bureaucracy required to develop and
implement a program may be too costly.
WHERE SHOULD WETLANDS BE LOCATED IN WATERSHEDS?
Where to restore wetlands in a watershed is, in part, a technical issue as outlined previously, but
it is also a social, economic, and political issue. Figures S and 6 illustrate some advantages and
disadvantages of siting wetlands at the base of a watershed or distributing them all over the
watershed when farm boundaries are considered. Siting one large wetland at the terminus of a
12
-------
watershed (Figure 5) will involve dealing with fewer landowners, possibly only one. Other
landowners in the watershed may not need to participate in the wetland restoration program at
all. Basal siting means taking more land out of production on farms at the lower end of a
watershed. Distributing the wetlands around the watershed (Figure 6) will mean having to deal
with more landowners, but it means that less land may be taken out of production on any one
farm. If the distributed approach is to work, nearly all landowners in a watershed will have to
participate. The distributed model has the advantages of fairness, as no one landowner normally
is doing more than what is required to treat the runoff from his or her land. Further, complex
substitution procedures for landowners within a watershed may be avoided.
WHO WILL MAKE SITING DECISIONS?
Ideally, decisions about how to reduce agricultural NPS pollution in a watershed will be made
voluntarily and collectively by the farmers who own property in the watershed and by
appropriate local officials using technical alternatives provided by the Soil Conservation Service
and Cooperative Extension Service. Public funds will undoubtedly continue to be used to install
a variety of best management practices to reduce agricultural NPS problems. Restored and
created wetlands are only one off-site control measure that could be incorporated into a
watershed-level NPS pollution reduction plan.
How best to organize watershed-level programs that include wetland restorations is a key issue.
One possibility would be for landowners within a watershed to organize a sponsorship group to
develop restoration plans and request funding for them. This is an unlikely scenario in the
immediate future because farmers know little about wetlands, their water-quality benefits, or
their restoration. Information, education, and demonstration programs on the role of wetlands in
agricultural watersheds are therefore needed. Alternatively, existing infrastructures at the county
level, e.g., Soil and Water Conservation Districts, could be used to set up watershed-level
-------
committees within each county that would develop a plan for its watersheds. Since watersheds
will often cross county lines, protocols for dealing with multicounty watersheds may need to be
developed. Existing procedures and protocols for dealing with flood control and other
multijurisdictional issues, however, may be adequate.
Although we think that it is preferable that both organizational planning and implementation be
done at the watershed level, this is not essential. Wetland restoration programs must be developed
at the watershed level, but the administration of these programs can be done using different
existing administrative or political boundaries, e.g., counties. If administrative boundaries other
than the watershed are used, coordination among different units within a single watershed will be
essential. Although it will be more complex to use existing administrative boundaries to plan and
implement watershed-level programs, the advantages of using the watershed as a unit will
outweigh the added administrative burden that this causes.
WHAT IS THE BEST WAY TO OBTAIN LANDOWNER COOPERATION?
Getting farmers to think in terms of watersheds will require educating them about wetlands and
water-quality issues and the technical reasons for organizing programs at the landscape level.
Surveys of farmers' attitudes to watershed-level programs and wetland restorations are needed to
determine what are the social and economic obstacles to such an approach and what kinds of
incentives will be needed to get farmers to adopt them. These surveys will need to be done in
several regions of the country; data should be collected and digested before any adjustments and
changes in regulations and public policy are implemented. The insights gained from these
surveys also should be used to develop educational programs for farmers on why wetlands are
beneficial in agricultural landscapes and why watersheds are environmentally, socially, and
economically the most natural and logical unit for dealing with NPS pollution problems.
14
-------
WHO WILL PAY?
We already know the answer to this question, at least in general terms. Public funds primarily
will compensate farmers for land taken out of production and restored to wetlands. At the
Federal level, there are several existing programs within the USDA, including the new Wetlands
Reserve Program, and within the Fish and Wildlife Service, including the various joint ventures
that are part of the North American Waterfowl Management Plan, for wetland acquisition,
restoration, and creation. Also, there are many State-level wetland protection, acquisition, and
restoration programs as well as water quality protection programs that are potential sources of
funding, e.g., the Reinvest in Minnesota (RIM) program. Lastly, there are many conservation
organizations that will fund habitat restorations projects, e.g., Ducks Unlimited, Inc. How best to
coordinate funding within Federal agencies and between the Federal government and State
governments needs to be explored. How best to fund watershed restoration programs needs to be
examined, and determination made as to what extent it is possible to internalize the costs of
wetland restorations.
HOW COST EFFECTIVE IS THIS APPROACH?
An evaluation of the potential cost effectiveness of this approach should be made before it is
proposed on a national scale. Before this approach can be evaluated, however, water-quality
goals for agricultural runoff need to be established nationally. Once national goals are proposed
or established, spatial models of selected agricultural watersheds developed using a Geographic
Information System (GIS) or other suitable modeling approach, can be used to determine the
contribution of wetlands of various sizes and at various locations to achieving these goals in a
particular watershed. NPS contaminant delivery models should be used with these spatial models
initially to determine the location and size of the wetlands needed. Economic models applied to
these spatial models can be used to determine the costs of restoring wetlands and the benefits
derived from improved water quality. Although they all have shortcomings, models currently
-------
exist that can assess pollutant delivery to a wetland (e.g., Knisel, 1980; Young et al., 1987, Braden
et al., 1989; Lane and Nearing, 1989; Novotny and Chesters, 1989). Linking watershed process
and economic models, however, remains a major challenge.
PROPOSED RESEARCH
A great deal is known about the fate of nutrients in agricultural runoff in wetlands (Neely and
Baker, 1989) and how to restore (Kusler and Kentula, 1990) and construct wetlands (Hammer,
1989). Nevertheless, as outlined above, using wetlands as contaminant sinks in agricultural
landscapes raises many technical, economic, social, and legal questions for which there are no or
only partial answers. Among the most important topics that need to be studied before guidelines
can realistically be developed for implementing this approach are the fate and effect of
contaminants other than nutrients, criteria for siting and designing restored wetlands, the rural
population's attitudes toward landscape-level approaches to NPS reduction, and the best way to
organize and fund wetland restoration programs at the watershed level.
The proposed research projects outlined below should be integrated as much as feasible. This can
best be done by establishing several major regional studies to research relevant technical, social,
and economic problems. Regional differences in landscapes (topography, soil characteristics,
precipitation patterns, etc.), farming practices, attitudes toward wetlands, and economic realities
may require different regional strategies.
There are eight topics that we think should be investigated. Each of these is described briefly
with only a few general recommendations for approaching these studies suggested where
appropriate or obvious. Specific recommendations for conducting most studies have not been
made because we do not think that we are the best qualified to make them. Five of the
-------
recommended research topics deal with technical issues, the remaining three with social and
economic issues. The most promising agricultural region in which to initiate studies is the
Midwestern corn belt because it is by far the largest geographic region where this approach is
appropriate (Figure 1) and it is the agricultural region with the poorest water quality (Omernik,
1977).
RECOMMENDED STUDIES
(a) Whole watershed demonstration studies are needed in several regions to establish the
feasibility and utility of using restored wetlands as sinks for contaminants in agricultural runoff.
The Food, Agriculture, Conservation and Trade Act of 1990 (S. 2830) established the Wetlands
Reserve Program (section 1438) that allows the enrollment of up to 1,000,000 acres of land for
restoring and protecting wetlands by 1995. This new program could be used with the agricultural
water quality incentives (section 1439) of the 1990 Act to help fund the demonstration studies.
Such watershed-level studies will provide essential information on (1) the costs of constructing
wetlands; (2) the best way to establish wetlands in different kinds of agricultural landscapes; (3)
how acceptable this approach is in different regions of the country; (4) how effective wetlands
are as sinks for contaminants in agricultural runoff, and (5) how best to organize and administer
a wetland restoration program. All the other recommended studies should be done as part of, or
in conjunction with, these studies.
