U.S. Environmental Protection Agency


NATURAL WETLANDS AND URBAN
STORMWATER: POTENTIAL
IMPACTS AND MANAGEMENT

February 1993

U.S. Environmental Protection Agency
Office of Wetlands, Oceans and Watersheds
Wetlands Division
Washington, D.C.

CONTENTS

  1. Introduction
    Background
    Purpose

  2. Wetland Characteristics
    Hydrology
    Water Quality/Benthic Processes
    Biologic/Habitat Functions
    Unresolved Issues

  3. Stormwater Characteristics
    Urban Activities That Affect Stormwater Characteristics
    Chemical Characteristics
    Hydrologic Characteristics

  4. Regional Differences in Stormwater Characteristics and in Wetland Types
    Regional Differences in Stormwater Characteristics
    Regional Differences in Wetland Types
    Relationship Between Regional Characteristics of Stormwater and Wetlands

  5. Potential Impacts of Urban Stormwater Runoff on Natural Wetlands
    Hydrologic Changes
    Water Quality Changes
    Wetland Soil Changes
    Biologic/Habitat Impacts
    Regional Differences
    Impoundments
    Unresolved Issues
    Unresolved Impoundment Issues

  6. Stormwater Management Practices and Natural Wetlands
    Federal and State Stormwater Management Programs
    Control of Adverse Impacts
    Conclusions
    Unresolved Issues

  7. Summary
    Wetland Functions
    Impacts of Stormwater Discharges
    Management of Stormwater Discharges
    Literature Cited
    Glossary

TABLES

  1. Wetland Types

  2. Examples of Pollutant Characteristics Found in Stormwater Runoff From Various Land Uses in the Great Lakes Region

  3. Ranges in Pollutant Concentrations Found in Urban Runoff

  4. Sources of Urban Runoff Pollutants

  5. Locations of Wetland Types in the United States

  6. Relationship of Wetland Type to Its Origin, Hydrology, Soils, and Vegetation

  7. Wetlands Present in SCS Type Rainfall Distribution Areas

  8. Comparison of Stormwater Runoff Quality

  9. Summary of Mean Soil and Sediment Chemistry Data as a Function of Sampling Location, December 1978

  10. Distribution of Selected Constituents in Water, Sediments, and Groundwater at the Silver Star Road Study Area

  11. Median Values of Selected Constituents in the Water Column and Values for One Sample of Bed Sediments at the Island Lake Wetland

FIGURES - Not available on this page

  1. Approximate geographic areas for SCS rainfall distributions

  2. Hourly fraction of total rainfall within a 24-hour period for each rainfall distribution type

  3. Major climatic regions of North America

  4. Average number of days each year on which thunderstorms are observed throughout the United States

  5. Month-to-month variation of precipitation in the United States

  6. Average annual runoff in the United States

  7. Physiographic regions of the United States

  8. Oxygen fluctuations in a shallow water impoundment in Minnesota

  9. Monthly distribution of fishes in Impoundment No. 12 and pond water levels during 1979

* - To view the Figures, please call the EPA Wetlands Hotline @ (1-800-832-7828 to order a copy of manual.

1. INTRODUCTION

The U.S. Environmental Protection Agency (EPA) is in the process of developing and implementing programs to reduce pollutants in urban runoff and stormwater discharges. The protection of natural drainage systems, including wetlands, is an important part of these efforts.  The need for a more complete understanding of the effects of stormwater impacts on wetlands has been recognized (Newton, 1989; Stockdale, 1991).

A draft of this  issue paper was prepared to focus discussion on these and other related issues at an EPA-sponsored Wetlands and Stormwater Workshop held in Clearwater, Florida, in January 1992.  The purpose of the workshop was to investigate and explore various issues, options, and opinions related to the protection of natural wetlands that receive stormwater and urban runoff.  The focus of workshop discussions was not on methods for assessing or improving the capacity of wetlands to control stormwater discharges, but on what is known and not known concerning the impacts to natural wetlands from urban stormwater discharges and the opportunities for protecting natural wetlands that receive urban stormwater.  The major themes discussed at the workshop include the following:

The information developed at the workshop is being used by the Wetlands Division of EPA's Office of Wetlands, Oceans and Watersheds (OWOW) and the Office of Wastewater Enforcement and Compliance (OWEC) to develop joint guidance to address urban stormwater discharges to natural wetlands.  This issue paper incorporates information generated at the workshop as well as comments received on the draft issue paper.

BACKGROUND

Urbanization dramatically alters the natural hydrologic cycle.  As urban structures such as roads and buildings are built, the amount of impervious area within a watershed increases. Increases in impervious area increase the volume and rate of runoff, while decreasing groundwater recharge.  Urbanization also increases the type and amount of pollutants in surface runoff.

Uncontrolled urban runoff can have adverse impacts on urban wetlands.  The dramatic increases in peak flow rates can cause erosion and channelization in the wetlands, which ultimately adversely impact the ability of the wetland to support aquatic habitat.  Reductions in groundwater recharge within a watershed can reduce dry weather flows in wetlands.  The hydrology of a wetland is considered one of the most important factors in establishing and maintaining specific types of wetlands and wetland processes (Mitsch and Gosselink, 1986).  Relatively little information has been compiled on the adverse impacts of stormwater on natural wetlands (Woodward-Clyde Consultants, 1991; Newton, 1989; Stockdale, 1991).

Older approaches to stormwater management have focused on efficiently collecting and conveying stormwater offsite.  This approach can increase downstream property damage and impacts on receiving waters.  Newer approaches to stormwater management seek to retain natural features of drainage systems and provide onsite management to address water quality and water quantity goals.  This approach views stormwater as a resource to be used to recharge groundwater and to supply fresh water to surface waters, including wetlands.  Properly managing stormwater can avoid problems with erosion, flooding, and adverse impacts on natural drainage features, including wetlands.

Efforts to develop State and local stormwater management programs have been inconsistent nationwide.  Stormwater management approaches have been varied, and the ability of some approaches to protect receiving waters is not well known.  Some stormwater management controls, such as wet ponds, are designed to preserve some of the features of predevelopment hydraulic patterns and to provide some of the hydraulic and pollutant removal features of natural receiving systems.  The adverse effects and benefits of siting these controls in or near wetlands are not well understood.

The 1987 amendments to the Clean Water Act (CWA) contain two provisions addressing the control of pollutants in urban runoff and stormwater discharges.  Section 402(p) of the CWA requires EPA to develop phased requirements for discharges from municipal separate storm sewer systems and stormwater discharges associated with industrial activity under the National Pollutant Discharge Elimination System (NPDES) permit program.  NPDES permits for discharges from municipal separate storm sewer systems are to  effectively prohibit nonstormwater discharges to separate storm sewers and require municipalities to reduce the discharge of pollutants in stormwater to the maximum extent practicable.

EPA issued NPDES permit application requirements for discharges from municipal separate storm sewer systems serving a population of 100,000 or more on November 16, 1990.  The municipal component of the regulations focuses on requiring affected municipalities to develop municipal stormwater management programs to reduce pollutants in stormwater and protect receiving waters.

The November 16, 1990, regulations also addressed which types of facilities would be required to obtain NPDES permit coverage for stormwater discharges associated with industrial activity and specified permit application requirements for these discharges.

Section 319 of the CWA amendments requires States to identify waters that, without further action to control nonpoint sources, cannot be expected to attain the water quality standards or goals of the Act.  States were also to submit programs for management of nonpoint source pollution.

PURPOSE

The Wetlands and Stormwater Workshop was conducted to investigate the status of the science regarding the impacts and potential for use of natural wetlands for the storage and treatment of stormwater.  To this end, EPA formed a panel of wetland scientists, engineers, and environmental managers to

The purposes of the workshop were to:

A draft of this issue paper was originally developed to provide a base for discussion and to support deliberations at the Wetlands and Stormwater Workshop held in January 1992 in Clearwater, Florida.  Chapter 2 of this paper presents a summary of the characteristics and functions of natural wetlands most likely to be impacted by stormwater discharges.  An understanding of wetland functions is necessary to be able to predict and measure impacts resulting from stormwater discharges.  The hydrologic and chemical characteristics of urban stormwater are summarized in Chapter 3, with a focus on urban development activities that affect the quantity and quality of stormwater.  Chapter 4 presents a discussion of regional differences in wetland types and stormwater characteristics.  Such differences will influence the degree and character of potential impacts on natural wetlands due to urban stormwater discharges.

Chapter 5 presents a discussion of what is and is not known about changes that can be caused in wetland systems by stormwater, including hydro<%-2>logic changes, water quality changes, changes in the soil, and observed responses in plants and animal communities.  Chapter 6 presents an overview of stormwater management practices that include natural wetlands as a component.  Examples of practices currently being used by different States are presented.  The summary and conclusions are presented in Chapter 7.

2. WETLAND CHARACTERISTICS

For the purpose of this paper, wetlands are defined  as "those areas that are inundated or saturated by surface water 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 CFR 230.3).   This definition is used by EPA and the U.S. Army Corps of Engineers (Corps) in implementing section 404 of the Clean Water Act.  Table 1 briefly describes major freshwater and coastal wetland systems.

Wetlands are subject to increased attention relative to receiving stormwater runoff because of their inherent water storage and water quality improvement capabilities. The role of wetlands as storage areas for stormwater discharges was investigated by EPA (1985a) and Reinelt and Horner (1990), while Richardson (1989) and EPA (1983) documented the role of wetlands in water quality processes.  The value of natural wetlands, however, extends beyond their water storage and water quality functions to include food chain support, erosion control, groundwater recharge/discharge, and habitat functions.  An understanding of these functions is necessary when contemplating the use of natural wetlands to store and treat urban stormwater discharges in order to predict and measure potential impacts on wetland functions.  The potential impacts of urban stormwater on natural wetlands are discussed in Chapter 5.

HYDROLOGY

Hydrology is probably the most important determinant for the establishment and maintenance of specific types of wetlands and wetland processes (Mitsch and Gosselink, 1986).  Precipitation, surface water inflow and outflow, groundwater exchange, and evapotranspiration are the major factors influencing the hydrology of most wetlands.  The balance of inflows and outflows of water through a wetland defines the water budget and determines the amount of water stored within the wetland.  A wetland experiences natural water level fluctuations (WLFs) that are closely associated with the wetland's morphology and the basin's hydrologic regime (Stockdale, 1991).  WLFs are also determined by specific factors including wetland-to-watershed area ratios, level of watershed development, outlet conditions, and soils (Reinelt and Horner, 1990).  Changes in activities within the watershed (e.g., urbanization) will affect these natural WLFs.  