(b) Studies of the effectiveness of restored or recently created wetlands as sinks for nutrients and
organic contaminants are essential. Such studies should determine the fate of different
contaminants and their sustainable loading rates. These studies can most easily and economically
be done using mesocosms, but field studies of the actual performance of restored or recently
created wetlands also should be done. Reliable sustainable loading rates are essential for
calculating the appropriate size of restored wetlands.
17
-------
(c) Studies are also required of the effects of NPS contaminants on the development of restored
wetlands and on existing wetlands. The most worrisome contaminants are sediments, herbicides,
and insecticides. There are already many studies that suggest that contaminants can have adverse
effects on wetlands (see Sheehan et al., 1987; Stuber, 1988; Facemire, no date). Dosage studies
using mesocosms are the most logical approach to impact studies, and ideally impact studies
should be combined with mesocosm loading studies. Field studies of impacts of contaminants on
wetlands are also needed to examine spatial patterns of impacts within wetlands and long-term
effects on wetlands.
(d) Landscape simulation models of the origin and movement of NPS pollutants in selected
agricultural landscapes should be developed to determine the number of wetlands needed and
where they should be located to reduce contaminants to acceptable levels. These models can be
used* to evaluate both environmental and economic effectiveness of different site selection criteria
and wetland designs in diverse watersheds. If feasible, these models should be designed so that
they can be used with minimal data for the routine planning of watershed-level restoration
programs.
(e) Site selection and design criteria for created and restored wetlands need to be established. A
literature review and synthesis plus expert opinions on siting and sustainable loadings from
scientists in different regions of the country is the best approach to developing siting and design
criteria. Among the questions that should be considered are: (1) in which regions of the country
should this approach be used? (2) what are the major contaminants in agricultural runoff in each
region? and (3) what will be the potential impacts of establishing wetlands on local and regional
hydrology, particularly groundwater levels?
18
-------
(f) Studies should be conducted of farmers' and local business and community leaders' attitudes
toward a landscape approach to NPS problem reduction and targeting watersheds as units in
government programs. This information is needed to develop educational programs for farmers
on wetlands and water quality, to plan how best to implement landscape approach solutions, and
to decide what public policy changes and administrative structures will be needed.
(g) Legal and public policy issues created by wetland restoration programs need to be researched.
This is needed to develop effective interagency programs for administering, funding, and
implementing these programs. One important legal issue that needs to be examined is the
implication of disrupting drainage networks to establish wetlands. What are the liabilities for
agencies and farmers who fund or restore wetlands if other landowners in the watershed believe
that this will negatively affect drainage of their land? Disputes between adjacent landowners and
government agencies over wetland restorations have already occurred in Iowa and resulted in the
cancellation of restoration projects. Another important issue is whether restored or created
wetlands that were established to treat agricultural runoff should be designated waters of the
United States and thus become subject to Federal regulation.
(h) Studies are needed to determine the economic costs and benefits of this approach to rural NPS
pollution reduction and how costs associated with it can be internalized. This information is
critical if effective public policies are to be designed and implemented. Recreational, wildlife,
and other ancillary values of having wetlands in agricultural landscapes should be considered in
these economic analyses. These studies should be done in conjunction with both the modeling
studies and the demonstration studies.
19
-------
SUMMARY
The restoration of wetlands in appropriate agricultural landscapes as sinks for contaminants in
runoff is feasible. Major technical issues that need to be resolved include the effects of
contaminants, particularly sediments and pesticides, on wetland ecosystem composition, structure,
and function; the fate of organic contaminants in wetlands; the development of site selection
criteria; and the development of design criteria. Much of the work on nutrient and contaminant
processing in wetlands has been done in natural or highly engineered wetland systems often
designed for tertiary sewage treatment (Hammer, 1989). To be economically acceptable, wetlands
in agricultural landscapes must be restored or created with minimal cost and effort. There is
little work on organic contaminant processing in recently restored or created wetlands. There is
even less work on the effect of contaminants and sediments on invertebrate production,
denitrification rates, litter decomposition rates, species composition of vegetation, etc.
There are many social, economic, and political barriers to the adoption of landscape-level
approaches, such as wetland restoration, to solving NPS problems. These include having
watersheds accepted as the basic landscape unit both legally and socially, establishing mechanisms
at the landscape level to develop restoration programs and gain acceptance for them, and
determining needed incentives for wetland restoration programs and ultimately their economic
effectiveness, considering both private and public costs and benefits.
Eight research projects are identified that need to be completed before guidelines for establishing
watershed-level wetland restoration programs can be developed. These include five technical
projects (watershed-level demonstration projects, landscape or watershed simulation models, site
selection criteria, sustainable loading rates for contaminants, and effects of contaminants on
wetlands) and three economic and social projects (attitudes of farmers and rural leaders, legal and
public-policy implications, economic costs and benefits).
20
-------
ACKNOWLEDGMENTS
We would like to thank Roger Link of the Soil Conservation Service in Des Moines for
information on various Federal and State programs dealing with water quality. We also would
like to thank Bill Crumpton, Tom Jurik, Bob Kadlec, Vernon Meentemeyer, Roger Link, and
Suzanne van der Valk for their comments on various drafts of the manuscript.
21
-------
REFERENCES
Bowden, W. B., 1987. The biogeochemistry of nitrogen in freshwater wetlands. Biogeochemistry,
4: 313-348.
Braden, J. B., G. V. Johnson, A. Bouzaher, and D. Miltz, 1989. Optimal spatial management of
agricultural pollution. American Journal of Agricultural Economics, 71: 404-413.
Dahl, T. E., 1990. Wetland losses in the United States 1780's to 1980's. U. S. Department of the
Interior, Fish and Wildlife Service, Washington, DC.
Davis, C. B, J. L. Baker, A. G. van der Valk, and C. E. Beer, 1981. Prairie pothole marshes as
traps for nitrogen and phosphorus in agricultural runoff. In: B. Richardson (ed.), Proceedings of
the Midwest Conference on Wetland Values and Management. Freshwater Society, Navarre, MN.
Facemire, C. F., No date. Impact of agricultural chemicals on wetland habitats and associated
biota with special reference to migratory birds: a selected and annotated bibliography. U. S. Fish
and Wildlife Service and South Dakota Agriculture Experiment Station, South Dakota State
University, Brookings, SD.
Grue, C. E., L. R. DeWeese, P. Mineau, G. A. Swanson, J. R. Foster, P. M. Arnold, J. N.
Huckins, P. J. Sheehan, W. K. Marshall, and A. P. Ludden, 1986. Potential impacts of
agricultural chemicals on waterfowl and other wildlife inhabiting prairie wetlands: An evaluation
of research needs and approaches. Transactions of the North American Wildlife and Natural
Resources Conference, 51: 357-383.
22
-------
Grue, C, E., M. W. Tome, G. A. Swanson, S. M. Borthwick, and L. R. DeWeese, 1988.
Agricultural chemicals and the quality of prairie-pothole wetlands for adult and juvenile
waterfowl-what are the concerns, pp. 55-64. In: P. J. Stuber (ed.), Proceedings of the National
Symposium on Protection of Wetlands from Agricultural Impacts. U. S. Fish and Wildlife
Service, Biol. Rep. 88(16), Washington, DC.
Hammer, D. A. (ed.), 1989. Constructed Wetlands for Wastewater Treatment. Lewis Publishers,
Inc., Chelsea, MI.
Howard-Williams, C., 1985. Cycling and retention of nitrogen and phosphorus in wetlands:
theoretical and applied perspective. Freshwater Biology, 15: 391-431.
Knisel, W. G., 1980. CREAMS: a field-scale model for chemical runoff, and erosion from
agricultural management systems. Conservation Research Report No. 26. U. S. Department of
Agriculture, Washington, DC.
Kusler, J. A. and M. E. Kentula (eds.), 1990. Wetland Creation and Restoration: The Status of
the Science. Island Press, Washington, DC.
Lane, L. J. and M. A. Nearing, 1989, USDA-Water Erosion Prediction Project (WEPP).
Hillslope Profile Model Documentation. NSERL Report No. 2. National Soil Erosion Research
Laboratory, Agricultural Research Services, West Lafayette, IN.
Martin, D. B. and W. A. Hartman, 1987. Effect of cultivation on sediment composition and
deposition in prairie pothole wetlands. Water Air and Soil Pollution, 34: 45-53.
23
-------
Moody, D. W., 1990. Groundwater contamination in the United States. Journal of Soil and
Water Conservation, 45: 170-179.
Neely, R. K. and J. L. Baker, 1989. Nitrogen and phosphorous dynamics and the fate of
agricultural runoff. In: A. G. van der Valk (ed.), Northern Prairie Wetlands. Iowa State
University Press, Ames, IA.
Novotny, V. and G. Chesters, 1989. Delivery of sediment and pollutants from nonpoint sources:
a water quality perspective. Journal of Soil and Water Conservation, 44: 568-576.
Omernik, J. M., 1977. Nonpoint source-stream nutrient level relationships: A nationwide study.
EPA-60013-77-105. U.S. Environmental Protection Agency, Corvallis, OR.
Pavelis, G. A., 1987. Economic survey of farm drainage, pp. 110-136. In: G. A. Pavelis (ed.),
Farm Drainage in the United States: History, Status, and Prospects. Miscellaneous Publication
No. 1455. Economic Research Service, U. S. Department of Agriculture.
Phillips, J. D., 1989. Fluvial sediment storage in wetlands. Water Resources Bulletin, 25:
867-873.
Ribaudo, M. O., D. Colacicco, A. Barbarika, and C. E. Young, 1989. The economic efficiency of
voluntary soil conservation programs. Journal of Soil and Water Conservation, 44: 40-43.
Sheehan, P. J., A. Baril, P. Mineau, D. K. Smith, A. Harfenist, and W. K. Marshall, 1987. Impact
of Pesticides on the Ecology of Prairie-Nesting Ducks. Technical Report Series 19. Canadian
Wildlife Service, Environment Canada, Ottawa, Ontario, Canada.
24
-------
Stuber, P. J., 1988. Proceedings of the National Symposium on Protection of Wetlands from
Agricultural Impacts. U. S. Fish and Wildlife Service, Biol. Rep. 88(16), Washington, DC.
U.S. Environmental Protection Agency, 1990. Water Quality Standards for Wetlands: National
Guidance. EPA 440/S-90-011. Office of Water Regulations and Standards, Environmental
Protection Agency, Washington, DC.
van der Valk, A. G., S. D. Swanson, and R. F. Nuss, 1983. The response of plant species to
burial in three types of Alaskan wetlands. Canadian Journal of Botany, 61: 1150-1164.
Young, R. A., C. A. Onstad, D. D. Bosch, and W. P. Anderson, 1987. AGNPS: a nonpoint source
pollution model for evaluating agricultural watersheds. Journal of Soil and Water Conservation,
44: 168-173.
25
-------
FIGURES
Figure 1. Wetlands drainage in the United States, 1980. Each dot represents 20,000 acres.
Figure taken from Pavelis (1987).
Figure 2. Siting of wetlands in a watershed. Basal model: one large wetland placed at the lower
reaches of the watershed so that all water leaving the watershed passes through it.
Figure 3, Siting of wetland in a watershed. Distributed model: a wetland placed at the lower
reaches of each subwatershed. This distribution pattern reduces the overall movement of water
and contaminants within the watershed.
Figure 4. Typical project design for restoring wetlands in areas with drainage-tile networks.
Figure 5. Siting of wetlands in a watershed when farm boundaries are considered. The basal
model minimizes the number of property owners who will be affected in a watershed.
Figure 6. Siting of wetlands in a watershed when farm boundaries are considered. The
distributed model maximizes the number of property owners who will be affected.
26
-------
Figure 1
-------
-------
Figure 3
-------
Figure 4
-------
Figure 5
-------
Figure 6
-------
RESEARCH AND INFORMATION NEEDS RELATED TO NONPOINT
SOURCE POLLUTION AND WETLANDS IN THE WATERSHED;
AN EPA PERSPECTIVE
Beverly J. Ethridge (6E-FT)
U.S. Environmental Protection Agency
1445 Ross Avenue, Suite 1200
Dallas, Texas 75202-2733
and
Richard K. Olson
ManTech Environmental Technology, Inc.
U.S. EPA Environmental Research Laboratory
200 SW 35th Street
Corvallis, Oregon 97333
ABSTRACT
Two related Environmental Protection Agency (EPA) efforts, wetlands protection and nonpoint
source pollution control, fail to fully consider landscape factors when making site-specific decisions.
This paper discusses the relationship of the two programs and the use of created and natural wetlands
to treat nonpoint source (NPS) pollution. Recommendations to improve the programs include
increased technical transfer of existing information, and more research on construction methods and
siting of created wetlands to effectively manage NPS pollution. Additional research is also needed
to determine (1) the maximum pollutant loading rates to assure the biological integrity of wetlands,
(2) the effectiveness of current land use practices in protecting habitat and water quality functions,
(3) wetland functions as pollutant sinks, (4) NPS pollution threats to wildlife, (5) practical watershed
models, and (6) indicators and reference sites for monitoring wetland condition. Model watershed
demonstrations, jointly implemented by the research and conservation communities, are
recommended as a means of integrating research results.
This article has been prepared with funding from the U.S. Environmental Protection Agency.
Portions of this document were prepared at the EPA Environmental Research Laboratory in
Corvallis, Oregon, through contract #68-C8-0006 to ManTech Environmental Technology, Inc. It
has been subjected to the Agency's peer and administrative review and approved for publication.
1
-------
INTRODUCTION
The U.S. Environmental Protection Agency (EPA) oversees many regulatory efforts designed to
protect U.S. surface waters, including wetlands (Fields, this volume). Some of these programs have
overlapping objectives, and efforts are made to ensure that these programs are well coordinated
during implementation. Coordination is achieved in part by emphasizing problem solving using a
holistic approach, as is embodied in watershed-, landscape- or ecoregion-based initiatives.
A watershed approach is commonly used in programs to control nonpoint source (NPS) pollution.
Through this approach, environmental agencies and individuals can efficiently target pollutant
abatement activities within the watershed. Knowledge of spatial and functional relationships
between wetlands and various land-use practices at the watershed scale helps to design integrated
NPS control strategies.
The purpose of this paper is to identify information gaps that hinder the inclusion of wetlands,
especially riparian wetlands, in NPS control strategies, and to recommend research and technology
transfer actions to help fill these gaps.
BACKGROUND
Riparian areas in the southeastern United States are typically forested areas along stream sides that
maintain the stability of the water course and buffer the parent waterbody from adjacent land uses.
Riparian areas also provide significant habitat functions.
Riparian systems (riparian, as used in this paper, generally refers to eastern riparian systems, which
are often wetlands) process and store large amounts of NPS pollutants from agricultural activities,
2
-------
particularly nutrients and sediment (Cooper et al., J986; Fail et al., 1987; Lowrance et al., 1984a,b;
Peterjohn and Correll, 1984; Schlosser and Karr, 1981; Whigham et al., 1986). Lowrance et al. (1985)
described the nutrient cycling processes of riparian areas as including (1) stabilizing sediments in
stream banks, (2) long-term storage of nutrients in woody material, (3) uptake of nutrients from
subsurface flow, and (4) denitrification. An earlier study illustrated the relationship between
riparian ecosystems and water quality (Lowrance et al., 1983). In a Georgia coastal plain, total
replacement of a riparian forest with a mix of crops similar to those grown on uplands was projected
to result in an estimated twenty-fold increase in loading of NOyN to the stream.