Table 1.  Wetland Types
NONTIDAL   FRESHWATER

Figure Page

Lacustrine -

 Associated with bodies of water greater than 2 m in depth, or less than 8 ha in area, or less than 30 percent covered by emergent plants.

Riparian -

 Associated with flowing water systems.  For example, bottomland wetlands are lowlands found along streams and rivers, usually on alluvial floodplains that are periodically flooded.  These are often flooded and termed bottomland hardwood forests.

Palustrine -

 Do not have channelized flow and either are not associated with bodies of water or form the headwaters of streams.  These wetlands include the following:

Marsh -

A frequently or continually inundated wetland generally characterized by emergent, soft-stemmed herbaceous vegetation adapted to saturated soil conditions.

Swamp -

Wetland dominated by woody vegetation.

Bog -

A peat-accumulating wetland that has no significant inflows and outflows and supports acidophilic mosses, especially sphagnum.

Fen-

A peat-accumulating wetland that receives some drainage from surrounding mineral soil and usually supports marshlike vegetation.

Wet prairie -

Similar to a marsh.

Wet meadow -

Grassland with waterlogged soil near the surface but without standing water for most of the year.

Pothole -

Shallow marsh-like pond, particularly as found in the Dakotas

Playa -

Term used in southwest United States for marshlike ponds similar to potholes, but with different geologic origin.

COASTAL

Tidal salt marshes -

Found throughout the world along protected coastlines in themiddle and high latitudes. In the United States, these wetlands are often dominated by Spartina and Juncus grasses.  Plants and animals in these systems are adapted to the stresses of salinity, periodic inundation, and extremes in temperature.

Tidal freshwater marshes -

Found inland from tidal salt marshes, but still experiencetidal effects.  These marshes are an intermediate in the continuum from coastal salt marshes to freshwater marshes.

Mangrove wetlands -

Found in subtropical and tropical regions.  These wetlands are dominated by salt-tolerant red mangrove or black mangrove trees.

SOURCE: Mitsch and Gosselink, 1986.

The time of year and the depth, frequency, and duration of inundation and soil saturation (wetland hydroperiod) are key factors in determining the impacts of water-level changes in wetlands (Stockdale, 1991).  Within the wetland, the wetland hydroperiod influences the biochemistry of the soils and is a major factor in the natural selection of wetland biota (Reinelt and Horner, 1990).  The hydroperiod is unique to each type of wetland, and its relative constancy ensures stability for that wetland.  Mitsch and Gosselink (1986) suggest characterizing hydroperiod by the ratio of flood duration divided by flood frequency (i.e., the amount of time a wetland is exposed to excess floodwaters over the average number of times a wetland is flooded in a given period).  Changes in the hydroperiod can affect such processes as nutrient transformation and availability (Hammer, 1992); responses of biota, including both enrichment of species and degradation of species diversity with succession to a different vegetative community, (Zimmerman, 1987); and amphibian egg and larval development  (Richter et al., 1991).  Changes in the hydroperiod can be measured by the average change in water level occurring in the wetland (Azous, 1991).  

Seasonality is also a characteristic of hydroperiod.  Some wetlands have water year-round, while others may become dry during the summer period.  Reduced groundwater base flows are frequently cited as a consequence of urbanization and may result in extending the length of the dry period in wetlands, with seasonally affected groundwater sources potentially impacting the life cycles of species dependent on the water column (Azous, 1991).

A major hydrologic feature of coastal salt marshes and freshwater tidal marshes is the periodic tidal inundation.  The tides act as a stress by causing submergence, saline soils, and soil anaerobiosis.  The tides act as a subsidy by removing excess salts, reestablishing aerobic conditions, and providing nutrients (Mitsch and Gosselink, 1986).  The periodic tidal inundations influence the species that occur in the wetland because of the water depth and duration of flooding.  Salinity is also a major factor in influencing what vegetation is found in the wetland, with a salinity gradient generally high in the low marsh and decreasing as the elevation increases. If the salinity in the adjacent waterbody is less than 5 parts per thousand (ppt), salt marsh vegetation is replaced by freshwater plants (Mitsch and Gosselink, 1986).

WATER QUALITY/BENTHIC PROCESSES

An important function of wetlands is their role in changes that occur in water quality. Many complex chemical and biological processes that affect water quality occur in wetlands.  The occurrence and timing of these processes are determined by the wetland type and the hydrologic regime of the wetland.  Wetland water quality processes include:

Sedimentation is the principal mechanism by which suspended solids are removed from the water column.  Sedimentation is directly related to the size of the particulate, the rate and type of flow through the wetland, and the residence time of the particulate.  Wetland systems that have long hydraulic residence times allow most settleable solids to be removed by sedimentation.  As particle size decreases, solids in the water column become more difficult to remove by sedimentation.  Wetlands with dense stands of vegetation also enhance sedimentation by decreasing the velocity of water flowing through them.

Filtration occurs as suspended pollutants are physically trapped by vegetation, biota, and sediments in the wetland.  Reduced velocities and dense vegetation promote greater pollutant removal.  Removal of pollutants by filtration through soils is effective in removing organic matter, phosphorus, bacteria, and suspended material.

Adsorption is a physical process by which dissolved pollutants adhere to suspended particulates or onto bottom sediments and the surfaces of vegetation.  It is also a factor in removing nutrients and heavy metals through sedimentation.  Suspended organic and inorganic materials have a strong tendency to adsorb other pollutants, such as refractory organics, hydrocarbons, bacteria, and viruses.  Since these substances are adsorbed onto suspended solids, they, too, are effectively deposited with the trapped sediment (Chan et al., 1981; Silverman, 1983 in PSWQA, 1986).  Both particulates and their associated contaminants can be considered pollutants.  Thus the removal of sediments from the water column by wetlands reduces the potential impact on receiving waters.

The excess water in wetland soils, along with biological and chemical activities, can change the soils from an aerobic to an anaerobic system, with many resultant chemical (reduction-oxidation) transformations in the wetland.  The chemical transformations are governed by pH and redox potentials (Eh) and determine the state of the nutrient, mineral, or heavy metal entering the water column in the wetland or infiltrating the groundwater. The relationship between Eh and pH manifests itself in chemical speciation;  e.g., the predicted pH level necessary to precipitate iron or manganese is much higher at low Eh levels than at higher Eh levels (Faulkner and Richardson, 1989).

Nitrogen and phosphorus speciation are two of the most important chemical transformations occurring in wetlands.  Of the many elements necessary to sustain biotic production in wetlands, nitrogen presents special research challenges because of its chemical versatility.  This versatility is expressed in the various valence states nitrogen can occupy (-3 to +5), in the intricate array of biotic and abiotic transformations in which nitrogen participates, and by the fact that, like few other elements, nitrogen occurs naturally in soluble and gaseous phases (Bowden, 1987).  In a wetland, only a fraction of available nitrogen is removed by plants, with the most effective removal by nitrification/denitrification (Knight et al., 1986).  A limiting factor for nitrogen removal is anoxia.  In aerobic substrates ammonia is oxidized to nitrate by nitrifying bacteria. Nitrates (NO3) are then converted to free nitrogen in the anoxic zones by denitrifying bacteria.

Phosphorus removal in wetlands systems occurs from adsorption, absorption, complexation, precipitation, and burial.  Removal rates are highest in systems where a significant clay content is present (Watson et al., 1988).  Another factor affecting phosphorus removal is the presence of iron, aluminum, or calcium.  For example, Richardson (1985) found that the phosphorus adsorption capacity of a wetland soil can be predicted by measuring the extractable aluminum content of the soil.  Removal effectiveness is limited by the contact surface area of the substrate and the root zone

BIOLOGIC/HABITAT FUNCTIONS

Wetlands provide a valuable source of food and habitat, and wetlands often become a focal point for varied wildlife populations within a particular region.  Wetland vegetation also provides nesting material and sites for numerous birds and mammals.  Some fish rely on vegetation clumps as sites for depositing their eggs and as nursery areas for fry (Atcheson et al., 1979).

Most wetlands receive extensive use by animals characteristic of terrestrial or purely aquatic environments, while many unique organisms are restricted to wetland environments (Mitsch and Gosselink, 1986).  Wetlands are also important habitats for a disproportionately high number of endangered and threatened plant, mammal, bird, reptile, amphibian, and fish species.  Some aquatic organisms may use wetlands seasonally as a spawning ground and nursery for their young, spending most of their adult lives in deeper waters.  Amphibians, reptiles, and invertebrates usually undergo an aquatic phase that requires water for breeding, egg development, and larval growth.  Some reptiles and amphibians are able to adapt to fluctuating water levels (Mitsch and Gosselink, 1986), whereas others may experience changes in breeding patterns and species composition due to water level fluctuations (Azous, 1991).  Wetlands are also used daily by birds and terrestrial animals for diurnal and nocturnal food foraging.  Many birds that inhabit both terrestrial and wetland habitats are frequently found in the highest numbers in the diverse, productive habitats of wetlands (NWTC, 1979).

The wetland vegetative community is determined by climate and wetland hydrology.  Wetland plant species are established based on their water regime requirements and on the natural hydroperiod of the wetland (van der Valk, 1981).  Plant species and diversity, in turn, have a direct effect on which wildlife will use the site.  Species diversity and abundance may vary greatly among different wetland locations and within a single wetland.  Some wetlands--acidic bogs, monotypic cattail (Typha) marshes, and many saltwater wetlands--can have high abundance but low plant species diversity.  Others, such as riverine swamps and fresh/brackish marshes, have high diversity.

Many emergent plant species are sensitive to changes in water levels in excess of the wetland's natural hydroperiod (Mitsch and Gosselink, 1986; Stockdale, 1991).  Excess depths, frequencies, and duration of inundation in wet seasons, or water deficiencies in dry seasons, have the potential to alter the vegetative community and, thus, the wildlife that use the wetland and the benthic and aquatic organisms that depend on the wetland.  These hydrologic changes can also directly affect some animals, such as amphibians, that have distinct preferences for placing their eggs in the water column (Richter et al., 1991).