While many studies of riparian wetlands and water quality have been conducted on individual sites
or relatively small watersheds, there is much less information on the role of riparian areas at the
landscape scale (Johnston et al., 1990; Schlosser and Karr, 1981; Kuenzler, 1989). Information on
the impacts of NPS pollution on the ecosystem health and habitat functions of wetlands is also limited
(van der Valk and Jolly, this volume).
Research to date has primarily addressed pollutant processing within existing riparian areas or
pollutant fate within the landscape if riparian vegetation is removed. The extent to which structural
restoration of riparian areas restores water quality functions is not well known (Kusler and Kentula,
1990). During the recent development of draft management measures for the newly reauthorized
Coastal Zone Management Act, EPA staff did not gain management approval to include restoration
of riparian systems as an enforceable technical measure to ensure water quality protection. This
decision was based on the paucity of data in the literature on the water quality benefits of restored
systems.
3
-------
CLEAN WATER ACT
Much of EPA's legislative authority to regulate surface waters comes from the Clean Water Act.
In accordance with the 1987 Amendments to the Clean Water Act (Section 319), activities resulting
in significant NPS pollution must be identified and addressed through source reduction by
application of best management practices (BMPs). In its 1989 Nonpoint Sources: Agenda for the
Future (U.S. EPA, 1989), EPA indicated that agricultural activities account for more than half of all
NPS impacts to surface waters, significantly impairing the quality and beneficial uses of the Nation's
lakes, streams, rivers, and estuaries. These conclusions were based on the States' biennial water
quality inventories, also know as Section 305(b) reports. Section 319 provides for funding State
efforts to reduce water quality impacts. Sections 314 (clean lakes) and 320 (National Estuary
Program) also place high priority on abating adverse effects from NPS pollution.
A particularly significant component of the Clean Water Act, Section 303, requires the establishment
of total maximum pollutant loads within watersheds that will ensure protection and propagation of
indigenous fish and wildlife resources. This section requires that all sources of pollution, including
both point and nonpoint, be addressed.
Implementation of each of these Clean Water Act sections requires a watershed-based approach to
problem solving. Even in the administration of Section 404 (dredge and fill permits), which has
traditionally been applied to wetlands by EPA and the Corps of Engineers (COE) on a site-by-site
basis, more emphasis is now being placed on cumulative impacts, or, more precisely, on the impacts
to wetland systems within a watershed or landscape context.
The obvious relationship between wetlands and NPS issues led EPA to issue program guidance
encouraging the linkage of NPS and wetland program objectives by joint implementation of projects
4
-------
(U.S. EPA, 1990). This has increased the need for technical guidance on the relationship between
NPS pollution and wetlands in the many watershed improvement projects being initiated throughout
the United States by the EPA, U.S. Department of Agriculture (USDA), Tennessee Valley Authority
(TVA), COE, and State and local agencies (see Whitaker and Terrell, this volume).
An example of the need for coordination between NPS and wetlands programs is described by
Kuenzler (1989). Small streams in a watershed usually comprise most of the total stream length.
Their riparian systems, therefore, provide the watershed's longest cumulative lengths of buffer
between the stream system and upland sources of NPS pollution. However, COE regulations allow
deposition of fill material in isolated and headwater wetlands with minimal review, which places
significant portions of these riparian wetlands at risk. The failure of wetland protection programs
to adequately protect riparian wetlands along low-order streams makes it more difficult for NPS
control programs to meet water quality objectives.
IDENTIFYING RESEARCH NEEDS
Regulatory agencies need additional information and research to effectively combine wetlands
protection and NPS control strategies. As a step toward defining these needs, regional program
managers and scientists from EPA as well as other Federal, State, and local agencies, and members
of the research community were surveyed. Inquiries were sent to people currently involved in NPS
program implementation, wetland protection program implementation, and related research.
Approximately 35 individuals responded to the survey, raising over 100 research and technology
transfer issues, many of which were closely related and have been consolidated and summarized.
Five main areas were identified where additional technical guidance and research is needed:
(1) relationships between wetland water quality and habitat functions, (2) wetland water quality
5
-------
functions at the landscape scale, (3) monitoring and evaluation techniques, (4) design criteria for
constructed wetland treatment systems, and (5) technology transfer. The most common research
needs within each category are listed below.
WATER QUALITY AND HABITAT FUNCTIONS
• Determination of thresholds (i.e., pollutant loading rates and cumulative loadings) above which
wetland structure and functions are degraded.
• Differences in loading thresholds for different wetland types and geographic regions.
• Long-term sustainability of wetland water quality functions.
• Recovery of water quality functions following perturbations such as severe flood events.
• Indirect effects of NPS pollution on the health of migratory waterfowl and wading bird
populations (e.g., greater vulnerability to parasites and disease).
LANDSCAPE FUNCTIONS
• Methods for determining pollutant loading rates to wetlands based on upland acreage, land
uses, and spatial configuration.
• Landscape design: Should restoration of wetlands within the watershed attempt to duplicate
historical wetland patterns?
• Site location criteria: Are wetlands near the pollutant source most critical, or are larger systems,
strategically located downstream, more important in the water quality relationship?
• Relationships between watershed hydrology, biogeochemistry, and wetland functions.
• Integration of wetlands with other BMPs for controlling NPS pollution.
• Models of landscape function for predicting water quality effects of different wetland
protection and restoration scenarios.
MONITORING AND EVALUATION TECHNIQUES
• Indicators of wetland function and condition; needed at scales from site to regional, and for
a variety of stressors including chemical, hydrological, and physical perturbations.
• Methods for monitoring the fate and effects of NPS pollutants within a watershed.
• Methods for selecting reference sites for use in monitoring programs.
6
-------
• Procedures for evaluating the effectiveness of BMPs at improving water quality and protecting
wetlands from degradation and NPS pollution.
• Post-restoration monitoring to help improve wetland restoration procedures.
DESIGN CRITERIA FOR CONSTRUCTED WETLAND TREATMENT SYSTEMS
• Improved design and operational criteria for animal waste treatment systems.
• Improved design and operational criteria for field runoff treatment systems.
• Improved capabilities for predicting treatment system longevity and maintenance needs.
TECHNOLOGY TRANSFER
• Synthesis of research results from the literature, and transmittal to program managers in the
form of clear and comprehensive guidance on integrating wetland and NPS regulatory programs.
• Additional watershed-level demonstrations of integrated NPS/wetland programs.
• Economic information relating the relative cost and benefits of different BMPs and wetlands
protection strategies relative to water quality improvement.
• Improved collaboration and information exchange among Federal, State, and private groups.
SUMMARY
The range and number of research needs identified from the survey make clear the great need for
research and technical guidance relating wetlands and NPS pollution. An opinion often expressed
is that watershed-scale water quality improvement projects are being funded and implemented, yet
these projects are proceeding without adequate guidelines from the Federal community. Program
implementers must also work with farmers who are understandably reluctant to invest monies on
efforts that lack clear, definable benefits. It is critical that farmers, other landowners, and
researchers all be included as members of the committees that oversee research and development of
technical guidance.
7
-------
It is well documented that wetlands and riparian systems perform a significant role in protecting
and maintaining water quality within the landscape. Considerable information is available showing
that wetlands improve water quality by filtering and processing pollutants. However, that function
has not been adequately described in holistic, watershed terms that also address other natural
functions of wetlands, particularly the relationship between water quality functions and habitat
quality.
These information gaps leave field program implementers (both NPS and wetland protection program
implemented) with little guidance on how to implement wetland/riparian restoration efforts on a
watershed scale. The gaps need to be filled by research, but just as importantly, the research results
must be packaged and distributed in formats that can be easily used by land managers.
RECOMMENDATIONS
• EPA should take a more active role in technology transfer. Many of the "research" questions
raised may, in fact, be answered to some extent in the literature, but the information is not
available to program personnel in a useful form. An interagency team, including other Federal
agencies, as well as State, local, and nongovernmental groups, should be established to determine
the most effective means of synthesizing and disseminating information.
• Collaborative watershed demonstration projects between EPA and other groups should be
implemented in multiple regions of the United States. Greater EPA participation in the USDA
Hydrologic Unit Areas program and other USDA water quality improvement demonstration
projects (see Whitaker and Terrell, this volume) is one possible approach. These projects would
provide a means of integrating a number of research tasks including:
- developing models that define the functional relationships between wetlands and other
land uses in the watershed;
- developing models for predicting the effects of different BMPs and landuse patterns on
the fate and transport of NPS pollutants;
- developing indicators (biological, chemical, etc.) of wetland condition, and identifying
reference wetlands.
8
-------
• USDA and TVA should continue their leadership roles in developing improved design criteria
for agricultural wastewater constructed wetlands. EPA should assist by assessing the ecological
condition and effects of these systems.
These recommendations are steps toward filling some of the information gaps that are hindering
the effective coordination of NPS control and wetlands protection programs. A strong link between
science and policy is necessary if the multiple objectives of these two programs are to be met.
9
-------
REFERENCES
Cooper, J.R., J.W. Gilliam, and T.C. Jacobs, 1986. Riparian areas as control of nonpoint pollutants,
pp. 166-192. In: D.L. Correll (ed.), Watershed Research Perspectives. Smithsonian Institution Press,
Washington, DC.
Fail, J.L., Jr., B.L. Haines, and R.L. Todd, 1987. Riparian forest communities and their role in
nutrient conservation in an agricultural watershed. American Journal of Alternative Agriculture,
2(3): 114-121.
Johnston, C.A., N.E. Detenbeck, and G.J. Niemi, 1990. The cumulative effects of wetlands on
stream water quality and quantity. A landscape approach. Biogeochemistry, 10: 105-141.
Kuenzler, Edward J., 19S9. Value of forested wetlands as filters for sediments and nutrients, pp.
85-96. In: Proceedings of Symposium on the Forested Wetlands of the Southern U.S., Orlando, FL.
Kusler, Jon A. and Mary E. Kentula (eds.), 1990. Wetland Creation and Restoration: The Status of
the Science. Island Press, Washington, DC. 591 pp.
Lowrance, R.R., R.L. Todd, and L.E. Asmussen, 1983. Waterborne nutrient budgets for the riparian
zone of an agricultural watershed. Agriculture, Ecosystems and Environment, 10: 371-384.
Lowrance, R.R., R.L. Todd, and L.E. Asmussen, 1984a. Nutrient cycling in an agricultural
watershed: II. Streamflow and artificial drainage. Journal of Environmental Quality, 13(1): 27-
32.
10
-------
Lowrance, R., R. Todd, J. Fail, Jr., O. Hendrickson, Jr., R. Leonard, and L. Asmussen, 1984b.
Riparian forests as nutrient filters in agricultural watersheds. Bioscience, 34(6): 374-377.
Lowrance, R., R. Leonard, and J. Sheridan, 1985. Managing riparian ecosystems to control nonpoint
pollution. Journal of Soil and Water Conservation, 40(1): 87-91.
Peterjohn, W.T. and D.L. Correll, 1984. Nutrient dynamics in an agricultural watershed:
Observations on the role of a riparian forest. Ecology, 65(5): 1466-1475.
Schlosser, I.J. and J.R. Karr, 1981. Water quality in agricultural watersheds: Impact of riparian
vegetation during base flow. Water Resources Bulletin, April: 233-240.
U.S. Environmental Protection Agency, 1989. Nonpoint Sources: Agenda for the Future. WH-
556, U.S. EPA, Office of Water, Washington, DC. 31 pp.
U.S. Environmental Protection Agency, 1990. National Guidance: Wetlands and Nonpoint Source
Control Programs. U.S. EPA, Office of Water Regulations and Standards and Office of Wetlands
Protection, Washington, DC.
Whigham, D.F., C. Chitterling, B. Palmer, and J. O'Neill, 1986. Modification of runoff from upland
watersheds—the Influence of a diverse riparian ecosystem, pp. 305-332. In: D.L. Correll (ed.),
Watershed Research Perspectives. Smithsonian Institution Press, Washington DC.
11
-------
FEDERAL PROGRAMS FOR WETLAND RESTORATION AND USE OF WETLANDS
FOR NONPOINT SOURCE POLLUTION CONTROL
Gene Whitaker
U. S. Fish and Wildlife Service
Division of Habitat Conservation
440 N. Fairfax Drive, Suite 400
Arlington, Virginia 22003
and
Charles R. Terrell
Ecological Sciences Division
Soil Conservation Service
Washington, D.C. 20013
ABSTRACT
A review of Federal wetlands programs shows that a number of agencies have significant wetland
restoration and creation efforts. Water quality improvement is not the main objective of most of
these programs, and areas with high nonpoint source (NPS) pollution may actually be avoided in
order to protect wetland values such as habitat. However, ancillary water quality benefits are
provided by many created and restored wetlands, and agencies such as the U.S. Department of
Agriculture are actively evaluating the use of created and restored wetlands as components of NPS
control strategies.
I
-------
INTRODUCTION
Wetland restoration has come into widespread use in the last few years. All Federal land
management agencies now have active programs to restore wetlands on lands under their control,
and most of them have programs to assist other agencies and private landowners restore wetlands.
In addition, several U. S. Department of Agriculture (USDA) agencies and the Department of the
Interior (DOI) Fish and Wildlife Service (FWS) are implementing specifically mandated programs
to help private landowners enhance, restore, and create wetlands. All of these agency programs
contribute to the control of nonpoint source (NPS) pollution. However, control of NPS pollution
is only one objective of most of these programs, and, in most cases, it is not a major objective.
This paper briefly reviews the goals and objectives of the restoration programs of the major
Federal agencies and what they are doing (see U. S. EPA (1989) for additional details).
In all these agency programs, the definition between wetland restoration and wetland
enhancement is blurred. Restoration can be anything from blocking a drainage ditch that only
removed a small fraction of the water from a wetland to major efforts that restore wetlands
drained many years ago. The objective of most restorations is to recover the original wetland
type, size, value, and vegetative community that existed before man's activities. Wetland
enhancement refers only to the restoration of partially damaged wetlands. However,
enhancement may be improving the value of a wetland for a function considered by the planners
to be more important. The following definitions of these terms recently adopted by the Soil
Conservation Service (SCS) are recommended for use (USDA, 1991a):
Wetland restoration is defined as the rehabilitation of a degraded existing wetland or a
hydric soil area that was previously a wetland.
2
-------
Wetland enhancement is defined as the improvement, maintenance, and management of
existing wetlands for a particular purpose or function, often at the expense of others.
Wetland creation is defined as the conversion of a non-wetland area into a wetland where a
wetland never existed.
Constructed wetlands are specifically designed to treat both nonpoint and point sources of
water pollution.
SUMMARY OF FEDERAL WETLAND PROGRAMS
FEDERAL LAND MANAGEMENT AGENCIES
Eight Federal agencies manage most of the wetlands and sites for wetland restorations owned by
the Federal government. Each has its own programs for restoring and enhancing wetlands. Most
importantly, each has its own legislatively mandated missions and wetland goals and objectives.
All the land management agencies emphasize the development of land management plans,
including wetland restorations, with a high level of local public input. They are responsive to the
needs and desires of the local communities.
Department of Defense
The Army, Navy, Air Force, and Coast Guard control approximately 25 million acres containing
many acres of wetlands and sites where wetlands could be restored. As compatible with their
primary mission, the land management plans for each installation address wetland management
and may include wetland restorations in cooperation with other State and Federal agencies. On
many bases, wetlands are restored in cooperation with the Fish and Wildlife Service to improve
3
-------
habitat for migratory birds. On many bases, the restoration and conservation of rare and unique
wetland ecosystems is emphasized.