UNRESOLVED ISSUES

Although much recent research has been directed at understanding the processes that control wetland functions, many questions remain to be resolved, particularly with respect to wetland functions as habitat.  Among these unresolved issues are the following:

3. STORMWATER CHARACTERISTICS

As human activities alter the watershed landscape, adverse impacts to receiving waters may result from changes in the quality and quantity of stormwater runoff.  Unmanaged storm surges increase discharges during runoff-producing storm events.  These discharges result in a predictable change of waters flowing to those receiving waters.  If left unmanaged, the hydraulic impacts associated with the increased water volumes may be several orders of magnitude higher than the impact of the undisturbed watershed.  In addition to causing runoff volume impacts, stormwater can also be a major source of nonpoint source pollution in many watersheds.

Six main source activities contribute to surface water runoff pollution:

The first five are the traditional sources;  the sixth, atmospheric deposition, has only recently been recognized as a major contributor of some types of nonpoint source pollution in certain regions of the country.  The type and quality of pollutants carried by storm runoff, commonly resulting in nonpoint source pollution of receiving waters, are highly variable (USEPA, 1984).  The pollutant characteristics of stormwater runoff are largely based on land use characteristics (as illustrated in Table 2) and vary with the duration and the intensity of rainfall events (Metropolitan Washington Council of Governments, 1980).  Table 2 illustrates the variability of pollutant loads associated with stormwater runoff.  For example, Table 2 shows that loads of suspended sediment vary considerably within a land use and between land uses.  Pollutant characteristics from stormwater runoff also vary regionally.

The remainder of this chapter focuses specifically on the chemical and hydrologic characteristics of urban stormwater.  Knowledge of these characteristics is necessary to understand and predict the potential impacts such discharges may have on natural wetlands.  The potential impacts of urban stormwater discharges on natural wetlands are discussed in Chapter 5 of this document.

Table 2.  Examples of Pollutant Characteristics Found in Stormwater Runoff From Various Land Uses in the Great Lakes Region
Land Use Suspended
Sediment
(kg/ha-yr)		 Total
Nitrogen
(kg/ha-yr)		Total
Phosphorus
(kg/ha-yr) 		Lead
(kg/ha-yr)

General Agriculture 5-8000 0.8-75 0.1-9 0.003-0.09
Cropland 30-7500 6-60 0.3-7 0.006-0.007
Improved Pasture 50-90 5-15 0.1-0.6 0.005-0.02
Forested/Wooded 2-900 1-8 0.03-0.7 0.01-0.05
Idle/Perennial 9-900 0.6-7 0.03-0.7 0.01-0.05
General Urban 300-2500 8-10 0.5-4 0.2-0.6
Residential 900-4000 6-9 0.6-1 0.08a
Commercial 75-1000 3-12 0.09-0.9 0.3-1.0
Industrial 750-2000 3-13 0.9-6 -b
Developing Urban >>10,000a 90a >10a 3.0-7.0

AOnly one value reported.

B Not measured

SOURCE : Novotny and Chesters, 1981.

URBAN ACTIVITIES THAT AFFECT STORMWATER CHARACTERISTICS

Urban runoff quantity and quality are significantly affected by watershed development.  Urbanization alters the natural vegetation and natural infiltration characteristics of the watershed, causing runoff from an urban area to have a much higher surface flow component, a much smaller interflow component, and a somewhat reduced baseflow component.  Urbanization also can create water quality problems because activities associated with urbanization create sources of pollutants for surface runoff.  Thus urbanization tends to increase runoff and pollutant loadings to the receiving waterbody (Woodward-Clyde Consultants, 1990).

The following sections describe the chemical and hydrologic characteristics of urban stormwater and the urban/development activities that affect those characteristics.

CHEMICAL CHARACTERISTICS

One significant effect of urbanization is to increase pollutant runoff loads over predevelopment levels.  During a storm event, land surfaces, including impervious surfaces, are washed clean by the rainfall and the resulting runoff creates an increased loading of pollutants to receiving streams (Livingston, 1989).  Pollutant concentrations in urban runoff vary considerably, both during the course of a storm event and from event to event at a given site, from site to site within a given urban area, and from one urban area to another across the country.  This variability is the result of variations in rainfall characteristics, differing watershed features that affect runoff quantity and quality, and variability in urban activities (Woodward-Clyde Consultants, 1990).  Table 3 presents ranges of urban runoff pollutant concentratons based on results of the Nationwide Urban Runoff Program (NURP) as reported in Woodward-Clyde Consultants (1990).  Values reported in Table 3 represent the mean of event mean concentration pollutant values for the median, 10th percentile, and 90th percentile sites in the NURP data.  Potential sources of urban runoff pollutants are presented in Table 4.  The principal types of pollutants found in urban runoff from these various sources include:

HYDROLOGIC CHARACTERISTICS

The most important factor in determining the quantity of runoff that will result from a given storm event is the percent imperviousness of the land cover.  Other factors include soil infiltration properties, topography, vegetative cover, and previous conditions (Woodward-Clyde Consultants, 1990).

The factors that influence the hydrologic characteristics of stormwater are dependent on the phase of urbanization of an area.  During the construction phase, the hydrology of a stream changes in response to initial site clearing and grading.  Trees that had interrupted rainfall are felled.  Natural depressions that temporarily ponded water are graded to a uniform slope.  The thick humus layer of the forest floor that had absorbed rainfall is scraped off or eroded away.  Having lost much of its natural storage capacity, the cleared and graded site can no longer prevent rainfall from being rapidly converted to runoff (Schueler, 1987).

After construction is completed, rooftops, roads, parking lots, sidewalks, and driveways make much of the site impervious to rainfall.  Unable to percolate into the soil, rainfall is converted into runoff.  The excess runoff becomes too great for the existing drainage system to handle.  As a result, the drainage network must be improved to direct and convey the runoff away from the site (Schueler, 1987).

Table 3.  Ranges in Pollutant Concentrations Found in Urban Runoff

Mean Concentration in Runoff
       10th Percentile Constituent         Median          90th Percentile Constituent
Urban Site                Urban Site                Urban Site

Total Suspended Solids (mg/L) 35 125 390
BOD (mg/L) 6.5 12 20
COD (mg/L) 40 80 175
Total Phosphorus (mg/L) 0.18 0.41 0.93
Soluble Phosphorus (mg/L) 0.10 0.15 0.25
Total Kjedahl Nitrogen (mg/L) 0.95 2.00 4.45
Nitrate-Nitrogen (mg/L) 0.40 0.90 2.20
Total Copper 15 40 120
Total Lead 60 165 465
Total Zinc 80 210 540

SOURCE:  Woodward-Clyde Consultants, 1990.<R>

Table 4.  Sources of Urban Runoff Pollutants
Source Pollutant of Concern

Erosion Sediment and attached soil nutrients, organic matter, and other adsorbed pollutants.
Atmospheric Deposition Hydrocarbons emitted from automobiles, dust, aromatic hydrocarbons, metals, and other chemicals released from industrial and commercial activities.
Construction Materials Metals from flashing and shingles, gutters and downspouts, galvanized pipes and metal plating, paint, and wood preservatives.
Manufactured Products Heavy metals; halogenated aliphatics; phthalate esthers; PAHs; other volatiles; and pesticides and phenols from automobile use, pesticide use, industrial use, and other uses.
Plants and Animals Plant debris and animal excrement.
Nonstormwater Connections Inadvertent or deliberate discharges of sanitary sewage and industrial wastewater to storm rainage systems.
Accidental Spills Pollutants of concern depend on the nature of the spill.
SOURCE:  Based in part on Woodward-Clyde Consultants, 1990.

The following changes in stream hydrology in a typical, moderately developed watershed were summarized by Schueler (1987):

4. REGIONAL DIFFERENCES
IN STORMWATER CHARACTERISTICS AND IN WETLAND TYPES

To comprehensively evaluate the impacts and potential for use of natural wetlands for the storage and treatment of urban stormwater runoff, it is essential to understand the regional variations, both in stormwater runoff and in natural wetland types, that exist throughout the Nation.  The following sections briefly summarize these differences.

REGIONAL DIFFERENCES IN STORMWATER CHARACTERISTICS

The characteristics of precipitation events control the timing, volume, and intensity of urban stormwater runoff. The U.S. Department of Agriculture, Soil Conservation Service (SCS) developed dimensionless rainfall distributions using U.S. Weather Bureau data (McCuen, 1989).  The distributions are based on generalized rainfall volume-duration-frequency relationships and indicate that there are four geographically distinct rainfall regions in the United States, illustrated in Figure 1.  Figure 2 is a dimensionless hydrograph that shows the hourly fraction of total rainfall that falls in a 24-hour period for each rainfall distribution type (Ferguson and Debo, 1990).

*Figure 1. Approximate geographic areas for SCS rainfall distributions (Adapted from McCuen, 1989))

*(Figure 2. Hourly fraction of total rainfall within a 24-hour period for each rainfall distribution type (Ferguson and Debo, 1990))

Climatic variations result in different storm intensities for each rainfall distribution type.  Figure 3 illustrates the major climatic regions of North America (Ahrens, 1982) and shows that the four SCS rainfall distribution types have very different climatic regimes.  Figure 4 shows the frequency of thunderstorms experienced nationwide, given in days per year when thunderstorms are observed (Ahrens, 1982).  It is obvious why Type IA, with 20 percent of the rainfall volume falling during the 8th hour of a 24-hour storm, and Type III, with 55 percent of the rainfall volume falling during the 12th hour of a 24-hour storm, have very different stormwater characteristics.  Type IA regions have Maritime and Mediterranean climates with onshore winds producing mild, wet winters with frequent, light precipitation; dry summers; and very few thunderstorms annually.  Typical Type IA storms are long, steady periods of relatively light rainfall.  Type III regions are coastal areas with a humid subtropical climate, with adequate precipitation throughout the year, cold to mild winters, and hot and humid summers with frequent thundershowers.   Typical Type III storms are short, high-intensity rainfall events.

* - To view the Figures, please call (202) 260-9043 to order a copy.

*(Figure 3 - Major climatic regions on North America (after Koppen)(AAhrens, 1982))

*(Figure 4 - Average number of days each year on which thunderstorms are observed throughout the United States (Ahrens, 1982))

Seasonality of precipitation, the time of year in which the precipitation falls, is another important factor that must be considered when evaluating stormwater runoff.  Water available during the growing season has a different impact from that of water that becomes available when the plants are dormant.  Precipitation that falls as snow during the winter months and melts during the spring thaw has a profound effect on local stream levels.

*(Figure 5 - Month-to-month variation of precipitation in the United States (U.S. Weather Bureau (Chow, 1964))

When precipitation intensity and frequency are combined with watershed cover characteristics, the runoff characteristics of a region can be estimated.  Runoff estimations are useful for estimating impact to a natural wetland from stormwater.  Figure 6 gives average annual runoff in inches for the United States (Chow, 1964).

(Figure 6 - Average annual runnoff in the United States (Chow, 1964))

REGIONAL DIFFERENCES IN WETLAND TYPES

The United States has a wide range of wetland types that result from the interaction of many separate environmental variables.  The characteristics of wetlands derive from and are controlled by two interrelated factors: (1) origin and (2) regional climatic factors.

The origin of a wetland, and the resulting topography, affects and determines critical wetland aspects such as elevation, drainage, and soils.  Wetlands are created by one or more basic processes: geological forces (tectonic, volcanic, and glacial activities);  erosion and sedimentation; animal activity; and human activities (OTA, 1984; Hammer, 1992).

The second major controlling factor that leads to the formation of regional wetland types is climate.  Because hydrology is critical in establishing and maintaining a wetland (Mitsch and Gosselink, 1986), the variable climates in the United States have contributed to the formation of distinctly different wetland types (OTA, 1984).  Figure 7 shows the major physiographic regions of the United States, and Table 5 gives the geographic locations of wetland types in the United States (OTA, 1984).  The relationships of the various wetland types to their origin, hydrology, soils, and vegetation are summarized in Table 6.

* - To view the Figures, please call (202) 260-9043 to order a copy.

RELATIONSHIP BETWEEN REGIONAL CHARACTERISTICS OF STORMWATER AND WETLANDS

The relationship between stormwater characteristics and wetland type will influence the degree and character of the impacts to natural wetlands that may result from urban stormwater discharges.  The regional differences in stormwater hydrology and wetland types summarized above can play a significant role in determining such impacts.   Table 7 presents the wetland types that occur in each of the SCS rainfall distribution areas to illustrate a method of describing the relationship between stormwater characteristics and wetland types.   By identifying those regions of the country in which  rainfall is characterized by the fraction of rain that falls per hour during a 24-hour period (e.g., rainfall is more or less evenly distributed during a 24-hour period or is characterized by a gradual build-up of rainfall, followed by a brief period of relatively intense rainfall and gradual dissipation) and the wetland types that occur in those regions, storm hydrology can be linked to wetland type.  Also, the actual hydroperiod characteristics of natural wetlands depend on specific watershed land use and wetland morphology, soils, and biological nature in addition to regional climate.  The effects that regional differences in wetland type and stormwater characteristics may have on impacts on natural wetlands that receive stormwater are briefly discussed in Chapter 5.

*(Figure 7 - Physiographic regions of the United States (Mitsch and Gosselink, 1986)

* - To view the Figures, please call the EPA Wetlands Hotline @ (1-800-832-7828 to order a copy of manual.

 Table 5.  Locations of Wetland Types in the United States
Wetland Type Primary Regionsa States

Inland freshwater marshes Prairie pothole region: Eastern Highlands (7); Upper Midwest (9);Dakota-Minnesota drift and lake bed flats (8); Central Hills andbed flats (8); Plains (10)West Coast: Pacific Mountains (13) New York and New Jersey to North Dakota and eastern Montana; Washington, Oregon, northern California
Bogs Upper Midwest (9); Gulf-Atlantic Rolling Plain (5); Gulf Coastal Flats (4); and Atlantic Flats (3) Wisconsin, Minnesota, Michigan, Maine, Florida, North Carolina Florida, North Carolina
Tundra Central Highland and Basin; Arctic Lowland; and Pacific Mountains Alaska
Wooded swamps Upper Midwest (9); Gulf Coastal Flats (4); A tlantic Coastal Flats (3); and Lower Mississippi Alluvial Plain(6) Minnesota, Wisconsin, Michigan, Florida, Georgia, South Carolina, North Carolina, Louisiana
Bottom land hardwood Lower Mississippi Alluvial Plain Atlantic Coastal Flats (3);Gulf Atlantic Rolling Plain (5); and Gulf- Coastal Flats (4) Louisiana, Mississippi, Arkansas Missouri, Tennessee, Alabama, Florida, Georgia, South Carolina, , North Carolina, Texas
Coastal salt marshes Atlantic Coastal Zone (1); Gulf Coastal Zone (2); Eastern High- Lands (7); Pacific Mountains(13) All coastal States, but particularly the Mid-and South Atlantic and Gulf Coast States
Mangrove swamps Gulf Coastal Zone (2) Florida and Louisiana
Tidal freshwater marshes Atlantic Coastal Zone (1) and Flats(3); Gulf Coastal Zone(2) and Flats (4) Texas, Louisiana, Mississippi, Alabama, Florida, all of the Atlantic coastal states
aNumbers in parentheses refer to the geographic regions in the United States identified in Figure 7 SOURCE: Adapted from OTA, 1984 and Mitsch and Gosselink, 1986.

To view Table 6, please call 1-800-832-7828 to order a copy the Natural Wetlands and Urban Stormwater: Potential Impacts and Management maunal.

Table 7. Wetlands Present in SCS Type Rainfall Distribution Areas

SCS Type Wetland Type State
Type I Tundra
Coastal salt marsh
Tidal freshwater marsh		 AK, HI
WA, OR, CA
CA, AK
Type IA Coastal salt marsh
Tidal freshwater marsh 		WA, OR, CA, AK
WA, OR, CA
Type II Inland freshwater marsh NY, PA, OH, MI, IN, WI, IL, MN,
ND, SD, MT, WA, OR
Bogs WI, MN, MI, ME, FL, NC
Wooded swamps MN, WI, MI, FL, GA, SC, NC, LA
Bottomland hardwood LA, MS, AR, MO, TN, AL, FL, GA, SC,
NC, TX
Coastal salt marsh DE, MD, VA
Mangrove swamps FL
Tidal freshwater marsh FL, VA, MD, DE
Type III Coastal salt marsh ME, NH, MA, RI, CT, NY, NJ, NC, SC, GA, northern FL, AL, MS, LA, TX
Mangrove swamp FL, AL, MS, LA, TX
Tidal freshwater marsh TX, LA, MS, AL, FL, GA, NC, NJ, NY,
CT, RI, MA, NH, ME
Inland freshwater marsh ME, NH, MA, RI, CT, NY, NJ, VA, NC,
SC, GA, FL, AL, MS, LA, TX, AK, OK

5. POTENTIAL IMPACTS OF URBAN STORMWATER RUNOFF ON NATURAL WETLANDS

Stormwater runoff has the potential of influencing natural wetlands in four major areas: wetland hydrology, wetland soils, wetland flora and fauna, and wetland water quality.  There is little doubt that urban stormwater discharges can affect wetlands; however, the long-term impacts on natural wetlands from urban stormwater discharges are not known at this time.  Perturbations to wetland hydrology can cause fluctuations in the character of the ecosystem that are seen as changes in the species composition and richness, primary productivity, organic deposition and flux, and nutrient cycling (Livingston, 1989).  Naturally occurring quantities of runoff with seasonal fluctuations are essential for the maintenance of a wetland, and moderate amounts of nutrients and sediment in the runoff can increase a wetland's productivity (Stockdale, 1991).  However, excessive stormwater discharge on a continuous basis has the potential to alter wetland hydrology, topography, and the vegetative community (Johnson and Dean, 1987 in Stockdale, 1991).  A few investigations that look at the potential impacts to natural wetlands from stormwater discharges have been initiated.  Some of these impacts have been identified and others require further investigation.  This chapter examines the nature of changes to wetland hydrology, soils, and water quality attributed to stormwater runoff and the perceived effects on the biologic community.

HDYROLOGIC CHANGES

As a result of urbanization, the quantity and quality of stormwater runoff change due to physical changes occurring in the watershed.  The quantity of water entering a wetland as stormwater runoff is dependent on factors such as drainage basin area, imperviousness of the drainage basin, routing of stormwater within the drainage basin, and climate (Lakatos and McNemar, 1987).  Increased impervious area in the watershed (from building construction, roadways, and parking lots), removal of trees and vegetation, and soil compaction can increase the quantity of urban stormwater runoff (Schueler, 1987).  Water velocity also increases, in general, as the degree of urbanization increases (Viessman et al., 1977).  These same activities will potentially cause decreased infiltration of stormwater to groundwater, resulting in decreased base flow.

One basis for determining the impacts to a wetland from stormwater runoff is the wetland's natural hydroperiod.  Impacts will also vary depending on the wetland type and size and whether the runoff is intercepted before entering the wetland.  Brinson (1988) characterized wetlands, geomorphologically, in three major categories:

Although these classifications are very general, Brinson (1988) acknowledges that classification of many wetlands is not clear-cut and the definitions tend to overlap.

Known impacts to wetlands associated with increased storm runoff include change in wetland response time, change in water levels in the wetland, and change in detention time of the wetland.  The response time is the time it takes for a wetland's water depth to begin to rise in response to a storm event occurring in the watershed.  The wetland's water depth will begin to rise sooner as the infiltration capability of the watershed decreases. The greater the amount of runoff entering the wetland soon after the storm event, the greater the water level fluctuation (WLF) (Azous, 1991).  On the other hand, the increased runoff at the expense of infiltration may cause local water tables to be reduced along with reducing base flows of local streams (USEPA, 1985).  Reduction in groundwater base flows has the potential effect of extending the length of dry periods in wetlands with seasonally affected groundwater sources, potentially impacting the life cycles of the species dependent on the water column (Azous, 1991).