U. S. Armv Coros of Engineers
The Corps manages 11 million acres of Federal land. Management of these lands, including
wetlands and wetland restorations,"... is directed toward the continued enjoyment and
maximum sustained use of public lands, waters, forests, and associated recreational resources
consistent with their aesthetic and biological values. . ." Most of the Corps' wetland restorations
are aimed at replacing the wetland functions and values lost during the construction of major
projects. They also have an active wetland restoration research program and provide training to
other agencies on wetland restoration.
USD A Forest Service
The estimated 9 million acres of wetlands on the 190 million acres of national forest land are
managed and restored for their multiple-use values. They have a very active program to restore
riparian wetlands providing significant control of NPS pollution.
DPI Bureau of Reclamation
Wetlands are managed and restored on the 8,5 million acres of land managed by the Bureau of
Reclamation (BOR) as an integral part of total water resources management. Wetland
conservation focuses on ". .. fish and wildlife habitat with equal consideration given to functions
such as sediment control, water and wastewater treatment, flood storage, and ground water
recharge." Most of BOR's wetland restoration activity is aimed at replacing wetland functions
and values lost as a result of their irrigation water delivery systems. BOR staff are actively
researching methods to minimize the effects of irrigation return water on wetlands and ways to
minimize adverse impacts. They are also researching wetland construction techniques for various
4
-------
purposes and are preparing a wetland manual with scientific and engineering guidelines and
procedures related to wetlands restoration and construction.
DPI National Park Service
Wetlands in units of the national park system are managed, protected, and restored to maintain
their original natural characteristics to the fullest extent possible. To accomplish this, the
National Park Service has an active water resource program designed to provide adequate water
quantity and water quality for protecting park wetlands. The quality of water entering the parks
is their single greatest problem.
DPI Bureau of Land Management
The Bureau of Land Management (BLM) manages 178 million acres of land in the lower 48 states
containing about 1 million acres of marshes, ponds, reservoirs, and lakes and 41,000 miles of
streamside riparian wetlands. BLM has an active program to restore and manage wetlands and
riparian areas on their lands, mainly in the arid west. BLM states that "The objective of riparian
area management is to maintain, restore, or improve riparian values to achieve a healthy and
productive ecological condition for maximum long-term benefits." A major objective of many of
their riparian area management plans is to control NPS pollution by reducing the delivery of
sediment to downstream reservoirs.
DPI Fish and Wildlife Service
The National Wildlife Refuge System has over 90 million acres of land in 462 refuges. The
largest portion is in Alaska. However, over 25 million acres are in the lower 48 states. Most of
the wetlands on refuges in the lower 48 states have been restored or enhanced during the last 20
years, primarily to improve habitat for waterfowl and other migratory birds. Active programs
are underway in the major waterfowl breeding, migratory, and wintering areas to acquire and
5
-------
restore wetlands. Although control of NPS pollution is not a major purpose of wetland
restorations on refuges, many refuge wetlands are adversely impacted by NPS pollution.
Sediment in incoming water is shortening the effective life of many refuge wetlands, and
contaminants, especially in irrigated areas, are making some refuge wetlands unusable.
FEDERAL AGENCIES WITH PRIVATE LANDS PROGRAMS
Several Federal agencies have programs targeted to protect, enhance, and restore wetlands on
private lands. The following sections list the Federal agencies with major programs and a brief
discussion of each program and its goals and objectives (see U.S. DOI (1991) for additional details
on the programs of Interior bureaus). The U.S. Environmental Protection Agency (EPA) has
several programs, mainly working through State agencies, to restore wetlands. However, these are
not discussed in this paper.
DOI Fish and Wildlife Service
The Fish and Wildlife Service (FWS), through its Private Lands Program, uses numerous avenues
to effect the restoration and protection of wetlands on private lands throughout the country.
Although restoration of wildlife habitat, especially for migratory birds, is a major objective in all
the activities, emphasis is placed on the restoration and protection of the wide array of wetland
functional values. However, the primary purpose of the restorations is to provide the maximum
benefits for the longest possible time to wildlife and the people that enjoy them. Restoration
sites are avoided where there is significant NPS pollution or other contaminant sources in the
watershed, or measures are taken to avoid adverse impacts. Over the last 3 years, a total of more
than 90,000 acres of mostly small wetlands have been restored, using funds appropriated for this
purpose.
6
-------
A concentrated effort is being made under the North American Waterfowl Management Plan
(NAWMP) to protect a minimum of 2 million acres of existing wetland habitat in the United
States by the year 2000. Under the plan, several million additional acres of wetlands will be
restored and enhanced for migratory birds in Canada, Mexico, and the United States. The
primary goal of the NAWMP is "To enhance and protect high-quality wetland habitat in North
America that supports a variety of wetland-dependent wildlife and recreational uses." The
NAWMP is being implemented through innovative Federal-State-private partnerships within and
between States and Provinces.
DO! Bureau of Mines
The Bureau of Mines actively supports the construction of wetlands to help treat acidic mine
water and regularly presents workshops on wetland construction for acid mine drainage
treatment. They have developed design criteria for sizing wetlands based on the volume of acid
mine drainage.
DPI Office of Surface Mining
The Office of Surface Mining strongly encourages and provides guidance on the development of
wetlands for the broad range of natural functional values as part of all surface mine reclamations.
They also have an active constructed wetlands research program to develop better
recommendations.
USDA Cooperative Extension Service
The national office provides leadership, encouraging each State extension service to promote and
provide assistance to private landowners for the management and restoration of wetlands. Most
State extension services do have active programs promoting wetland conservation and provide
information and technical assistance to landowners.
7
-------
USDA Agricultural Stabilization and Conservation Service
Under the Cropland Reserve Program (CRP) of USDA1s Agricultural Stabilization and
Conservation Service (ASCS), approximately 950,000 acres of cropped wetlands and associated
uplands have been re-established in natural vegetation under 10-year contracts. In addition, a
couple million acres of natural and partially drained wetlands that were cropped when dry
enough are included in fields entered in the CRP under the highly erodible criteria, generally
because of wind erosion. More than 13,000 acres of wetlands have been restored under the CRP,
in addition to wetlands the FWS has worked with farmers to restore on lands in the program,
mainly for their wildlife and water quality benefits.
The Waterbank Program administered by the ASCS is presently protecting, in agricultural areas,
480,000 acres of natural wetlands and adjacent buffer areas under 10-year rental agreements.
The Wetland Reserve Program (WRP), authorized by the 1990 Farm Bill, is scheduled for
implementation starting early in 1991, pending authorization of funds. The objective of the WRP
is to restore and protect, through easements, up to 1 million acres of wetlands in cropland on the
Nation's farms and ranches. The law requires that priority shall be placed "... on acquiring
easements based on the value of the easement for protecting and enhancing habitat for migratory
birds and other wildlife." Technical assistance for the WRP is being provided by the FWS and the
SCS. The emphasis will be on restoring wetlands to natural conditions for the multiplicity of
functions and values provided. Areas with a high level of NPS pollution will be avoided to
prolong the useful life of the wetlands and protect wildlife and other functional values of the
restored wetlands.
8
-------
Under the Agricultural Conservation Program (ACP), ASCS will cost share with farmers up to 75
percent of the cost of numerous practices that help control NPS pollution. For the "Creation of
Shallow Water Areas" (wetland restoration) practice, cost share has been provided for the
restoration of 555,000 acres of wetlands over the last 30 years. A new practice, eligible for up to
75 percent cost share and titled "Constructed Wetlands for Agricultural Waste Water Treatment
(WP6)," is being designed to encourage farmers to use artificially created wetlands to control
pollution by animal wastes. These constructed wetlands will be specifically designed as waste
treatment systems that possess wetland characteristics. Their primary purpose will be to control
NPS pollution. However, they will have some wetland wildlife value and provide some of the
other functional values of natural wetlands. The technical requirements for their design will be
supplied by the SCS, and SCS specialists will do the actual designs and supervise construction.
Monitoring will be required for each constructed wetland installed.