Increased impervious surface areas have the effect of increasing flood peaks during storms and decreasing low flows between storms (Stockdale, 1991).  Larger peak flows can result in scoured streambeds as the beds enlarge to accommodate larger flows.  Associated impacts include increased sediment loading to bordering vegetated wetlands and reduction of fish spawning habitat (Canning, 1988).  In addition to increased flows, urbanization can increase the velocity of the stormwater entering the wetland, which can result in biotic disturbances (Stockdale, 1991).  Disrupted flow patterns and channeling can result in decreased pollutant removal efficiencies (Morris et al., 1981), and the changes in velocity will determine deposition as well as eroded areas (USEPA, 1985).

Changes in average water levels or duration or frequency of flooding will also alter species composition of plant and animal communities and distribute pollutants more extensively throughout the wetland (Stockdale, 1991).  Cooke (1991) states that species richness is affected by increases in water level fluctuation, with decreased species richness associated with higher water level fluctuations than are found in natural systems.  The flood tolerance and sensitivity of different plant species vary greatly and will dictate the response to flooding stress.  Responses of vegetation to WLF are discussed in the Biologic/Habitat Impacts section.

WATER QUALITY CHANGES

As stormwater runoff passes through a wetland, its quality often changes and the changes tend to be variable and difficult to predict.  The ability of a wetland to remove pollutants from water has typically been the predominant reason cited to promote the use of wetlands for stormwater runoff treatment.  Studies have been conducted to determine the pollutant removal capacity of natural wetlands (Schiffer, 1989; ABAG, 1979; Hickok et al., 1977), and recent studies have tried to address the impacts of stormwater runoff on wetlands (Horner, 1988; Cooke, 1991; Reinelt and Horner, 1991).  Table 8 gives an example of seasonal water quality characteristics for stormwater from various land uses.   Note the differences from season to season (e.g., total phosphorus for undeveloped land was 2.2 mg/L, 0.37 mg/L, and 0.30 mg/L from spring to fall) and between land uses.  Changes in water quality as stormwater runoff passes through a natural wetland are examined in this chapter by discussing physical, chemical, and biological changes separately.

The predominant physical water quality parameters of concern are temperature, conductivity, and suspended solids (Reinelt and Horner, 1991).  As urbanization increases, these parameters typically increase in stormwater runoff and likewise in wetlands (Reinelt and Horner, 1991).  Increases in water temperature are attributed to warming of runoff as it passes over warmed impervious surfaces.  Conductivity increases are related to increases in the total dissolved solids that typically are found in stormwater runoff.

Of these three physical parameters, suspended solids are typically the pollutant of concern, primarily because solids tend to settle within the wetland.  A good example of sedimentation rates achievable in wetlands, in which a wetland was found to trap 16,009 kg of sediment per year, is illustrated by Hickok et al. (1977).  These results represented a reduction of 94 percent of the suspended solids entering that wetland on an annual basis.  Hickok et al. (1977) pointed out that total suspended solids are often the parameter that exceeds effluent requirements of stormwater runoff.  Other authors (ABAG, 1979; Schiffer, 1989) have also reported high percentage removals of suspended solids from stormwater runoff passing through a wetland.  

Table 8.  Comparison of Stormwater Runoff Quality
SPRING 1975 SUMMER 1975 FALL 1975
Drainage
Group		 TP
(mg/L) 		NH3-N
(mg/L)		 TSS
(mg/L)		 TP
(mg/L)		 NH3-NTSS
(mg/L)		 TP
(mg/)L 		NH3N
(mg/L) 		TSS
(mg/L)

374
Undeveloped 2.2 3.33 780 0.37 4.13 1200 0.30 4.33 70
Low-Density 2.4 3.87 559 0.73 5.44 3800 0.42 4.97 60
Residential Business/
Commercial 2.0 2.85 580 0.22 5.130.22 4.00 68
Urban Roadway 1.9 2.55 614 0.09 2.86 200 0.25 3.81 100
Average 2.1 3.15 633 0.35 4.39 1394 0.30 4.28 75

Note: Above concentrations are based on weighted values calculated from specific runoff events that occurred during the study period.

TP = Total phosphorus
NH3-N = Ammonia nitrogen
TSS =Total suspended solids<R>SOURCE:  Hickok et al., 1977.

Chemically, water quality parameters of concern can be broken down into nutrients, metals, and other toxics.  Nutrients include phosphorus and nitrogen and are generally linked to eutrophication problems in receiving waters.  Metals present in stormwater runoff may include copper, chromium, cadmium, nickel, lead, iron, manganese, and zinc. Other metals may be present depending on the specific activities within the drainage basin feeding the wetland.  Miscellaneous toxics that may be present in stormwater runoff include pesticides, hydrocarbons, and organic compounds.  Table 9 compares three wetlands used to treat stormwater runoff and gives an indication of the variability of pollutant removal between wetlands.  This table shows some of the variability found between different wetlands.  For example, phosphorus decreases about 79 percent in the Wayzata wetland, increases about 6 percent in the Palo Alto wetland, and decreases about 87 percent in the Island Lake wetland.

The fate of chemicals entering a wetland is highly variable and depends on many chemical and physical factors (Richardson, 1989).  At times, wetlands serve as a sink for pollutants, which are stored in the wetland.  Wetlands can also transform pollutants from one form to another.  The transformation may be from a desirable to an undesirable state, or the converse can occur.  The complex chemical reactions that occur in wetlands change with time.  For example, a pollutant being stored in a wetland can become a transformed pollutant that is subsequently exported from the wetland (Richardson, 1989).

Biological changes in water quality for wetlands receiving stormwater runoff typically are reported as changes in the bacteriological quality of the water.  Horner (1988) reported that bacterial indicatorn (fecal coliforms and enterococci) within wetlands increased in numbers in more highly urbanized watersheds.  The levels reported by Horner (1988) did not exceed water quality standards and were in areas not typically used directly by humans.  However, wetlands with elevated bacterial levels that discharge to shellfish areas may be of concern.  Reinelt and Horner (1991) found that the water columns in wetlands with a flow-through (more channelized) character in urbanized areas had higher bacterial levels than more quiescent open-water systems.  The difference was attributed to settling of sediment, and the adsorbed bacteria, out of the water column in the open-water systems.  Reinelt and Horner (1991) compared levels of chlorophyll a, an indicator of algal growth, in several wetlands and found that open-water wetland systems had higher levels than those of other systems.

WETLAND SOIL CHANGES

Physical, chemical, and biological qualities of the soil substrate change in wetlands as they are subjected to stormwater runoff.  Soils are storage facilities for many potentially toxic compounds including heavy metals.  Urban stormwater input has the potential to change the pH and redox potential of soils, rendering many toxins available from the storage pool so that they can have an immediate effect on wetland soils, both in situ and potentially downstream (Cooke, 1991).  The rate of metal accretion and the degree of burial in the sediments are critical factors in determining the loadings that can be endured by wetlands without damage (USEPA, 1985).  Physical property changes of wetland soils due to stormwater runoff such as texture, particle size and distribution, and degree of saturation are not well documented in the literature.  Some of the physical properties will be affected by other processes that occur due to stormwater runoff.  For example, as sediment is deposited in the wetland, the soil will take on the characteristics of the sediment addition. As the hydrology changes in a wetland, the soil moisture patterns may also change to reflect new conditions.

Similar to the physical properties, the chemistry of wetland soils change as processes change in the wetland.  Chemical property changes typically reflect sedimentation patterns as documented by Schiffer (1989) and ABAG (1979) and are illustrated in Table 10.  The findings of Horner (1988), relative to the greater accumulation of some metals in some zones, were high heavy metal accumulation occurring in the inlet zone of wetlands affected by urban runoff.  Wetland soils typically act as a sink for nutrients and metals, as evidenced in Tables 11 and 12.  Note the large differences in constituent concentrations for phosphorus, nitrogen, and some metals.  Another chemical process that occurs in wetlands is the adsorption of some chemicals to the existing soil particles in the wetland (Richardson, 1989).  Chemical processes in wetlands are also transient.  As water chemistry changes, pollutants that are stored in wetland soils can be transformed from solid to dissolved phases and become exported from  a wetland.  For example, as the soil-water interface becomes anaerobic, the redox potential changes, and pollutants like phosphorus are transformed from solid to dissolved phases.

Table 10.  Summary of Mean Soil and Sediment Chemistry Data as a Function of Sampling Location, December 1978 (in mg/kg unless noted)

Source of
Validation		 pH EC
(mmhos/cm) 		Organic
Carbon 		TKNNH3-N Total-P Available
P		 Cu Pb Zn

Lateral Positio
Lower Marsh 4.9 21 2.0 458 72605 2.7 14 28 23
Middle Marsh 5.9 10 2.1 37237 667 7.3 14 38 26
Upper Marsh 5.7 18 2.3 1,320 52 700 7.4 23 70 25
Vertical Position
0 - 8 inches 5.9 10 2.0 1,002 39 710 10.0 19 48 27
8 -6 inches 5.5 17 2.4 495 70 707 4.5 17 56 25
16 - 24 inches 5.2 22 2.1 670 51 550 2.9 15 33 22
Vegetative Cover
Pickleweed 4.7 25 2.0 948 40 719 8.7 15 32 23
Salt Bush 5.1 13 2.0 388 30 747 7.2 17 41 25
Rye Grass 4.7 9 2.0 995 30 645 5.9 13 27 24
No Vegetation 7.7 17 2.6 545 120 506 1.2 23 88 27 (Stream Channel)

SOURCE: ABAG, 1979.<D>

Table 11.  Distribution of Selected Constituents in Water, Sediments, and Groundwater at the Silver Star Road Study Area

Water Column (mg/L) Sediments (mg/kg) Constituent
Pond Inlet Wetland Inlet Wetland Outlet Pond Wetlands

Specific conductancea 145 144 153 -- --
pH-labb 7.2 7.1 6.9 -- __
Ammonia nitrogen 0.8 0.2 0.4 92 14
Nitrogen ammonia 0.10 0.10 0.10 9 6
 plus nitrite
Phosphorus 0.06 0.10 0.08 1,100 260
Total organic 15 15 15 -- --
 carbon
Cadmium <0.001c <0.001c <0.002 <6 <1
Chromium <0.003 <0.001c <0.002 20 2
Copper 0.01 <0.01c <0.01c 49 3
Iron -- -- -- 4,400 640
Lead 0.034 0.026 0.010 620 20
Zinc 0.06 0.05 0.03 250 14
aMicrosiemens per centimeter at 25 °C.
bpH units.
cDetection level
SOURCE:  Schiffer, 1989.

Biological activity within wetland soils is also subject to change due to changing conditions.  The only documented change found in the literature was soil microbial activity.  Hickok et al. (1977) examined changes in soil microbial activity due to changes in soil moisture conditions and found microbial activity to be directly related to increases in soil moisture.

BIOLOGIC/HABITAT IMPACTS

The impacts of urbanization and stormwater discharge on wetland systems are interactive and not clearly understood at this time.  The changes in hydrology, vegetative productivity and community structure, and water quality are inseparable.  Changes in the vegetative community structure appear to be correlated with the time of year, water depth changes, and frequency and duration of inundation experienced in the wetland from excess stormwater discharge (Stockdale, 1991; Azous, 1991; Cooke, 1991, USEPA, 1985).  USEPA (1985) reports that marked changes in water depth and frequency of inundation can result in changes in plant species composition and can affect plant production, as well as influence dissolved oxygen in the water column and in the soils.  The tolerance to water depth changes varies with each plant species and will dictate the response to flooding stress (Stockdale, 1991).  Local ecotypes within a species may also vary in their tolerance to flooding and soil saturation (Tiner, 1991).

Table 12.  Median Values of Selected Constituents in the Water Column and Values for One Sample of Bed Sediments at the Island Lake Wetland

Water Column (mg/L)

Constituent Inlet Outlet Sediments (mg/kg)
Specific conductancea 140 100 --
pH -laboratoryb 7.1 6.9 --
Nitrogen, ammonia 0.23 0.01 80.5
Nitrogen, ammonia plus organic 1.4 0.82 9,600
Phosphorus 0.23 0.03 2,250
Total organic carbon 7.5 20.5 --
Cadmium 0.001c 0.001c 2
Chromium 0.0075 0.0045 40
Copper 0.008 0.001 26.5
Iron 0.605 0.490 4,550
Lead 0.018 0.003 390
Zinc 0.075 0.025 175
aMicrosiemens per centimeter at 25 °C.
bpH units.
cDetection level.
SOURCE:  Schiffer, 1989 .

Increasing flood frequency or water level fluctuations could cause the mortality of certain plant species while favoring the productivity of others.  Stockdale (1991) in his literature review states that the character of wetland vegetation and riparian areas is primarily governed by the flooding regime (Thibodeau and Nickerson, 1985), with periodic inundation promoting richer and more abundant species composition than either constant dry or constant flooded conditions (Conner et al., 1981; Gomez and Day, 1982).  Determining plant responses to these stresses is difficult because direct responses  (physical damage) and indirect responses (physiological responses to direct impacts) are numerous and often simultaneous (Koslowski, 1984 in Azous, 1991).

Plant species are generally specific in their requirements for germination, and many are sensitive to flooding effects once established (Niering, 1989 in Azous, 1991).  Mature trees may survive inundation, whereas the same water level fluctuations may retard or limit the establishment of seedlings and saplings (Stockdale, 1991).  Newton (1989) and Stockdale (1991) list the relative flood tolerance of woody plants.  Little information is available on the effects of hydroperiod on emergent plants, though Kadlec (1962) found that several species of emergent plants were tolerant of lengthy dry periods (Azous, 1991).   Because the tolerance to flooding, intermittent and prolonged, varies so widely among and within plant species, it is hard to extrapolate from the literature what the impact on a certain plant species within a community will be.  Some information, however, is known about hydroperiod impacts on individual species (Stockdale, 1991):

Horner (1988) found that emergent zones of palustrine wetlands receiving urban runoff in the Pacific Northwest were dominated by an opportunistic exotic grass (Phalaris arundinaceae) while unimpacted wetland plant communities were composed of a more diverse group of species.  Ehrenfeld and Schneider (1990) found a relationship between stormwater discharge and changes in plant community composition in the New Jersey Pinelands; there was a reduction in indigenous wetland species and colonization of exotic species due to changes in hydrology, water quality, or both.  Wetland plant species may have a limited ability to migrate in the face of persistently raised water levels; many species can spread only through clonal processes under such conditions because of seed bank dynamics (van der Valk, 1991).  The result may be lowered plant diversity over the wetland-to-upland gradient.

Azous (1991) reports that many Pacific Northwest amphibians undergo an aquatic phase that requires water for breeding, egg development, and larval growth.  Changes in wetland water level may alter the quantity and quality of amphibian habitat, triggering changes in breeding patterns and species composition (Minton, 1968 in Azous, 1991).  Egg development may be impacted by a decline in WLF by potential exposure and desiccation when stranded on emergent vegetation (Lloyd-Evans, 1989 in Azous, 1991).  WLF may also cause changes in water temperatures, which may impact egg development (Richter et al., 1991).

Freshwater hydrologic disturbances were also correlated to responses of fish and macrobenthic assemblages (Nordby and Zedler, 1991).  In the study, two coastal marshes with different hydrology, one of which was impounded from tidal action, were compared.  Results show that the fauna was most depleted where the hydrologic disturbances were the greatest, with the trends over the course of the study being reduced species richness and abundance.

Among potential impacts brought up by workshop participants was the mortality of eggs or young of waterfowl due to flooding during the nesting period.  Also, continuity of habitat around wetlands receiving stormwater may be important in allowing wildlife free movement and refuge during storm events.

Wetland mammal populations may potentially be affected by change in hydroperiod because of the loss of vegetative habitat and the increased potential for disease organisms and parasites due to shallower, warmer base flow conditions (Lloyd-Evans, 1989 in Azous, 1991).

Changes in water quality (chemistry and sediment loading) have the potential to affect the vegetative community structure and to reduce the availability of plant species preferred by fish, mammals, birds, and amphibians for food and shelter (Lloyd-Evans, 1989 in Azous, 1991; Mitsch and Gosselink, 1986; Weller, 1987 in Azous, 1991).  For example, Azous (1991) found that plant community richness, evenness, and dominance were negatively correlated with the presence of total organic carbon in the water column.  Further studies are needed to determine the levels of heavy metal concentrations in the water column that will affect the plant species diversity in the wetland.

Despite the fact that little work has been documented on the effects of water quality changes on aquatic organisms, such changes have the potential to impact life cycles.  The ability of aquatic organisms, especially amphibians, to readily absorb chemicals suggests that they are responsive monitors of wetland conditions (Richter and Wisseman, 1990).  Richter et al. (1991) state that significant negative correlations were found between amphibian species richness and water column conductivity.  Negative changes in water quality and potential accumulation in soils and macrobenthic organisms suggest that bioaccumulation may occur in the bird and mammal communities.  Further studies are required to determine whether bioaccumulation is occurring and to what degree.

The habitat requirements, life histories, and species assemblages of wetland communities are relatively unknown at this time, requiring further investigation before impacts from stormwater discharges into wetlands can be determined.

REGIONAL DIFFERENCES

The degree and character of impacts to natural wetlands due to urban stormwater runoff described above will vary from region to region and even from site to site.  These impacts will vary due to regional differences in storm events, wetland types, watershed characteristics, and pollutant loads.  For example, geographical areas with Type II and III rain distributions (see Chapter 4) are those in which relatively intense rainfall occurs over a relatively brief period of time.  Certain wetland types that occur in these regions (e.g., coastal wetlands and seasonally wet areas) may be particularly vulnerable to stormwater discharges characterized by this type of rainfall.  In addition to regional climate, other factors including watershed land use and wetland morphology, soils, and biological nature influence actual wetland hydroperiod characteristics and changes that may occur in these characteristics as a result of stormwater discharges.  For example, Reinelt and Horner (1990) found that water level fluctuation patterns of wetlands depend on such factors as watershed use (e.g., level of urbanization), wetland bathymetry, vegetation, inlet and outlet conditions, and others.  The authors found that the level of urbanization and wetland outlet conditions appear to be the most significant factors influencing water level fluctuation.

Regional differences not only will result in differences in observed changes to wetland hydrology, water quality, and soils, but also will influence changes that occur in wetland vegetation, benthic organisms, and the wildlife functions of the wetland.  No attempt will be made in this paper to characterize the regional differences in impacts to wetlands that occur or may occur as a result of stormwater discharges.  In most cases, the exact nature of such impacts is not known.  Regional differences in impacts will occur, however, and will need to be considered in any stormwater management program.

IMPOUNDMENTS

Wetlands have historically been impounded for a variety of management purposes, but the primary reason has been wildlife and habitat management.  Treatment of stormwater runoff in natural wetlands by impounding all or part of a wetland is currently being used as a method for stormwater management (Livingston, 1988; ABAG, 1991).  For example, some impounded tidal wetlands in California are used to detain stormwater during rainfall events for later gradual release to the San Francisco Bay during low tides (ABAG, 1991).  When a wetland is partly or wholly impounded for stormwater management, its water quality improvement, flood attenuation, sediment retention, or groundwater recharge capabilities are being exploited, possibly at the expense of other wetland functions such as habitat for fish and wildlife.

An impoundment is defined as a body of water confined by a dam, dike, floodgate, or other barrier (USEPA, 1989).  Often the impoundment of a wetland (e.g., for stormwater treatment) results in changes in the wetland.  These changes may result in such extreme modifications that the functional characteristics of a wetland, such as hydrology, soils, or water quality, are affected.  Such modifications may include the placement of water control structures within a wetland, or at its outlet, which changes the natural hydrology established in that wetland.  As noted in the previous sections of this document, many functions of a wetland depend on its characteristic hydrologic regime.  Thus, changes in the hydrology of a wetland can result in changes in other functional attributes.   For example, impoundment of coastal wetlands may decrease water circulation; in some cases, circulation may become almost nonexistent (Devoe and Baughman, 1986).  As water circulation changes, water quality may change as well, which can result in changes in temperature, dissolved oxygen, salinity, pH, and nutrient levels of effluents from wetlands. Another impact noted in impounded wetlands includes increased sedimentation, which produces changes in the characteristics of the substrate and smothers vegetation (Copeland, 1974; Dean, 1975).