USDA Soil Conservation Service
The SCS, with offices covering every county in the country, provides technical assistance to
private landowners for wetland restoration. SCS provides detailed training to their field
personnel in the planning and design of wetland restorations. SCS also helps develop the plans
for some of the FWS's restorations and helps train FWS field people in the engineering aspects of
designing wetland restorations. Working cooperatively with the FWS, Corps of Engineers, and
EPA, SCS has just completed a detailed manual for the restoration, enhancement, and creation of
wetlands (USDA, 1991a).
INTERAGENCY USDA ACTIVITIES TO CONTROL NPS POLLUTION
The SCS with the Extension Service and the ASCS has been preparing to provide increased
technical assistance for agricultural water quality problems. In February 1990, SCS issued its
9
-------
"Water Quality and Quantity Five-year Plan of Operations" (USDA, 1990a), which was followed
later that year by the USDA "Policy for Water Quality Protection (USDA, 1990b)."
Hvdroloeic Unit Areas (HUAs>
In 1990, 37 HUAs were selected across the country where farmers and ranchers can participate in
correcting water quality problems in their agricultural operations. In 1991, 37 additional HUAs
were added to the list for a total of 74. Hydrologic Unit Areas are located in agricultural
watersheds where the goal is to provide increased assistance to farmers and ranchers in
voluntarily applying conservation practices to improve the water quality of the area (USDA,
1991b).
In each area, cost-sharing provides agricultural operators with the incentive to install
conservation practices, such as animal waste control facilities, grassed waterways, irrigation water
management systems, or integrated crop management for water quality improvement. Cost-
share funds are from the ASCS and State programs. SCS provides technical assistance to farmers
and ranchers in the HUAs and the Extension Service provides information and educational
assistance, including specific recommendations on the use of nutrients and pesticides (USDA,
1990c).
Demonstration Projects
In 1990, eight demonstration projects were approved across the country with an additional eight
added in 1991. These projects demonstrate new ways to minimize the effects of agricultural NPS
pollution, including the effects of nutrients and pesticides on groundwater. The goals are to
demonstrate cost-effective water quality-oriented practices that can be used and shared by
farmers and ranchers. Also, the projects are to accelerate the adoption of newly developed water
10
-------
quality technology. These projects are implemented under the joint leadership of SCS and the
Extension Service, with financial assistance provided by ASCS.
ACP Water Quality Special Projects
In this program, funds are reserved by ASCS at the national level to fund Water Quality Special
Projects developed by county Agricultural Stabilization and Conservation Committees. Project
emphasis is on improving the quality of surface and ground waters that are impaired by
agricultural NPS pollution. The projects are administered by ASCS with educational and
technical assistance from the Extension Service and SCS.
It is obvious that overall integration between SCS, ASCS, and the Extension Service is necessary
to accomplish the needed goals of agricultural water quality improvement. To that end,
flexibility is built into the above three project types. For example, ACP Special Water Quality
Projects can be used to solve water quality problems identified in HUAs and demonstration
projects and also those identified locally that may provide significant public benefits to
nonagricultural interests. These projects may be designed to support State NPS objectives
developed to meet the requirements of Section 319 of the Water Quality Act (see Fields, this
volume).
SCS CONSTRUCTED WETLANDS PROGRAMS
The term "constructed wetland" is one that has come into use in recent years. Constructed
wetlands, as we use the term, are constructed on non-wetland sites for the specific purpose of
water quality improvement.
11
-------
This definition distinguishes constructed wetlands from restored wetlands, which are "restored" or
"recreated" where they once had existed. Created wetlands generally refer to wetlands developed
where none previously existed to provide a variety of functional values.
This definition of constructed wetlands does not distinguish between point and nonpoint source
pollution treatment. Constructed wetlands are used for treating both point and nonpoint sources
of pollution. To a constructed wetland, the source of pollution does not make much difference.
The wetland will do the job of cleaning water no matter what the source. Also, the definition
does not mention surface and ground waters for similar reasons; water, whether above or below
ground, really is one resource.
BASIC DESIGN CONSIDERATIONS
This paper will touch on this subject briefly because other papers in this volume (Hammer,
Mitsch) and Cooper and Findlater (1990) cover the design and construction of constructed
wetlands in detail. There are four basic considerations relative to a constructed wetland and its
functioning: (1) Construction, (2) Water, (3) Plants, and (4) Water Quality Improvement. Soils
can be part of the equation or they can be left out entirely. For example, a constructed wetland
can be lined with a synthetic liner or built in a concrete trough. Water hyacinths with their
floating aspect never need to touch the soil.
Construction design and water considerations are key to a successful constructed wetland.
Constructed wetland design is tied to water height; it is difficult to separate the two components.
Water height, water residence time in the wetland, and exit of water from the wetland are
critically important engineering considerations. The specifics of these characteristics must be
developed if the constructed wetland is to have the desired functions.
12
-------
Plants, the third component, need some examination. A misconception is that constructed
wetlands in cold climates do not work. However, the evidence does not confirm that viewpoint.
While it is true that photosynthesis and other plant functions are reduced in cold climates,
wetland plants even below the snow or ice keep working, if at a reduced level. Actually, it
appears that snow and ice act as an insulating blanket, so vital plant functions continue. Sunlight
can penetrate ice to allow some photosynthesis. Further, it should be remembered that, in
agricultural situations, water quality cleanup is needed largely in spring and summer and into
autumn when organic production is at its peak and wetland plants are also very active.
Finally, the type of water quality improvement needed is vitally important and must be
thoroughly understood before a shovel of soil is moved. The major decision to be made is
whether the constructed wetland is for protection or a cure. Will the constructed wetland act in
concert with other conservation practices to achieve an overall water quality improvement, or will
the wetland serve as a device of last resort when other conservation practices are unable to
achieve the desired goals of water quality improvement?
LANDSCAPE CONSIDERATIONS
It is possible to design for the four characteristics discussed above; but additionally a constructed
wetland should fit the landscape. Does a constructed wetland need to be a perfect square or a
rectangle? Consider the shapes of wetlands that Nature has provided. Is the constructed wetland
to be located in the Midwest where ox-bows from old river bends are common? Why not create
the new wetland in a similar shape? Is the wetland going to be located in Appalachia, where
drowned river valleys are long and narrow? How about that shape? Rather than being bound by
tradition that may dictate a so-called "standard" shape for a constructed wetland, today's
designers should consider how to make the wetland an integral part of the land, not just another
geometric configuration.
13
-------
Regarding siting, some believe that constructed wetlands can be located at the bottom of the
farmer's land, just before field runoff water enters the stream, river, or pond. Basically, this
makes the wetland a structure of last resort and may lessen the perceived need to address water
quality on a field-by-field basis. Without other measures, the constructed wetland would offer
the only chance for cleanup before the runoff merges with receiving waters. It offers the
tempting possibility to the farmer or rancher that other conservation practices need not be
employed: "The constructed wetland will take care of my nonpoint source problems," he says.
An analogy can be made with a sewerage treatment plant. What happens with human waste is
that clean water is intentionally contaminated with human waste and sent through a pipe; the
contaminated water is intercepted at a sewerage treatment plant, often with the intention of
cleaning the water just before that water empties into receiving waters. Thus, the sewerage
treatment plant is a device of last resort. While alternative methods could be used that would not
contaminate the water in the first place, they often are not. In some cases, millions of additional
dollars are spent to protect the receiving waters because those same waters are used by the next
town downstream for drinking purposes. Therefore, we must be absolutely certain that the
sewerage treatment plant works 100 percent of the time or we endanger ourselves and our
neighbors because of contaminated water.
As with the treatment plant, a constructed wetland located at the bottom of the farmer's land or
at the bottom of the watershed becomes a structure of last resort. The wetland must work 100
percent of the time in all kinds of weather and under all conditions. This is an overly optimistic
expectation. As with soil erosion protection, water quality protection must be implemented on a
field-by-field basis. This practice reduces the pollution load at the edges of fields and does not
transfer an increasingly larger load to the bottom of the watershed where one constructed wetland
might be the only means of purifying the water before it empties into receiving waters.
-------
ONGOING ACTIVITIES
Presently, the SCS Plant Material Centers are conducting activities that enhance or restore
existing wetlands or help establish constructed wetlands. These activities include: (1) developing
propagation, establishment, and management techniques for potentially useful native wetland
plants; (2) developing commercial sources of native wetland plants; (3) identifying the best plants
and methodologies for using wetland plants for improving water quality associated with
agricultural operations; (4) using wetland plants for drawdown areas of large impoundments; and
(5) restoring coastal marshes and wetland riparian areas. These activities are planned or being
conducted in Mississippi, Georgia, New Jersey, Michigan, Kansas, Oregon, Idaho, New York,
and Texas. Most of this work involves cooperative projects with other Federal and State
agencies.
In Mississippi and Alabama, SCS is working with the Tennessee Valley Authority and other
agencies to see how constructed wetlands can be used to treat animal waste problems associated
with livestock and poultry (Hammer, in this volume, and Hammer, 1989). Generally, in these
types of situations the constructed wetland is used in combination with other structures, such as
anaerobic and aerobic lagoons. The wetland acts as a polishing filter because the wetland would
not be able to receive animal wastes directly.
In Maine, seven "Nutrient/Sediment Control Systems" have been constructed. Each system
combines a series of conservation practices, including a constructed wetland that minimizes NPS
pollution leaving the farms. The objective is to intercept cropland runoff from potato fields in
the northernmost county of the lower 48 states, treat that runoff, and release clean water to a lake
or stream. The basic system consists of a sediment basin, level-lip spreader, primary grass filter,
constructed wetland, pond, polishing grass filter, and outlet (see Hammer, this volume, for
IS
-------
details). Typical results show total phosphorus and suspended solids reduced by more than 90
percent, even during intense storm events (Wengrzynek and Terrell, 1990).
In Maryland, SCS is working with the State Department of Natural Resources on using
constructed wetlands to control urban NPS runoff problems. These shallow marshes are created
to remove pollutants, increase wildlife habitat, and provide educational and recreational
opportunities. Maryland has designed and built more than SO of these marshes across the state.
Monitoring data show that total suspended solids were reduced by 63 percent; nitrate, 40 percent;
total nitrogen, 12 percent; and total phosphorus, 38 percent.
In Pennsylvania, SCS has been working with the DOI Office of Surface Mining and the Bureau of
Mines in the construction of new wetlands on abandoned coal mine lands. Several new wetlands
have been constructed in the Pittsburgh-Sumerset area. These constructed wetlands were built on
non-wetland soils to improve the quality of the water coming from abandoned coal mines.
Monitoring data show that 90 percent of some pollutants such as iron are removed. The Bureau
of Mines, which is the research arm of the 1977 Abandoned Mine Lands Act, maintains
monitoring stations on these lands. Similar activity is ongoing in Kentucky and West Virginia.
CONCLUSIONS
In conclusion, we would like to point out several cautions and considerations to researchers and
those charged with developing and implementing programs using constructed wetlands to control
NPS pollution.
16
-------
SINGLE-PURPOSE VS. MULTIPLE-PURPOSE WETLANDS
Wetlands serve many valuable purposes to society; assimilation of various pollutants is just one of
them. As this paper has discussed, most Federal agencies' wetland management and restoration
programs are concerned about wildlife benefits of wetlands and the multiplicity of other
functional values provided. They are concerned about the long-term maintenance of healthy
functioning wetland ecosystems. Assimilation of moderate amounts of nutrients, pesticides, and
other contaminants is a normal function of wetlands. However, in most places where there is a
recognizably serious NPS problem, the water going into or through a wetland may reduce the
other functional values of the wetland and shorten its useful life.
On the other hand, it is generally recognized that wetlands constructed specifically for the
purpose of controlling NPS pollution do provide other benefits to society. SCS has often heard
the comment, "You are constructing wetlands under the guise of water quality improvement so
there will be more wetlands; water quality is only a ruse to get more wetlands for wildlife."
While more benefits than water quality improvement can be realized from building constructed
wetlands, such as wildlife habitat enhancement, there also are other advantages to constructed
wetlands that have little to do with wildlife, such as flood control and recreation. One of the
farmers collaborating with SCS was glad to have the water quality of the runoff leaving his fields
improved, but he was even more delighted that he could raise baitfish in his constructed wetland
with which to go fishing. Different people appreciate the different functions of wetlands and
want constructed wetlands for different reasons. We should be open to those varied reasons,
respect them, and see how constructed wetlands can accommodate a variety of needs, functions,
and uses without impacting the functions and values of natural wetlands.
17
-------
SECTION 404 CONSIDERATIONS
Farmers will not invest in constructed wetlands if they fear that in a few years a new generation
of thinking might say that it looks like a wetland so it must be a wetland and therefore it falls
under 404 (see Fields, this volume, for a description of regulations under Section 404 of the
Clean Water Act). The wetland/environmental community must be able to assure farmers and
ranchers who honestly invest in a constructed wetland as a conservation practice that they will
not find themselves facing Section 404-related problems in the future.
THE FUTURE
Constructed wetlands offer an excellent potential for agriculturally related water quality
improvement, but we do not know all the answers as to how they work and how they can work
most effectively in agricultural situations. We need to foster more research and investigations
that will provide some of these answers, while at the same time allowing us to proceed with using
constructed wetlands as one strategy for controlling NPS pollution. The investigations must be
conducted not only on a scientific basis, but also on a policy basis. Even though we do not have
all of the answers to questions concerning constructed wetlands, we should continue to think of
them as one more tool in the toolbox of conservation practices, rather than as a panacea that will
cure all of our water quality problems.
18
-------
REFERENCES
Cooper, P. F. and B. C. Findlater (eds.), 1990. Constructed Wetlands in Water Pollution Control:
Proceedings of the International Conference on the Use of Constructed Wetlands in Water
Pollution Control, Cambridge, England, Sept. 24-28, 1990. Pergamon Press.
Hammer, D. A. (ed.), 1989. Constructed Wetlands for Wastewater Treatment: Municipal,
Industrial and Agricultural: Proceedings of the First International Conference on Constructed
Wetlands for Wastewater Treatment, Chattanooga, TN, June 13-17, 1988. Lewis Publishers, Inc.,
Chelsea, ML
U. S. Department of Agriculture, 1990a. Water Quality and Quantity: SCS Five-Year Plan of
Operations, October 1, 1989-September 30, 1994. SCS National Bulletin No. 460-0-1.
U. S. Department of Agriculture, 1990b. USDA Policy for Water Quality Protection. Soil
Conservation Service (SCS) National Bulletin No. 460-1-1.
U. S. Department of Agriculture, 1990c. Water Quality Education and Technical Assistance Plan:
1990 Update. Agricultural Information Bulletin No. 598.
U. S. Department of Agriculture, 1991a. Chapter 13. Wetland Restoration, Enhancement, or
Creation. SCS Engineering Field Handbook. In press. 94 pp.
U. S. Department of Agriculture, 1991b. Water Quality Activities of the Soil Conservation
Service: 1991 Update.
19
-------
U. S. Department of the Interior, 1991. Wetland Stewardship; Highlights of the Department of
the Interior's 1990 Wetlands Activities. Department of the Interior, Washington, DC. 37 pp.
U. S. Environmental Protection Agency, 1989. Wise Use and Protection of Federally Managed
Wetlands: The Federal Land Management Agency Role. Results of a Workshop. U. S.
Environmental Protection Agency, Washington, DC. 120 pp.
Wengrzynek, R. J. and C. R. Terrell, 1990, Using constructed wetlands to control agricultural
nonpoint source pollution. In: P. F. Cooper and B. C. Findlater (eds.), Constructed Wetlands in
Water Pollution Control: Proceedings of the International Conference on the Use of Constructed
Wetlands in Water Pollution Control, Cambridge, England, Sept. 24-28, 1990. Pergamon Press.
20
------- |