Although there are few research data available describing the impacts of impounding wetlands for stormwater treatment, it is likely that research from general studies on impounded wetlands may be used to predict potential changes in natural wetlands resulting from the combination of impoundments and increased stormwater inflows.  Stormwater, depending on the land uses within the watershed over which it flows, can vary in water quality and quantity.  This section begins with a description of how impoundments change the hydrology, water quality, soils, and wildlife/habitat of natural wetlands.  Then, by considering the typical constituents found in stormwater coupled with changes found when wetlands are impounded, the resulting impacts on wetlands that are impounded for treatment of stormwater runoff are discussed.

HYDROLOGIC CHANGES

Modifying natural wetlands with impoundments may result in radically different hydrologic regimes that are not ecologically sound (Frederickson, 1982).  Manipulating a wetland to enhance certain habitats or to attract certain species has been shown to degrade the wetland over a long period of time.  Managed wetlands often lack seasonal and long-term water level fluctuations (Frederickson, 1982).  The ability to vary water levels allows the depth of water in the impounded wetland, the length of the drawdown period, and the amount of exchange during flooding and drawdown to be controlled (Thompkins, 1986).  Studies have shown, however, that water circulation patterns within impounded wetlands appear to be responsible for many of the differences between these systems and natural wetlands.  The degree to which wetlands export carbon and nutrients is dependent in large part on the hydrologic characteristics of a system.  Changes in these characteristics as the result of impoundment or hydrologic manipulation can change this export.  Reduced circulation in impoundments can result in higher water temperatures and increased evaporation rates during the summer, as well as fluctuations in dissolved oxygen and salinity and other changes associated with water quality.  The manipulation of wetland hydrology can also directly influence the availability of aquatic habitat and indirectly affect invertebrates through the physiological responses of hydrophytes (Reid, 1982).  Therefore, an understanding of the amount and timing of water exchange is important to the success of these systems.

WATER QUALITY CHANGES

Although few studies directly relating stormwater inflow to water quality changes in impounded wetlands were identified in the literature, some comparisons between impounded and open wetlands can be considered.  Suspended particles are typically the pollutant of concern.  Sedimentation of suspended particles is one of the principal mechanisms of pollutant removal in a wetland.  The rate and degree of sedimentation are directly related to the flow characteristics of a particular wetland (Brown, 1985).  The ability to manage many of the flow characteristics within an impounded wetland can result in more efficient removal of suspended sediments in the system.  However, the loss of water storage capacity in impounded wetlands as the result of increased sedimentation (i.e., filling in of the impounded basin due to settling solids) must be considered to properly manage impounded wetlands used for stormwater control.  High influent concentrations of sediment from stormwater runoff can also reduce light penetration, reduce oxygen content, and affect the overall productivity of the wetland system (Stockdale, 1991).

Studies conducted in South Carolina tidal marsh systems indicated that the magnitude of nutrient and organic flux varied between different open marsh systems and between different open and impounded systems (McKeller, 1986).  The variation can be attributed in large part to the biological, geochemical, and hydrological characteristics of individual systems.  The differences in nutrient exchange between the impounded and natural tidal marshes were attributed in large part to the extent of water movement.  Summer  periods of restricted water flow in the marsh impoundments resulted in reduced tidal nutrient exchange.  Nutrient exchange in the open marsh system was at a maximum during the summer.  In another example, over a 6-year study period, a stormwater treatment marsh adjacent to Clear Lake in Minnesota was shown to remove 54 percent of total phosphorus from the influent water after a minimum of 3 days of detention time (Barten, 1986).  The sedimentation of particulate phosphorus was determined to be the primary removal mechanism in the system.  Table 13 shows the removal rates of phosphorus and other measured parameters by the impounded wetland over the 6-year period.  The impoundment of wetlands also removes acreage from open wetland systems and may diminish the overall export of detritus from the systems (Devoe and Baughman, 1986). Other differences in nutrient exchange were associated with basic biological and geochemical differences that characterize the impounded versus intertidal environments (McKeller, 1986).  Wetland impoundments in the South Carolina study were dominated by submerged benthic communities rather than open emergent intertidal marsh systems. These basic differences exert considerable control over the ways in which nutrients are processed and exchanged in wetland systems.

The impounded and open tidal marshes in the South Carolina study (McKeller, 1986) were both shown to export particulate and dissolved carbon.  The nature of the carbon export was not determined.  Differences in the quality of organic matter being exported are important in determining the overall impacts of impounded versus open wetlands on adjacent receiving waters (McKeller, 1986).

Salinity in impounded systems has been shown to fluctuate as the result of several factors, including reduced circulation, increased evaporation, and a lack of exchange.  Diked wetlands in the San Francisco Bay exhibit alternating periods of hypersaline and freshwater conditions in response to winter rains and summer evaporation.  Winter rains dilute and leach salts in the upper soil profile, and summer evaporation brings the salts to the surface (ABAG, 1991).  Conditions of varying salinity, pH, and oxygen in impounded wetlands also occur as a result of water control and management techniques (Wenner, 1986).  Decreased flushing rates in late summer can contribute to deteriorating oxygen levels within a system.  In addition, lack of vertical mixing due to water depth can result in the development of anoxic conditions in an impoundment.

Shallow impoundments have dissolved oxygen dynamics different from those of unimpounded streams or lakes.  In north central Minnesota shallow impoundments have been shown to lose much of their oxygen during ice-over (Verry, 1982).  Resulting low redox potentials cause massive migrations of nutrients out of the bottom sediments into the overlying water.  The enriched layer of water can encompass the entire depth of the impoundment (Verry, 1982).

Dissolved oxygen levels are also affected by the surges in organic matter that can occur in wetland impoundments.  Strong development of thermal stratification does not occur in shallow impoundments, and as a result the decay of organic matter can occur throughout the water column.  When organic matter is introduced into the system due to natural conditions or management practices, large fluctuations in dissolved oxygen can occur (Verry, 1982).  Figure 8 shows oxygen fluctuations in a shallow water impoundment in Minnesota over the period of a year.  During the summer, concentrations move between the upper and lower shaded areas rapidly as organic matter is processed and wind mixing or photosynthesis occurs.  Dashed lines depict areas of limited data.

Water temperatures in shallow impoundments have also been shown to vary compared to those of adjacent natural marshes and free-flowing streams in north central Minnesota. While minimum temperatures in impoundments were shown to be the same as those in natural systems, the maximum temperatures were as much as 5 degrees Celsius higher in the impounded wetlands (Verry, 1982).  The higher maximum temperatures were shown to be associated with surface-water-fed impoundments that were stagnant, with diminished depths and little or no water flow.  Decreased water depths and flow were associated with dry weather conditions or intentional management drawdowns (Verry, 1982).  The effect of higher water temperature in shallow impoundments on downstream water temperatures was also examined by Verry (1982).  Temperatures were shown to drop from 24.5oC to 20.5oC within 20 meters of a shallow impoundment outlet (Verry, 1982).  Temperatures remained the same farther downstream.  The rapid decrease in water temperature leaving the impoundment was attributed by Verry to streamside shading and groundwater influx. The difference in temperature downstream from the outlet was close to the temperature difference between the impoundment and the natural stream.

SOIL CHANGES

The increased capability to control flow characteristics and exchange in impounded wetlands can result in a more efficient removal of suspended sediment from the water column.  Increased sedimentation and management of water levels in impounded wetlands can affect the soils within the system.  Lack of daily flushing by tides in impounded wetlands in South Carolina resulted in a greater accumulation of organic material in the soils as compared to adjacent natural marshes (May and Zielinski, 1986).  Surface sediments within the tidal wetland impoundments ranged from silty clays to clayey silts. Sediments that accumulate in impounded wetlands may contain a higher percentage of fine-grained particles because of decreased energy levels and increased water retention time.  The higher percentage of finer-grained sediments affects the textural characteristics of the soils developing in the system.

Suspended organic and inorganic materials have a tendency to adsorb pollutants, such as heavy metals, nutrients, hydrocarbons, bacteria, viruses, and refractory organics (Stockdale, 1991).  These materials may then be deposited with the sediments, affecting the overall characteristics of the soils in the impoundment.  Toxicants are generally found to be associated with finer-grained particles that are less than 246 microns (Oberts, 1977).  Impoundments that cause sedimentation of finer-grained particulates would result in the incorporation of these toxicants into developing soils.  Bioaccumulation of contaminants by fish and wildlife could occur as a result of the buildup of materials in the soil.  Improper management techniques in impoundments used for stormwater runoff could result in the reintroduction of these toxicants to the water column  during turbulent conditions associated with storms or high-flow events.

High percentages of silt-sized particles in combination with low sedimentary flushing of smaller particle sizes can result in decreased oxygen levels in impounded wetlands. Oxygen depletion can result and in brackish or saltwater environments may be accompanied by the accumulation of sulfides (Wenner, 1986).  The accumulation of organic material can also result from oxygen depletion.  Management of water levels in impounded wetlands can also cause leaching and oxidation of marsh soils.  If soils are not kept moist, sulfides can become oxidized to form sulfuric acid and cat clays (Wenner, 1986).  The development of acid sulfate soils or cat clays can result in a soil pH of 3.5 or less.  Soil samples taken in a brackish marsh impoundment that had been dewatered on South Island in South Carolina had  pH values ranging from 3.2 to 8.3, depending on whether the soils were kept wet or allowed to dry (Wilkinson, 1970).

BIOLOGIC/HABITAT IMPACTS

Changes in the types and diversity of vegetation in wetlands have been shown to occur as the result of the impoundment of these systems.  Additional changes in vegetation could be expected as the result of stormwater discharges to impounded wetland systems.  As mentioned above, plant communities and individual species appear to be affected by water depth, frequency and duration of flooding, and water quality.  Studies in impounded marshes along Florida's east coast showed that excessive or prolonged flooding in wetland impoundments resulted in the stressing or killing of existing high marsh vegetation in the systems (Carlson and Carroll, 1985).

Another basic change associated with the impoundment of intertidal marshes is the conversion from a wetland dominated by emergent vegetation to a system dominated largely by submerged macrophytes, benthic algae, and phytoplankton (Kelly et al., 1986). Although total community production in managed wetland impoundments in South Carolina was shown to be similar to total production in adjacent open marshes, the contributions to productivity of the various plant communities (marsh grasses, benthic macrophytes and macroalgae, and phytoplankton) were shown to differ considerably between the impounded and open marshes (Marshal and McKeller, 1986).

Wilkinson (1970) conducted studies on a newly flooded brackish impoundment on South Island in South Carolina to determine vegetative succession in the system.  Water depths in the wetland impoundment ranged from 12 to 24 inches.  During a 3-year study period, the relative abundance of some species changed drastically with the distribution of plants into zones associated with water depth.  Ruppia maritima, a submerged aquatic grass, became the most successful plant after flooding.  Scirpus robustus was the most successful emergent plant.  Distichlis spicata, a salt grass associated with higher portions of salt marshes, decreased in abundance after flooding and eventually disappeared from the impoundment; Spartina cynosuroides, a shallow water emergent, was reduced in area of coverage to the very shallow margins of the impoundment (Wilkinson, 1970).

In freshwater impoundments in South Carolina where water levels are maintained, floating and submergent species have been shown to become the dominant vegetation in succession.  The dominant species vary according to water depth, but Utricularia, a submerged aquatic plant, Lemna, a floating aquatic plant, Nymphaea, a floating leaved aquatic plant, and Ceratophyllum, a submerged aquatic plant, are usually the most common species (Miglarese and Sandifer, 1982).

Changes in salinity associated with the impounding of salt marshes also result in changes in vegetation patterns.  Restriction of tidal inundation resulting from the impoundment of a 20-hectare tidal marsh in Stonington, Connecticut, resulted in a succession from a Spartina-dominated marsh to one dominated by Phragmites australis and Typha angustifolia.  Lower soil water salinity associated with tidal restrictions in the study area resulted in a change in vegetation from salt marsh to an emergent freshwater wetland to a brackish water wetland (Sinicrope et al., 1990).  Brackish conditions resulted from an attempt to control Phragmites by raising salinities in the impoundment.  Similar or more rapid changes could occur in the vegetation of salt marsh impoundments with the introduction of stormwater discharges to these systems.

Changes in the hydrologic character of impounded versus open wetland systems can result in a depletion of fauna. The introduction of stormwater runoff or water control objectives, resulting in hydrologic disturbances in impounded wetlands, could result in the development of stressful habitat conditions.  Since a limited number of species can adapt to conditions of changing salinity, pH, temperature, and dissolved oxygen, low species richness could result.  The lack of interchange between impounded wetlands and adjacent waters could also result in a reduction of species richness due to the inability of many fauna to access the impounded wetland.  Studies have shown that wetland impoundments can significantly affect the diversity and richness of fish species in comparison to adjacent natural wetlands (Devoe and Baughman, 1986).

Studies to determine fish species diversity were conducted on a salt marsh impoundment in Indian River County, Florida, in 1979.  Prior to the study the impoundment was managed for mosquito control.  During the study period the impoundment did not receive pumped estuarine water and the water levels in the impoundment were allowed to fluctuate with weather conditions.  Initial surveys conducted on the impoundment indicated that at least 11 species of fish were present in the impounded marsh (Gilmore et al., 1981).  Arid conditions during the study period resulted in hypersaline conditions and the dewatering of large portions of the impoundment.  Salinities ranged from 2 to 125 ppt, water temperatures were from 14 to 34 oF" , and oxygen levels varied from 1.2 to 14.4 ppm (Gilmore et al., 1981).  The number of species collected in the impoundment after dewatering was reduced to four.  Figure 9 shows the monthly distribution of fish in relation to salinity, rainfall, and water levels in the impoundment during the study period (Gilmore et al., 1981).  Fluctuations in rainfall, salinity, and water depth during the month of September were the result of Hurricane David.

Earlier studies conducted on the marsh by Harrington and Harrington in 1966 prior to and 30 months after initial impoundment showed a reduction in the number of fish species from 16 to 5.  Studies also indicated a change in feeding habits to a reliance on plant materials by three of the remaining species in the impoundment (Harrington and Harrington, 1982).

Shallow impoundments in north central Minnesota with reduced or stagnant water flows were determined not to be well suited for fish populations during the summer and over winter because of rapid and wide fluctuations in dissolved oxygen levels (Verry, 1982). Maximum water temperatures in the impoundments were also shown to be above the upper level for normal trout growth.  Maximum temperatures in wetland impoundments with flowing water did not exceed maximum growth temperatures for several other species of fish in the region including walleye, yellow perch, large-mouth bass, northern pike, carp, shad, crappie, white perch, spotted bass, white bass, and catfish (Verry, 1982).

Potentially rapid or large changes in water temperature associated with stormwater inflow could cause thermal stress to fish in shallow impoundments.  Impacts to aquatic insects resulting from temperature fluctuations in a system are possible because of their general inability to compensate for or acclimate to the temperature changes.  Fluctuations in water temperature regimes of from 2 to 3oC could potentially eliminate some sensitive species (Galli and Dubose, 1990).

Changes in the pH of water in wetland impoundments associated with management practices or the introduction of stormwater can also affect the biota in impounded systems.   Most organisms are adapted to function within particular pH ranges, and abrupt or substantial variations in pH can have adverse effects on aquatic life usually in the form of reduced productivity and increased mortality (Newton, 1989).  Most urban stormwater is slightly acidic.  The variable nature of stormwater inflow could result in abrupt changes in the pH of an impoundment.  Lowered soil water pH associated with drawdown in impounded brackish or saltwater marshes can affect densities of invertebrates such as molluscs and crustaceans.  Species that depend on alkaline conditions for shell development may be affected if low pH occurs at the sediment or soil-water interface (Wenner, 1986).

The use of impounded wetlands by water birds has been shown to be high in several systems.  Studies conducted on South Carolina impoundments indicated that high water bird use was directly related to season, management practices, impoundment size, and availability of resources (Epstein and Joyner, 1986).

Newly impounded brackish wetlands on South Island, South Carolina, were studied by Wilkinson (1970) over a 3-year period to determine plant succession and waterfowl use.  Five impoundments with different hydrologic controls--fully flooded, slowly rising, tidal, saturated soil, and dry--were observed.  Use of the impoundments by waterfowl was rated based on the estimated number of observed waterfowl.  Observed numbers of 1 to 10 were rated as poor, 10 to 30 as fair, 30 to 60 as good, and above 60 as excellent (Wilkinson, 1970).  Waterfowl observations were made twice a week during the fall and winter.  The fully flooded impoundment was the most used by waterfowl.  Use of the wetland impoundment with rising water levels was rated good, and use of the tidal impoundment was good to poor.  Use of the impoundment with saturated soil was rated as fair to poor, and use of the dry impoundment was rated as poor (Wilkinson, 1970).

In Minnesota, Wisconsin, and Michigan, surveys of impoundments indicated that after initial flooding the diversity and density of birds increased due to increased edge, productivity, nest cavities, and perch sites (Rakstad and Probst, 1982).  The increase in amount of edge and degree of interspersion of habitat types also resulted in use by greater numbers and kinds of wildlife including muskrats, racoons, red fox, river otter, mink, and water shrew (Rakstad and Probst, 1982).  After several years, however, the density and diversity of wildlife has been shown to have decreased in many impoundments.  This decrease has been shown to be due in part to vegetative succession in the impoundments.

Some management techniques applied to wetland impoundments have been shown to be successful in maintaining or enhancing use by wildlife in several cases.  The water quality salinity and hydrology requirements of different fish and wildlife species vary, and therefore management techniques applied to wetland impoundments to increase or enhance habitat for one species may have adverse impacts on others (Hynson et al., 1985).

REGIONAL DIFFERENCES

Regional differences that affect impounded wetland systems are similar to those that affect natural wetlands.  The methods, timing, and period of drawdowns depend largely on the geology, hydrology, soils, and climate of an impoundment site.  For example, soils in arid regions with low rainfall tend to accumulate salts in their upper profiles.  As a result, drawdowns or evaporation in arid-region impoundments can result in the development of hypersaline conditions.  Such conditions would be less likely to occur in humid regions.  In addition, northern regions are more likely to be affected by the ice-over of impoundments in winter than are southern regions.  These regional and site-specific characteristics, in addition to others, all exert controls on the inflow, outflow, and quality of water in an impoundment.

SUMMARY

Shallow-water impoundments have been shown to be both potentially beneficial and potentially detrimental to the functions of the impounded wetland systems.  The increased ability to manipulate the hydrology in impoundments (i.e., water levels and flow) allows management techniques to be designed to enhance or control specific aspects of the systems.  For example, water levels can be controlled to enhance the growth of certain vegetative species and in turn attract certain waterfowl or wildlife.  Flow within the impoundments can be controlled to promote increased sedimentation of pollutants from inflowing stormwater.  However, altering the hydrology in a natural system by impoundment or through the management of impounded systems can change the functional processes of the system.  As mentioned, techniques applied to impoundments to enhance or control one aspect within the system can result in adverse impacts to others. Changes in the characteristics of the hydrology, water quality, soil, vegetation, and fauna in the impoundment can result.

As the result of urbanization, in many areas low- to moderate-intensity storms can produce large volumes of runoff.  Because of the variable nature of stormwater runoff flow, the ability of impounded wetlands to remove nutrients, suspended solids, and heavy metals may vary by season, from storm to storm, or within the same storm (ABAG, 1991). Impoundments may act as a sink for the constituents of stormwater under certain conditions or as a source under others.  Variations in the characteristics of stormwater inflow will also have varying impacts on the components of impoundments.  Changes in the characteristics of the soil, water quality, and hydrology in impoundments will occur and, in turn, will affect the biota in the impounded wetland.  The potential bioaccumulation of pollutants for fish and wildlife as the result of stormwater inflows remains unclear (Meiorin, 1986).  The effects of impounding wetlands and manipulating impoundment conditions, along with the potential impacts of stormwater discharges on the characteristics of the soil, vegetation, water quality, and fauna in the systems, need to be further studied.

UNRESOLVED ISSUES

Because the use of natural wetlands for stormwater management purposes is relatively new, considerable uncertainty exists concerning the impact of the quantity and quality of stormwater runoff on natural wetlands.  Several issues related to this topic are presented below.

Several unresolved issues were raised at the January 1992 workshop in Clearwater, Florida.  These include:

UNRESOLVED IMPOUNDMENT ISSUES

As mentioned, the potential use of wetland impoundments for the enhancement of treatment of stormwater runoff has been considered as an option for stormwater management.  Several information needs related to this practice (in addition to those discussed in the previous section) are listed below: