Institute for Urban Environmental Risk Management

Marquette University, Milwaukee WI 53201 - 1881

TECHNICAL REPORT NO. 2

WATER QUALITY, ECOLOGICAL, AND FLOOD RISKS TO

RECEIVING WATERS DUE TO URBAN RUNOFF AND
URBANIZATION

Neal O’Reilly, MSc

Vladimir Novotny, PhD, P.E.

Primary Investigator

Project sponsored by the Grant No. R82-5759 to Marquette University from the
USEPA/NSF/USDA STAR Watershed Program

Barbara M. Levinson
EPA Project Officer
 Acknowledgment

The research contained in the technical report was partially sponsored by the US Environmental
Protection Agency/National Science Foundation/US Department of Agriculture STAR Watershed
Program by a Grant No R82-5759 to Marquette University and by the university’s scholarship to the
author. The authors greatly appreciate this support. The findings and conclusions contained in this
report are those of the authors and not of the funding agencies nor the STAR program.

In addition to the Primary Investigator on the project, Dr. Vladimir Novotny, the first author would
like to acknowledge the assistance of Professors C. J. Crandall and D. Zitomer from Marquette
University for the review of his essay. The author would also like to thank his wife Bonnie who also
helped in preparation of this document.
 Table of Contents

Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i

List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii

List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii

Chapter 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Changes of Receiving Water Bodies Due to Urbanization . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

The Stream as an Ecosystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Integrity of Receiving Waters and Watershed Management . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Chapter 2 Hydrological/Morphological Changes By Urbanization . . . . . . . . . . . . . . . . . 7

Change of Channel Morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Sediment and Sedimentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Stream Bank Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Habitat Integrity Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Impacts of Stream Channelization and Armoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Impact of Floodplain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Effect of Bridge, Culverts and Other Channel Constrictions . . . . . . . . . . . . . . . . . . . . . . . . . 22

Chapter 3 Stream Corridor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Chapter 4 Ecological Impact of Urban Stormwater . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Impact of Urbanization on Aquatic Habitat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Impact of Urbanization on Biotic Communities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Water Quality Impacts of Urban Runoff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Elimination of Riparian Vegetation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Summary of the Ecological Impacts of Urban Stormwater Runoff . . . . . . . . . . . . . . . . . . . . 38

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

i
 List of Figures
Chapter 1

1.1 Land use change and stresses on aquatic ecosystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.2 Stream channel and flood plain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 High flow–flood plain schematics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

Chapter 2

2.1 Hydrologic effects of urbanization on a small stream (Mud Creek, Appleton) . . . . . . . . . 7

2.2 Impact of urbanization on peak stormwater discharges . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3 Relationship of flows to the recurrence interval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.4 Channel stability as a function of imperviousness and flow increase in pacific
north-west US . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.5 Idealized natural channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.6 Bank erosion failure process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.7 Hydraulic effect of bridges on upstream flooding caused by channel narrowing . . . . . . 22

Chapter 3

3.1 A cross-section of a river corridor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

3.2 Food web schematic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.3 The river continuum concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Chapter 4

4.1 Relationship of macroinvertebrate diversity to watershed imperviousness . . . . . . . . . . . 33

4.2 Relationship of population density to the stream Index of Biotic Integrity . . . . . . . . . . . 34

ii
 List of Tables
Chapter 2

2.1 Channel enlargement multipliers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

2.2 Relation of hydrololic stream parameters to physical integrity and habitat conditions . . 19

Chapter 3

3.1 Physical, chemical and biological components that influence stream ecological
Integrity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

Chapter 4

4.1 Water quality characteristics of urban runoff identified in the national urban runoff
program study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.2 Percent of occurrence of priority pollutants in urban runoff in Milwaukee and
Madison, Wisconsin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.3 Pollutants found in the bottom sediments of three Milwaukee streams . . . . . . . . . . . . . 36
4.4 Summary of impacts of urban stormwater runoff on stream ecosystems . . . . . . . . . . . . 39

iii
 CHAPTER 1

INTRODUCTION

Changes of Receiving Water Bodies Due to Urbanization

Urbanization represents the most profound global change of the terrestrial and aquatic ecosystems.
Urbanization itself is only the final phase of the land use transformation process that begins with
conversion of the native lands - prairie, forest, wetland or arid land. Processes that transform the land
and affect the aquatic ecosystems include (Figure 1.1) drainage, irrigation, urban development with
associated construction erosion and pollution discharges, and many other.

Figure 1.1. Land use change and stresses on aquatic ecosystems (from Novotny and Olem, 1994)

Urbanization and land use modifications caused by urbanization are the last steps in the land
modification process that started many years or even centuries ago by reclamation of virgin lands
by man.

1
Urbanization has numerous effects on urban receiving water bodies. The most important impact is
the alteration of the hydrological cycle and rainfall-runoff transformation process. By urbanization,
watersheds become more impervious, resulting in a dramatic shift of flow components and water
origin in the streams. In a hydrological situation of watershed located in Midwestern and Eastern
United States, streams that originally (pre-development stage) received significant portion of flow
from groundwater-base flow contribution have been hydrologically and morphologically changed
into drainage channels that receive most of their flow from surface runoff.

The increased magnitude and frequency of high flows has several major adverse effects on
floodplain, stream channel and on the ecology of the urban stream. The hydrologic effects include:
(1) floodplain enlargement; (2) increase of the frequency of flooding inside the floodplain; (3)
increase of peak flows during storm events; (4) increase of the magnitude and frequency of all runoff
events of all sizes; (5) as a result of increased medium floods channels become unstable and more
erosive (degrading); (6) imperviousness impedes recharge of shallow groundwater aquifers, which
diminishes base flow contributions (after urbanization some streams may become ephemeral or
effluent dominated); (7) more flow moves on the surface with a faster velocity which increases the
volume of surface runoff contribution.

The most important adverse ecological impacts are: (1) loss of bank habitat by increased stream bank
erosion and channel alteration to accommodate increased flows; (2) siltation of the channel by
increased sediment loads; (3) water column and sediment contamination by pollutant discharges
from point and nonpoint sources; (4) increased temperature due to more warmer surface flows and
loss of stream shading by vegetation; (5) loss of pollution intolerant species.

The effects of urbanization are not only limited to the stream channels. The entire stream corridor
consisting of the stream channel and floodplain must be considered in a comprehensive analysis of
urban streams. The major impacts of urbanization on integrity of urban streams can be divided into
three categories:

I. Hydrologic/hydraulic changes of flow regime and their effect on stream morpholgy.

II. Water and sediment quality degradation and their effects on composition and survival of
aquatic species.

III. Ecological/habitat changes, including modifications of channels and floodplain.

Larger urban settlements are almost always located near rivers. Rivers provide water for living and
economic processes and provide, in some cases, means for transportation. The processes comprising
urbanization that affected the stream progressed in several steps (Figure 1.1):

I. Land reclamation by deforestation, and/or conversion of lands to urban landscape, including

building roads, drainage, sewers, filling wetlands and floodplains, etc.

II. Construction activities and regrading.

2
III. Use of developed lands, runoff conveyance and disposal of wastes.

IV. Engineering works to cope with increased flooding caused by urbanization, storm runoff
drainage, sewerage, and reclamation of floodplains for development.

Urban engineers and developers recognized the adverse effects of urbanization from ancient times.
Urban streams were straightened, channelized, diverted, covered, dredged, lined with concrete,
stream bank vegetation was cut down, and dams were built to provide head for energy production
and running mills. The changes from a natural unimpacted stream to a fully urbanized channel took
many years, however, most profound changes occurred in the twentieth century.

This report will focus on the impacts of urbanization on the overall ecological system of receiving
streams.

The Stream as an Ecosystem

A stream ecosystem is a corridor that often is well defined. Stream corridors have diverse
characteristics that vary both in longitudinal view, from head waters to river mouth, and laterally,
from stream channel outward through the floodplain. The stream has also spatial scales ranging from
regional, landscape, stream corridor, stream and individual reach. Hydro-morphologically, stream
are classified numerically from first order headwater streams to higher order downstream sections.
A second order stream has at least one first order tributary, a third order stream has one or more
second order tributaries , and so forth. The stream corridor is an ecosystem that usually consist of
three major components (Figure 1.2):

! Stream channel
! Floodplain
! Upland floodplain fringe

Main stream channel is typically well defined channel and, in perennial streams, contains a base
flow channel called thalweg (Dunne and Leopold, 1978) that carries flow all the time and adjacent
gravel/sediment filled portions of the channel that carry some wet weather flows. In stable rural
streams, bankfull or greater flows take place at a frequency ranging from 1.2 to 2.4 years (Wolman
and Leopold, 1957; Leopold, Wolman and Miller, 1992; Carlson, 1965; Dury, 1973, Annable, 1996).
A channel that does not have a permanent flow is called ephemeral. A perennial channel receives
permanent base flow contributions. Perennial streams are typically fed by groundwater discharges
and, in urban areas, by dry weather effluents. However, in some arid areas, former ephemeral streams
became perennial after urbanization by discharges of treated and untreated effluents (for example,
Las Vegas Wash, Los Angeles River). Such streams are classified as effluent dominated. Some
former perennial stream became ephemeral or effluent dominated when, as a result or urbanization,
flow was withdrawn from the stream and none was left during low flows or when shallow
groundwater table due to urban and/or agricultural groundwater use dropped below the elevation of
the stream channel (for example, in Tucson, Arizona).

3
Floodplain is the land adjacent to the channel that is generally without flow during dry weather
periods but has flow or standing water on occasions during wet weather. Two types of floodplain
have been defined (Federal Interagency Stream Restoration Working Group, 1998):

! Hydrologic floodplain – the land adjacent to the base flow channel residing below bankfull
elevation. The base flow channel is also refereed to as the thalweg (Dunne and Leopold,
1978).
! Topographic Floodplain – the land adjacent to the channel including the hydrologic floodplain
and other lands up to an elevation reached by a flood peak of a given frequency (for example, the
100-year floodplain).

Figure 1.2. Stream channel and flood plain

The hydrologic floodplain is formed by frequent bankfull flows. As stated in the preceding section, in
stable rural streams, bankfull or greater flows take place at a frequency ranging from 1.2 to 2.4-years.

The extent of the topographic floodplain has been (more or less arbitrarily) limited to a borderline that
is reached by flood flows on average once in hundred years. This is so called 100 - year floodplain.
Similar definition can be extended to 10, 25, 50, or 500 year floodplains. Using recurrence interval in
years to define the floodplain is misleading to public who expect no flooding for 100 years after a major
flood.. A better definition can be obtained by expressing the floodplain extend in terms of risk of flooding

4
which is a probability that a location away from the stream will be flooded at any given year. A risk
(probability) is reciprocal of the recurrence interval, i.e., r(x) = 1/T. This means that a point located at the
boundary of a 100 floodplain has a risk of flooding of 1/100 = 0.01, or 1 percent.

A flood flow, defined as a flow that exceeds the capacity of the channel, is divided into three hydraulic
parts (Figure 1.3): (1) a high flow and high velocity flow in and above the stream channel; (2) slower
velocity flow in a floodway, and (3) mostly stagnant water in the floodway fringe zone. The flat floodway
zone that often surrounds the channel is flooded frequently. It contains pools, riparian wetlands and
temporary or permanent ponds. This riparian floodway zone has a great ecological significance and
flooding is an essential part of the stream ecology (see Chapter 2 for more discussion)

Figure 1.3. High flow - flood plain schematics ( based on FEMA).

During a flood event, the stream channel and floodway carry most of the flow. Although damages to
habitat may occur during flooding, sometimes to an extreme of forming an entirely new channel,
ecological recovery is rapid and, to some degree, floods are beneficial to the stream ecology. High
concentrations of suspended material that the flows may carry during rare flooding events are also of little
relevance to the integrity of the stream. Deposition occurs mostly in the floodway fringe and in large
oxbow ponds and wetlands. However, because of the flood plain encroachment by increased flows
caused by urbanization, real estate properties that were originally located outside of the flood plain are
now found to be in the flood plain fringe. The cost of clean up of the sediment from the homes located
in the floodplain after a major flood may be considerable.

Any stream is a dynamic system that is undergoing constant changes. At any given moment the
morphology and ecology of the stream system is in an equilibrium with boundary (allochthonous) and
internal (autochthonous) stresses. Example of an internal stress is the shear stress of the flowing water
that acts on the bottom and banks and erodes particles or lack of turbulence in quiescent reaches and
impoundments that results in deposition. External stresses are discharges of stormwater and wastewater

5
into the stream, withdrawals of water for power production and water supply. Many kinds of engineering
works that modified the streams should also be included as a stress.

Streams respond to these stresses by modifying their morphology, water and sediment quality, and
habitat/ecology. The erosive processes form the valley and the channel over the geological times. Man’s
actions can adversely modify the stream in a period of decades.

Integrity of Receiving Waters and Watershed Management

The term watershed management has a broad meaning and has been used in various contexts for more
than one hundred years. The management is always connected to its objectives. The first objectives of
watershed management focused primarily on providing water in adequate quantities to various users
(e.g., irrigation, water supply, navigation, hydro power, etc.) throughout the watershed and/or flood
control. In typical quantity watershed management, water flows are manipulated in time and space by
storing water during the times of hydrological excesses and supplementing flows during the times of
hydrological shortages in places of water use. The tools of quantity watershed management include
storage reservoirs, river modifications, and water distribution and transfer systems.

Until about twenty years ago, the major and often the single objective of management of urban streams
was drainage and flood control to accommodate the increased flows caused by urban development. The
remnants of these management objectives in many urban areas are streams that have been straightened,
lined or covered but devoid of a balanced aquatic life..

Under today’s philosophy and ensuing laws and regulations, preservation and restoration of ecological
integrity of receiving water bodies is the main objective of watershed management and restoration.
Further more, the antidegradation rule of the present regulations does not allow downgrading the
integrity of the receiving water bodies even when the objective is drainage or flood control. Therefore,
the sometimes conflicting concurrent objectives of drainage/flood control and restoration of ecological
integrity of urban streams of urban stream projects must be reconciled.

EPA defines “ecological integrity” and “the condition of the aquatic community inhabiting unimpaired
water bodies of a specific habitat as measure by community structure and function”. The ecological
integrity of a stream is made up of the relationship of three components : physical, chemical and
biological (Plafkin et al., 1989). Physical integrity is achieved when the stream provides the needed
physical habitat structure for the indigenous group of organisms. Chemical integrity involves providing
the needed water and sediment quality for the organisms such as dissolved oxygen, nutrients, and a lack
of toxic compounds. Both physical and chemical integrity affect the biological integrity that involves
the proper composition and densities of the organisms corresponding to unimpacted reference reaches
within the same ecoregion. For organisms to survive they rely on the interactions with other organisms,
or biological integrity. These interactions include provisions of food, symbiotic relations between
organisms and needed competition that helps keep the system in balance.

6
 CHAPTER 2

HYDROLOGICAL/MORPHOLOGICAL CHANGES BY
URBANIZATION

As watersheds are urbanized, the surface of the landscape changes. As forests, prairies, wetlands and
agricultural fields are converted to urban landscape, the hydrology of the watershed is modified. In
undeveloped watersheds, rainwater and snowmelt are captured by vegetation and released by evaporation
back to the atmosphere (interception storage), and/or allowed to infiltrate into soils and subsequently
recharge the aquifer or is released back to the atmosphere by vegetation in a form of transpiration. The
consequence of urbanization is more impervious area (roads, roofs, driveways, parking lots, etc.) within
the watershed . The hydrologic urban (urbanization) allochthonous stresses and their effects on streams
and resulting impacts are:

1. Floodplain enlarges. Figure 2.1, shows that as the magnitude of the 100 year flood increases,
areas that were outside the 100 year flood plain would become a part of it.

Figure 2.1. Hydrologic effects of urbanization on a small stream located in

east-central Wisconsin (Mud Creek near Appleton) with soils of
medium permeability (SCS hydrologic category B and C). The
flows were calculated by the SCS TR-55 runoff curve model.

7
2. The frequency of flooding inside the floodplain increases. Under natural conditions, a river
channel can hold a flow with a recurrence interval of 1.2 to 2.4 years (Leopold, Wolman and
Miller, 1992). Figure 2.1 documents that the channel of the Mud Creek, as a result of
urbanization, could be presently and in the future overtopped several times each year.

3. Hydrological losses such as surface storage, evaporation and transpiration are decreased by
development which increases the flow volumes.

4. Since surface flow moves faster the time of concentration is decreased. As a consequence, peak
flows during storm events are increased.

5. Magnitude and frequency of all runoff events of all sizes increases. This is especially important
for rainfalls of smaller and medium magnitudes. Before urbanization these smaller rainfalls
mostly infiltrated into soil and the flows in the stream were smaller and could be easily contained
in the natural channels of the stream. After urbanization the same medium rainfall could results
in a flood.

6. As a result of increased medium floods channels become unstable and more erosive (degrading)
(Leopold, 1994; Booth and Jackson, 1997). This has an adverse impact on habitat. The result of
this degradation is a wider channel.

7. Imperviousness of the watershed impedes recharge of shallow groundwater aquifers. This

diminishes the base flow contributions (Masterson and Bannerman, 1994). After urbanization
some streams may become ephemeral (without base flow) or effluent dominated.

8. More flow moves on the surface with a faster velocity. This increases the peak rate and volume
of the surface runoff contribution.

9. Sedimentation. Urban runoff caries more polluted sediment originating from construction sites,
street surfaces and combined sewer overflows.

Impervious surface in urban areas dramatically increases surface runoff during storm events. The changes
affect both the rate and volume of flow. Schueler (1994) claimed that depending on the degree of
imperviousness and soil type, annual volume of runoff of fully developed urban watersheds can increase
2 to 16 times the pre-development rate (see Figure 2.1 for example for east-central Wisconsin.).

While impervious watershed surfaces affect runoff rates and volumes, the degree of connectivity to the
drainage system also plays a role. Impervious surface that are directly connected to the drainage system
produce more downstream flow than impervious surface that discharge onto pervious areas. When
impervious surfaces such as roof tops or parking lots are allowed to discharge onto pervious landscaped
surfaces, a portion of the runoff can infiltrate into the ground and be held in surface storage and
subsequently evaporate. When the runoff from an impervious surface discharges directly into the drainage
system, little opportunity is provided for infiltration.

8
Effects of the Drainage System on Peak Runoff Rates
The type of drainage system also affects the rate and volume of runoff. The drainage system, in urban
areas can be a mixture of grass swales, open channels, and storm sewers that drain into one another.
A grass swale is a wide open channel. Vegetation within the swale helps to retard the velocity of flow and
decreases downstream peak flows (Thayer and Westbrook, 1990). Properly designed swales can facilitate
infiltration, resulting in loss of water from the surface runoff. Infiltration in a swale is dependent on the
soil type, compaction during construction, and degree of pore space clogging caused by fine materials
being discharged into the channel.

Large open channels provide limited opportunities for infiltration during storm events due to the large
ratio of cross-sectional area to wetted perimeter. Vegetation along the channel can provide roughness that
impedes flow and reduces velocities. Channel with hard linings such as stone or concrete increase the rate
of flow. Storage can be provided within then open channels that decreases downstream [peak flows.

Closed conduit systems of storm sewers and culverts provide drainage components that convey
stormwater in a rapid fashion. The small roughness coefficient for pipe materials such as concrete
(Mannings n= 0.013), corrugated steel (n = 0.024) and plastics (n = 0.01) makes them ideal for conveying
stormwater at rapid rates (Ferguson, 1998). However, efficient fast conveyance reduces the time of
concentration at which runoff reaches the stream. Decreasing the time of concentration increases the peak
rate of flow downstream. Figure 2.2 illustrates the results of urbanization on several hydrologic
components for a theoretical 40 - ha watershed (Walesh, 1989).

Figure 2.2. Impact of urbanization on peak stormwater discharges. Flows

were calculated by the Rational Formula (Walesh, 1989)

Change of Channel Morphology

9
In natural streams, the morphological characteristics are in a quasi equilibrium with the hydrology and
geology. Leopold, Wolman and Miller(1992) documented that channels of rivers in eastern and
Midwestern US have a channel capacity that can contain a flow (a bankfull stage) that has an
approximate recurrence interval of about 1 ½ years as shown on Figure 2.3. This figure is essentially
a channel rating curve normalized by the bankfull capacity flow and bankfull depth obtained from

Figure 2.3. Relationship of flows to the recurrence interval. Flow and

depths are normalized by the bankfull flows and depths
(Leopold, Wolman and Miller, 1992).
USGS and
other observations in the eastern US. In their observations the authors noted that the
recurrence interval of bankfull flows ranged from one to 1.9 years. Leopold, Wolman and Miller noted
that “there is a remarkable similarity in the frequency of bankfull stage on a variety of rivers in diverse
physiographic settings and differing greatly in size”.

Analyses of the data from Western humid regions of the US (Oregon and Washington) by Williams
(1978) more or less confirmed this finding but the author pointed out a large scatter of the recurrence
intervals around the mean of 1 ½ years (80 percent of intervals were within the range from about 1.0 to
3 years with a geometric mean of 1 ½ years). Consequently, floods, defined as a flow exceeding the
bankfull flow occurs in natural channels on average once in two years.

While the streams are in some state of instability, a major problem with determining the factors that
influence the morphological characteristics of urban streams is the fact that most are in a very unstable

10
condition unless dramatic engineering modifications were made in the past to stabilize the channel. Even
then such measures, such as lining with concrete, have only temporary effect.

Due to the fact that urbanization and suburban development in North America is progressing at relatively
fast pace many streams have not reached their equilibrium with the changing and increasing hydrological
stresses and are still reacting to the changes in the watershed. The percent of the population of the United
States that resides in metropolitan urban areas increased from 53 % in 1940 to 77% percent in 1990 (U.S.
Bureau of Census, 1991). In the same time the US population increased from about 200 million to 300
million at the end of this millennium. The increased urban population has resulted in more land converted
to urban use. Most urban streams have not had enough time to adjust to the new conditions. Other factors
such as stream channelization, water diversion, and dams have also affected the ability of urban streams
to reach their quasi stable conditions ( Allan, 1995; Collier and Schmidt, 1996)

Urban communities, in the past and today, have tried to restrict the enlargement of eroding channels by
armoring and lining the banks or even the entire channel. In other places, channel straightening, diking
and enlarging was implemented. In extreme, cases, urban streams were diverted into enlarged closed
conduits and put away from the sight. While these practices may slow down or eliminate bank erosion
and locally reduce flooding they do not provide conditions for habitat preservation and propagation of
aquatic biota.

To accommodate the increased high stream flow caused by urbanization, streams respond by increasing
their cross sectional area. This is done by stream bed downcutting, stream widening, or a combination
of both. Robinson (1976) in a study of eight watersheds in Baltimore, MD found that urban streams have
channel cross-sections areas approximately twice those or rural streams and width-to-depth ratios about
1.7 times those of rural streams. Other studies have found that cross-sectional area increases by a factor
of 2 to 5, depending on the degree of impervious cover in the watershed and the age of the development
(Arnold et al., 1982; Gregory et al., 1992; and Macrae, 1996). Stream channels react to increased urban
runoff not only by adjusting their widths and depths, but also by changing their gradients and meander
pattern (Riley, 1998).

As urban streams widen their cross-sectional area to accommodate increased peak flood flows, it also
affect the wetted perimeter covered by flowing water during wet weather flow which decreases the
velocity and capacity in the channel even more. In the same time, the base flow is diminishing;
sometimes, urban streams become ephemeral. Simmons and Reynold (1982) noted that dry weather flows
in urban watersheds of Long Island, NY, were reduced by 20 to 85 percent after development. The end
result in many urban areas is a very shallow low - flow channel that wanders across a wider stream bed,
often changing lateral position in response to storms. The smaller wetted perimeter combined with the
smaller and shallower base flow can effect stream habitat for aquatic organisms.

During the process of increasing the channel cross-section, active erosion takes place. Channel erosion
can be a major component of the annual sediment discharge of urban streams. In some urban streams,
channel erosion has been documented to constitute as much as 75 percent of the total sediment load
(Crawford and Lenat, 1989; Trimble, 1997). The higher flow conditions in urban streams make them
more capable of moving sediment than before (Wolman, 1964). Many urban channels have coarser
channel bottom material than rural streams (Robinson, 1976).

11
Both (1991); Both and Reintl (1993) and Booth and Jackson (1997) analyzed channels stability in the
Seattle (WA) metropolitan area east and south east of lake Washington. The watersheds ranged from 38
km2 to 181 km2 in area (14.7 to 70 sq mi). They noted that channels remain relatively stable with good
habitat conditions when the degree of imperviousness of the watershed is less than 10 percent (Figure
2.4). When, as a result of urbanization, the impervious portion of the watershed exceeds the 10 percent
threshold the channel become unstable and enlarging by accelerated bank erosion. They correlated the
channel stability to a 10 year flow in a pre-development forested watershed and degree of imperviousness.
Apparently, the forested watersheds of the Pacific North-west are in an equilibrium with the 10-year flow
while watersheds situated in eastern and Midwestern US and urban basins that lost their forests in the
North-west are more in equilibrium with the 1 ½ to 2 year flows. The studies concluded that a threshold
for urban stream stability exists at approximately 10 percent imperviousness of the watershed.

Figure 2.4. Channel stability as a function of

imperviousness and flow increase in the Pacific
North-west US (from Booth and Jackson, 1997)

Combined with the hydrological changes of increased flush runoff and diminishing base flow, urban
stream have significantly larger dimensions than their rural counterparts. Because of diminished
infiltration and shallow aquifer recharge, urban streams have either very little flow between events or base
flow is supplied by infiltration into storm sewers. Unmodified streams respond to increased flows and
frequency of floods due to urbanization by enlarging their cross-section until a new equilibrium is reached
in the future. For urbanized watersheds, Leopold (1968, 1994) stated that the channels tend to enlarge
proportionally to the increase in the 1 ½ year flow. The transitional period may last decades and, during
this transitional period, the channel banks become unstable and eroding. The eroded materials from the
banks and from the watershed may also silt the bottom of the streams. Consequently, fish and aquatic life
habitat is diminished.

12
Hammer analyzed 78 small watersheds located in the Philadelphia metropolitan area. About two thirds
of the watersheds contained some degree of large scale urbanization, including commercial, residential,
and industrial developments. Most of the impervious developments in the urban watersheds was suburban
and recently constructed. The remaining watersheds were rural. Hammer found that many cross-sections
of urban channels were unstable and enlarging, therefore, measurements were only made at cross-sections
that appeared to be at apparent quasi equilibrium with stream flow.

For rural watersheds Hammer obtained the following relationship for the channel cross-sectional area,
Cr [m2 or sq ft], and the area of the watershed, A [km2 or sq mi],

Cr = c A0.657

The conversion coefficient c = 13.29 if the watershed area is expressed in km2 and c = 24.8 if the area
is in square miles, respectively. Hammer then he found a dimensionless channel enlargement, R, ratio
from
Cu
R=
σ A 0.657
where Cu is the cross-sectional area for urban streams.

The channel enlargement parameter was then correlated to several watershed parameters involving
surface and imperviousness characteristic. The standard multiregression formula was

R = ?1 X1 + ?2 X2 + ?3 X3 + .... + ?3 X3

where ?’s are the regression coefficients and X’s are the fractions of the watershed with certain land use
(imperviousness) characteristics. The regression coefficients were then converted to multipliers given in
Table 2.1.

No matter what the channel form, most streams that were not drastically altered by engineering works
share a similar feature of alternating, regularly spaced, deep and shallow areas called pools and riffles.
The pools and riffles are associated with the low-flow channel, or thalweg. Pools typically form in the
thalweg near the outside bank of bends. Riffles usually form between tho bends where the thalweg
crosses over from one side of the channel to the other. Figure 2.5 illustrates an idealized natural stream
made up of riffles and pools. For larger streams, alternating deeper bends and shallower straight runs
represent the same morphological characteristics.

13
 Table 2.1 Channel enlargement multipliers

Independent parameter, X Multiplier ?

Fraction of the watershed (regression coefficient)

Land in cultivation 1.29

Wooded land 0.75
Land in golf courses 2.54
Area of houses on sewered streets
Houses 4 - 15 years old 3.36
Houses 15-30 years old 5.16
Houses > 30 years old 3.76
Area of sewered streets
Streets 4-15 years old 4.20
Streets 15-30 years old 5.16
Streets >30 years old 3.76
Other impervious area
Area 4-15 years old 6.26
Area . 15 years old 7.99

Figure 2.5. Idealized natural channel ( After Brookes, 1988)

14
Sediment and Sedimentation
All streams receive some sediment loads. The sediment moves in the streams in two forms:

! as a suspended fine sediment or washload

! in a form of a constantly shifting bottom or bed load of coarser and heavier particles

Sediment originates from many natural and anthropogenic sources (see Novotny and Olem, 1994). The
most important sources of sediment in urban watersheds are combined sewer overflows, erosion at
constructions sites, dust and dirt from streets, and erosion of degrading streams (bank erosion and bottom
scour).

Sediment movement in the stream channel affects ecological integrity of the stream system. In more
extreme cases, sediment deposition reduces the channel flow capacity. For example, sediment
accumulation in the Milwaukee’s North Avenue Dam over its more than one hundred years operation
almost completely filled the impoundment volume. The depth of water in the impoundment which
originally was more than 4.5 meters (15 ft) was less than 0.6 meters (2 ft) at the time of the drawdown
and dam removal.

The sediment can be also classified as cohesive and noncohesive. Cohesive sediments (mud) in riverine
and estuarine systems are composited of clays, silt and organic particulates. Cohesion of particles is more
pronounced for clay particles than for more coarse silt particles. Noncohesive sediments are mostly made
of sand and gravel. Because of their smaller density, organic particulates are more associated with
cohesive sediments than with sand and gravel. Sand and gravel bottom are an indication of faster flow
velocities that prevent settling of clay and organic particulates. Sediment mixtures with a fraction of clay
particles larger than 10 percent have cohesive properties .

Fine sediment itself is a nuisance pollutant that can affect aquatic organisms (Alabaster, 1982). Cohesive
sediments (clays and organic particles) can adsorb and/or co-precipitate many other pollutants, including
ammonia, phosphates, many organic chemicals and metals. Consequently, many bottom sediments can
be contaminated and can store pollutants that were deposited in them many years ago. Also, the cohesive
sediments containing organic particulates exhibit sediment oxygen demand (SOD) due to the
decomposition processes occurring in the sediment layer. Because the decomposition in the sediments
is mostly anaerobic, methane, ammonium and dissolved phosphates are released from the sediments.

Deposition of cohesive sediments occurs in reaches where the bottom shear stress is smaller than a
certain critical shear stress limit, gcs, which is about 0.1 Newtons/m2 but can be as high as 1 Newton/m2
(Mehta et al., 1989). If the bottom shear stress is greater (see the next section) no deposition of cohesive
sediments occurs and the bottom substrate may be made mostly of sand and gravel. However, the process
of cohesive sediment movement, deposition and erosion is complex and the models are highly empirical.
When larger quantities of coarser materials such as sand and larger organic detritus are present in the
sediment its behavior, interaction, scour and deposition become quite complex and not well understood
(Mehta et al., 1989).

15
Stream Bank Stability

Unstable stream banks can be a major source of

sediment, especially when the stream becomes
unstable due the change of hydrology due to
urbanization. Stream banks can be eroded by
scour of soil particles or by collapse. Collapse or
mass failure occurs when the strength of the
bank material is too low to resist the shear
stresses and gravity by flowing water. Natural
stream banks are frequently composed of distinct
layers caused by the successional deposition of
material in the floodplain. Each layer can have
its own physical properties. The bank profile
will respond according the physical properties of
each layer. Since the stability of stream banks
with respect to failures due to gravity and shear
stress depend on the geometry of the bank
profile and physical properties of the bank
materials, dominant failure mechanisms tend to
be closely associated with the characteristics
stratigraphy or secession of layers (The Federal
Interagency Stream Restoration Workgroup,
1998).

Quantitative analysis of bank instability is

considered in terms of force (weight) and
resistance (friction). The shear strength of the
bank material can be used to predict the
resistance of the boundary to erosion by gravity
and shear. Shear strength of the stream banks is
composed of cohesive strength and frictional
strength. For planar failure, the Couloumb
equation is applicable (The Federal Interagency
Stream Restoration Workgroup, 1998), where

Sr = c + ( N - W) tan k

where Sr = shear stress in Newtons/m2

c = cohesion, in Newtons/m2

Figure 2.6. Bank erosion failure process

(Source Hagerty, 1991)

16
 N = normal stress in Newtons/m2
W = pore pressure in Newtons/m2
k = friction angle in Newtons/m2

N = W cos k

where W = weight of the failure block in Newtons/m2

k = angle of the failure in degrees

The gravitational force acting on the bank is

Sa = W sin k

Factors that determine the erosional resistance (Sr) include excess pressure from saturation and the
development of vertical cracks. Increased bank height ( due to channel incision) and bank angle (due to
undercutting) favor bank failure by increasing the gravitational force component. In contrast, vegetated
banks generally are drier, and provide improved bank drainage, which enhances stability. Plant roots
provide tensile strength to the soil, resulting in reinforced earth that resists mass failure at least to the
depth of the root zone ( Yang, 1996).

Habitat Integrity Parameters

The changes in channel morphology, sediment scouring and accumulation and sediment texture are the
major parameters affecting the physical integrity of the water body.

The Rapid Bioassessment Protocols by Plafkin et al. (1989) that describe methodology for development
of the Index of Biotic Integrity list nine habitat parameters. Some of them can be correlated to flow
characteristics. The matrix of parameters for physical habitat evaluation are:

1. Bottom substrate and available cover. Rocks and gravel in flowing streams are generally
considered as the best substrate for the most desirable habitat.

2. Embeddeness is the degree to which boulders or gravel are surrounded by fine sediments.

3. Stream flow and/or velocity. The minimum flow to support healthy aquatic biota is 0.15 m3/s. In
larger streams, velocity and depth are more important than flow. This parameter is directly
related to the minimum flow characteristics such as 7 Q 10; however, a higher critical flow value
such as 7 Q 3 may be more realistic. A 7 Q 10 low flow parameter has been defined as the
smallest annual moving average of 7 consecutive flows that has a ten years recurrence interval
of being less or equal. 7 Q 3 low flow characteristic has a recurrence interval of being less or
equal of three years.

4. Channel alteration. Channel and watershed alteration by man results in channel instability,
watershed and bank erosion and also deposition of eroded materials in places where velocity is
reduced.

17
5. Bottom scouring and deposition. Scoured substrate and siltation in slow section are the
characteristics that fall under this IBI parameter. The potential for scouring is increased by
channelization.

6. Pool/riffle or run/bend ratio. A stream with riffles or bends provides more diverse habitat than
a straight (run) or uniform depth stream. Low gradient streams may not have riffle areas, but
excellent habitat conditions can be provided by the cutting action of water at bends.

7. Bank stability. Streams with poor and eroding banks will often have a poor instream habitat.
Steeper banks are more susceptible to erosion and failure, and may not support bank vegetation.

8. Bank vegetative stability. This parameter is related to some degree to the previous one. Stream
banks are often stabilized by roots of stream bank vegetation and also by boulders and gravel.

9. Stream side cover. Stream side vegetation provides shading and escape cover or refuge for fish.
Stream side cover consisting of shrub provides a higher fish standing crop than similar size
stream having tree or grass stream side cover (Platts, 1974; Plafkin et al., 1989)

The first three stream habitat condition parameters have the highest weight - 44.44 percent of the overall
maximum assessment, the second three parameters count for 33.33 percent of the overall assessment and,
the last three count for 22.23 percent, respectively.

Morphological characteristics that are related to the habitat quality and, hence, to the integrity of a stream
are (Plafkin et al., 1989) and changed by urbanization are:

! Channel cross-section, width and depth are increased during high flows
! Base flow is diminished
! Sinuosity of the channel is affected
! Pool/riffle or bend/run structure may be lost
! Channel substrate (bottom) characteristics are impaired
" Embeddedness, fine sediments become embedded into the coarse
substrate
" Texture of the bottom substrate is changed
" Scouring and deposition of fine sediment is increased.
" Frequent stream bed turnover
! Velocities and bottom scour is increased during higher flows
! Sediment is contaminated with toxic materials
! Loss of riparian vegetation and organic debris

Many of the above morphological, parameters are related to changing of hydrology (Table 2.2).

18
Table 2.2 Relation of hydrologic stream parameters to physical integrity and habitat conditions

Condition parameter Desirable state Key determinant Determining hydrologic

conditions
1. Bottom substrate and >30 % rubble, gravel, stable habitat Bottom shear stress greater than Mean annual flow with a
available cover the deposition shear stress for fine recurrence interval of once in
particles three years
2. Embeddedness Gravel, cobble and boulder more than Bottom shear stress greater than Mean annual flow with a
25 percent the deposition shear stress for fine recurrence interval of once in
particles three years
3. Flow/velocity Flow>0.05 m3/s and at least 3 habitats Small flows and resulting velocity Small flow at 7Q3 level
based on velocity (>< 0.3 m/s) and and depth
depth (>< 0.5 m) should be present

4. Channel alteration Minimum channelization Degree of past manmade No direct relation to hydrology
channelization

5. Bottom scouring and <10 % of bottom affected, mostly near Medium high flows should result 1 ½ to 2 year high flow
deposition constrictions and change of slope in shear stress that is less than the
critical erosive bottom stress.
6. Pool/riffle, run/bend ratio < 15 Degree of past manmade No direct relation to hydrology
(Distance between riffles channelization
divided by stream width)

7. Bank stability Moderately stable, small eroded areas Channel capacity flows 1 ½ to 2 year high flow, stream
mostly healed. bank stability parameter (Eq. 7)

8. Bank vegetation Over 50 % of bank surface covered by Tree/brush cutting and stream 1 ½ to 2 year flow if bank
stable vegetation, gravel or larger channelization efforts, Bank vegetation loss was caused by
material erosion bank erosion
9. Stream side cover Dominant vegetation is shrub and/or Past tree/brush cutting and stream No direct relation to hydrology.
tree channelization efforts

19
Bottom and bank stability parameters. Most habitat quality parameters are, to various degrees, affected
by erosion/deposition properties of the sediments carried by the stream and in the bottom layer. The scour
and deposition of sediments is related to the bottom shear stress, g, defined as

τ = γ R Se
where E is the specific weight of water ( E=9,810 N/m3)
R is the hydraulic radius of the channel in meters, and
Se is the energy slope of the channel.

The unit of the shear stress is N/m2. The science of sedimentology and stream erosion relates scour and
deposition to the difference between the shear stress and critical shear stress for either scour or
deposition. In this case if g < gc1 , deposition of sediments will occur and if g > gc2 bottom scouring can
be expected. gc1 and gc2 are then critical shear stresses for deposition and scouring, respectively.
Obviously, the critical shear stress depends on the size of the bottom material (e.g., fine sand, gravel).
However, for cohesive fine bottom sediments that have the greatest impact on the quality of the benthic
habitat, the critical shear stresses were found to be in the range of 0.1 to 1 N/m2 (Partheniades , 1972;
Aruthurai and Arulanandan , 1978; Ziegler and Lick, 1988). The literature is not unified on the magnitude
of the critical shear stresses and the models are not accurate.

Substituting well known Mannings equation ( v = 1/n R2/3 Se1/2 ) for the energy slope the shear stress
formula becomes
v2 n2
τ = 9810 1
3
H
where v is the average flow velocity, n is the Manning roughness factor for the channel and H is the
depth of flow.

It is interesting to note that if the typical Manning roughness is selected as n = 0.03, velocity
v = 0.3m / s and H = 0.5 m then the magnitude of the shear stress is about 1 N/m2 that is the
magnitude of the critical shear for deposition of cohesive sediments. One would then expect that if g >
1 the bottom of the stream may be composited mostly of sand.

Sediment movement affects the biological community by changes in habitat and physical impacts on the
organisms. Suspended sediment causes increased turbidity that affects living aquatic organisms in several
ways. As bed load moves, habitat for small invertebrates that inhabit the stream is affected and the
organisms are dislodged.

Impacts of Stream Channelization and Armoring

Developers and urban drainage agencies often modify channel morphology in urban environments in an
attempt to control the stream channel. Channel straightening, and/or lining is often done to improve
drainage and reduce flooding risks. Stream straightening, or channelization, results in steeper bed
gradients, higher velocities of flow, and increased capacity to transport sediments (Simon and Senturk,

20
1977; Linder, 1976; Parker and Andres, 1976; Brookes, 1988). In the absence of bank protection
measures, many channels that have been straightened had a tendency to regain former sinuosity (Brooks,
1988). Protection of the banks reduces the tendency of the channel to meander, but can also cause
instability in the stream bed (Ryckborst, 1980). Total lining of the channel with concrete or rock rip-rap
can produce stability within the armored reach, however, this can affect upstream downcutting and
downstream sediment deposition. (Parker and Andres, 1976).

The antidegradation rule and existing Clean Water Act mandates do not allow worsening of the
ecological integrity of the nation’s (navigable) waters even when it could bring about an improvement
of flood control. In other words, hydraulic works such as channel lining, straightening, channelization and
others that have strictly flood control benefits but are detrimental to the habitat, are not feasible and
permissible today.

Citizens in many communities that previously installed concrete stream lining in their communities are
today pressing public officials to “re-naturalize” the stream and “re-green” the stream corridor which
implies removal of the lining and replacing it with a bioengineered stream bank protection and restoring
habitat conditions. The present Lincoln Creek (Milwaukee, WI) flood control protection and stream
restoration project is an example of such a restoration effort.

Impact of floodplain
Floodplain, even in urban setting, is an important hydrological component. Floodplain provides storage
for excess flows and, hence, reduces the peak flow. Reclamation and development of the floodplain has,
therefore, a dramatic effect on increasing the flood flows downstream.

One can use a simple mass/flow balance equation to illustrate the effect of the floodplain storage. For
a stream section with of without a floodplain storage the mass balance equation is

dS
=I−O
dt
or
dS
O=I−
dt
where S = channel and floodplain storage in the section, including channel and floodplain
I = flood inflow into the stream section
O = flood outflow from the section

In simpler flow routing models, the storage is related to the outflow as S = a Ob because both O and S are
a function of the same variable, the flow depth in the section. If there is a large floodplain storage
available during a time interval Ft the change of storage volume in the stream section, FS, is larger,
therefore, the outflow is smaller. If the available storage is small (e.g., limited only to the channel), the
storage change is small and the outflow from the section is larger. Floodplain storage adds also lag time
to the flood, i.e., it slows the flood wave down and reduces in this way the peak flow rate.

21
Floodplain is also a place where sediments carried by the flood flow may deposit, especially when the
flood flow hydrograph is receding. This sediment deposition makes little or no harm to natural vegetated
floodplains; however, it is troublesome to owners of properties located in the floodplain because of the
clean-up cost and damagaes.

Effect of bridges, culverts and other channel constrictions

Often, channels have adequate capacity to convey high flows but are constricted by bridges and culverts,
siphons, and other obstructions that have or cause an insufficient flow capacity. Many older bridges in
urban areas were designed years ago using simple Rational Formula and underestimated the extent of the
future development. For example, stone bridges built in the 1930s during the times of Depression on
Milwaukee’s Lincoln Creek when the upper sixty percent of the watershed was undeveloped have present
capacity to convey only a flow that is less than the ten year flood. Thus, these bridges represent a
“bottleneck” causing extensive flooding of the surrounding watershed.

Constricting a channel by a bridge and its piers and embankments also increases water level upstream,
which a well known hydraulic phenomenon based on the flow energy conservation (Figure 2.7).

Figure 2.7. Hydraulic effect of bridges on upstream flooding

caused by channel narrowing

22
 CHAPTER 3

STREAM CORRIDOR

As discussed throughout this report, the stream corridor can be divided into the following zones

! Stream channel
! Floodway
! Flodplain fringe

Within the stream corridor, a variety of ecosystem habitats can form based on seasonal inundation. Figure
2.1 illustrates a cross-section of a series of stream corridor habitats. In this example, the floodplain is
seasonally inundated and includes wetland communities such as floodplain forest, emergent shallow and
deep marshes, and wet meadows. The transitional upland fringe includes an upland forest and a hill
prairies. The various plant communities have unique moisture tolerance and requirements and
consequently occupy distinct land forms.

Figure 3.1. A cross-section of a river corridor (Source Sparks, 1995; reprinted from Federal
Interagency Stream Restoration Working Group, 1998)

Stream Channel

The stream channel includes several biological components, including plants, invertebrates, and fish. The
ecosystem is made up of the movement of energy and materials. The sun provides the primary energy
input to the system. Primary producers use the process of photosynthesis to convert the sun’s energy into
chemical energy. Organisms, such as attached algae and rooted aquatic plants, form the base of the food
pyramid. Allochthonous materials from outside the channel, such as leaves and twigs, add to the organic
matter that feeds the stream system. In fact, in most stream environments, allochthonous material is the
major source of carbon and nutrients to the stream system (Hynes, 1970; Allan, 1995; Cowx and

23
Welcomme, 1998). Insects and insect like organisms known as macroinvertebrates, work to convert the
plant and debris material into body mass (Allan, 1995). The invertebrates provide a food source for fish.
Microbes within the system help facilitate the process of breaking down organic waste. Figure 3.2
illustrates a schematic of the aquatic food web

Figure 3.2. Food Web Schematic (from Novotny and Olem, 1994)

Algae, in the form of both attached algae, known as periphyton and free floating algae, known as
phytoplankton, play an important role in regulating nutrients within the water column (Allan, 1995).
Algae are important producers of oxygen in the stream. However, too much algae, caused by excess
nutrients, can cause oxygen depletion during night and early morning hours when respiration dominates
(Welch, 1992).

Both rooted aquatic, that live in the stream, and riparian plants, along the banks, play important roles in
the stream ecosystem. In streams with limited physical structure, vegetation is likely the most important
feature for the creation of fish and aquatic life habitat. The following is a list of the ecological values of
rooted aquatic plants (Cowx and Welcomme, 1998):

• Water purification –
a. direct: by oxygenation and conversion of toxic ammonia to usable nitrates
b. indirect: provision of huge surface area for microbes

• Nutrient re-cycling – removal during growth season, return during die back (senescence)

24
 • Physical link between water and air for many invertebrates, e.g. many invertebrates
have aquatic larval stages and aerial adult stages

• Refugia for zooplankton – which graze on phytoplankton and keep water clear

• Cover for invertebrates – many of which are food for fish

• Cover for fish – value and type varies with age/species of fish as well as type of vegetation

• Spawning areas – for fish species such as northern pike and perch

• Food source (living and dead) – direct source for vegetarian fish such as carp

• Affects flow patterns – accretes sediment and deflects flows to provide quieter waters and
faster shallows

• Creation of discrete habitat structure – functions as physical structure

Benthic invertebrates act in the system to break down coarse organic matter and provide food for fish.
Invertebrates can be divided in various groups based on how they consume food (Allan, 1995). Common
categories include collectors, shredders, grazers, and predators. Invertebrates live in organic sediments,
on top of rooted aquatic plants and attached to gravel and cobble material. Invertebrates play a major role
in the breakdown of coarse organic matter such as leaves and other organic debris (Wotton, 1994).
Examples of benthic invertebrates include caddis fly, mayfly, and chronmids.

Fish form the top of the aquatic food chain. To survive, fish require a variety of habitats to coincide with
their various life stages. Each species will have its own set of unique habitat requirements. The following
is a general list of habitat needs for most riverine fish (Cowx and Welcomme, 1998):

1. REPRODUCTION

a. Access to spawning area

• Provision of suitable depths and water velocity
• Absence of barriers to movement

b. Spawning
• Suitable spawning substrate

c. Incubation of eggs
• Stability of substrate
• Provision of adequate temperature and oxygen supply and water movement
1. FEEDING AND GROWTH

25
 a. Availability of food organisms
• Bankside and aquatic vegetation
• Substrata suitable for invertebrate production
• Supply of allochtonous organic material

a. Best use of energy for maintaining position and food gathering

• Cover and shade, e.g. rocks and or tree trunks
• Diversity of flow types
• Pool-riffle sequences
• Aquatic and bankside vegetation
• Appropriate temperature range

1. SELF PROTECTION

a. From physical displacement by current

• Shelter and visual isolation, e.g. varied bed profiles through: under cut banks, rocks,
tree trunks, roots, accumulated debris, aquatic vegetation, weedy shallow marginal
slacks (for juveniles); including backwater and lateral systems

Many fish exist within specific habitat ranges. To understand the importance of various habitat
requirements, the U. S. Fish and Wildlife Service (USFWS) has developed the Habitat Evaluation
Procedure (HEP). Within the procedure, suitability of habitat for a given species is described by a habitat
suitability index (HSI) constrained between 0 (unsuitable habitat) and 1 (optimum habitat) (Schamberger
et al., 1982). Within each HSI model, each of the individual habitat factors is weighted based on its
importance to the specific life stage of the species. The total HSI is based on an average of the weighted
values. HSI models are available for a number of aquatic and terrestrial species. The HSI modles are
published by the USFWF (Schamberger et al., 1982).

A simplified river continuum model for the stream environment is illustrated in Figure 3.3. The concept
proposes a relationship between stream size and the progressive shift in habitat types and speciesmakeup.
In the headwater areas the riparian forest shades the stream. The shading limits growth of algae,
periphyton, and other aquatic plants. Since energy in this reach is not created by photosynthesis, it comes
from materials from outside the channel, such as leaves and twigs. Macroinvertebrate communities in
the headwater areas is dominated by grazers and shredders who help breakdown the course organic
material

As one moves downstream the channel becomes more open to sunlight and the stream develops
increased reliance on primary produces such as algae and rooted aquatic plants. The stream continues
to receive inputs of dissolved and ultra-fine organic particles from upstream. Fine particle collectors such
as benthic worms and zooplankton dominate the invertebrate populations in the downstream areas.

26
Figure 3.3. The River Continuum Concept - A relationship between stream size and
the progressive shift in structural and functional attributes (reploted
from the Federal Interagency Stream Restoration Workgroup, 1998;
source Vannote et al., 1980)

27
Floodplain

The floor of most floodplains, or riparian corridors, in the north central US is relatively flat. Over time,
stream channel move back and forth across the stream valley in a process called lateral migration
(Leopold et al., 1992). In addition, periodic flooding causes sediments to be deposited on the valley floor.
These two processes continually modify the floodplain.

The floodplain has then following functions:

! It provides storage for flood flows, thus, reducing downstream peak flows and flood
damages.
! It provides habitat for a variety of ecosystems (Figure 3.1).
! It provides buffers to reduce nonpoint pollution inputs from the surrounding lands.
! It receives and removes sediments from the flood flow.
! In urban areas it provides open space for parks, recreation, nature areas and enjoyment.

The ecological integrity of the stream corridor is directly related to the integrity and ecological
characteristics of the plant communities. The plant communities provide valuable and physical habitat
for other organisms that inhabit the corridor. The distribution and characteristics of the vegetative
communities is a function of climate, water availability, topography, and soil properties (The Federal
Interagency Stream Restoration Workgroup, 1998).

The quantity of terrestrial vegetation, as well as species makeup, can effect the characteristics of the
adjacent stream channel. Root systems in the streambank bind sediments and regulate erosion. Material
that falls in the stream can deflect flows and induce erosion at some points and deposition at others. The
plant material regulates nutrients in the water that flows from upland areas through the floodplain into
the stream.. Studies conducted by the North Carolina Agricultural Research Service have shown that
vegetated riparian stream corridors can remove 80% of the sediment and 50% of the total phosphorus
exported from adjacent agricultural fields, illustrating their importance in trapping material (Copper and
Gillian, 1987, Copper et al. 1987). A wider corridor provides more effective filtering, and a contiguous
corridor provides a filter along its entire length.

Stream corridors are used by wildlife more than any other habitat type (Thomas, 1979). Like man,
wildlife need water as an essential element of survival, therefore wildlife tend to concentrate within
corridors near stream channels or lakes (Harris, 1984). In Arizona, 60% of the wildlife species reside in
riparian areas (Ohmart and Anderson, 1986). In the Great Basin of Utah and Nevada, 288 of the 363
terrestrial vertebrates depend on riparian zones (Thomas, 1979).

The faunal composition of a stream corridor is a function of the interactions of food, water, cover, and
spatial arrangement (Thomas, 1979). These habitat components interact with each other and produce the
following eight habitat features:

• Presence of water

28
• High primary productivity and biomass

• Dramatic spatial and temporal contrasts in cover types and food availability
• Critical microclimates

• Horizontal and vertical habitat diversity

• Maximized edge

• Effective seasonal migration routes

• High connectivity between vegetated patches

Stream corridors often include two general types of habitat structure: interior and edge habitat. Habitat
diversity is increased by a corridor that includes both edge and interior conditions. Interior plants and
animals will differ considerably from those that prefer or tolerate the edge’s variability. The interior
habitats are more stable, sheltered environments where the ecosystem may remain relatively the same for
prolonged periods. The interior is generally inhabited by larger mammals and can provide habitat for
certain bird species that only inhabit forest interiors. The edge community is more dynamic and exposed
to more variable environmental gradients. Edge communities tend to be dominated by smaller sized
mammals and a larger variety of insects.

More species tend to find suitable habitat in wide continuous and plant diverse corridors. Habitats
provide organisms or communities of organisms with the necessary elements of life, such as space, food,
water, and shelter. For some species, such as woodland birds, space is as important habitat feature.

Fragmentation is the breakup of the vegetated corridor into small isolated pieces often caused by urban
development. Fragmentation of the corridor can cause stress or even the elimination of some species.

Floodplain corridors are inhabited by a diverse variety of species, including insects, reptiles, amphibians,
birds, and mammals. Nearly all amphibians (salamanders, toads, and frogs) and insects depend on
aquatic habitats for reproduction and over-wintering. Many species use ephemeral or temporary ponds
that form after inundation of the floodplain. Without these temporary pools, several species of
salamanders, toads, frogs and certain insects would not survive.

Birds are one of the most common wildlife species in the floodplain corridor. Nationally, over 250
species have been reported using riparian areas during part of the year (The Federal Interagency Stream
Restoration Workgroup, 1998). The highest known density of nesting birds in North America occurs in
southwestern cottonwood habitats along streams (Carothers and Johnson, 1971). Seventy-three percent
of the 166 breeding bird species in the Southwest United States prefer riparian habitats (Johnson et al.
1977). To have a diversity of breeding birds, the riparian corridor needs to provide a diversity of vertical
complexity (Carothers et al.,1974). Vertical diversity is provided by a variety of trees, shrubs, and
herbaceous subshrubs. Most large mammals include riparian corridors in their territory. The combination

29
of cover, water, and food make them desirable for species such as deer, raccoons, and mink. Some
mammals live almost exclusively in the riparian corridor, such as otters, beavers, and muskrats. Many
bat species such as the brown bat, big brown bat, and pallid bat roost in floodplain areas (Brinson et al.
1981). The diversity of the mammal community is directly related to the diversity of the habitat in the
corridor.

SUMMARY

Ecological integrity of a stream is governed by several complex interacting mechanisms. Table 3.1
summaries the physical, chemical, and biological components that influence stream ecological integrity.

TABLE 3.1
Physical, Chemical and Biological Components that Influence Stream Ecological Integrity.

Hydrology Channel Processes

• Total annual discharge • flow characteristics
• seasonal (monthly discharge) • channel dimensions, shape, profile and pattern
• peak flows • substrate composition
• minimum flows • floodplain connectivity
• annual flow durations • evidence of entrenchment and/or deposition
• rainfall • lateral bank erosion
• size and shape of the watershed • floodplain scour
Erosion and Sediment Yield • channel avulsion/realignments
• watershed cover and soil stability • meander and braiding processes
• dominant erosion processes • depositional features
• rate of erosion and mass wasting • scour-fill processes
• sediment delivery ratios • sediment transport class (suspended, bedload)
• channel erosion processes and rates Aquatic and Riparian Species and Critical Habitats
• sediment transport functions • aquatic species of concern and associated habitats
Floodplain/Riparian Vegetation • riparian species of concern and associated habitats
• community type • native vs. introduced species
• type of disturbance • threatened or endangered species
• surface cover • benthic, macroinvertebrate, or vertebrate indicator
• canopy Water Quality
• community dynamics and succession • temperature
• recruitment/reproduction • dissolved oxygen
Corridor Dimension • suspended sediment
• width • nutrients
• linearity • pH
• connectivity • alkalinity
Source: The Federal Interagency Stream Restoration Workgroup, 1998

30
 CHAPTER 4

ECOLOGICAL IMPACT OF URBAN STORMWATER

Impacts of Urbanization on Aquatic Habitat

The ecology of urban stream habitats is influenced by the changes in hydrology, geomorphology and
water quality that are consequences of the development process. Urban streams generally have poor in-
stream habitat quality as compared to rural streams that are not heavily impacted by agriculture. Habitat
degradation in urban streams is often caused by the following factors:

• Loss of pool and riffles structure

• Embedding of streambed sediments

• Loss of base flow, resulting in more shallow depth

• High velocities during storm events

• Frequent streambed turnover

• Sediment contamination with toxic materials

• Loss of riparian vegetation and organic debris

Heat absorbed by urban runoff from impervious surfaces and in open detention ponds, combined with
the loss of shading by the reduction in riparian vegetation, can increase summer stream temperatures. by
1 to 6°C (2 to 10 degrees Fahrenheit) (Galli, 1991). Since temperature play an important role in the life
cycle of many aquatic organisms, even slight shifts in temperature can impact growth and survival.

During the development phase of an urban watershed, fine sediment export from construction sites
increases stream sediment loads dramatically (Novotny and Chesters, 1981). As fine sediment is washed
into stream channels, a portion of the material becomes embedded into the coarse substrate. Pore spaces
between the gravel and cobbles become filled with finer sediment. Surface habitat for organisms that
inhabit the substrate material is lost. As pore spaces become clogged, less stream water flows through the
course material. The micro-layer of algae (periphyton), bacteria, and fungus on the surface of the sediment
is reduced, reducing the ability of the stream to breakdown dissolved organic material. Macroinvertebrate
habitat is also reduced as available surface on the substrate is reduced.

As discussed above, urbanization results in a decrease in stream base flow and changes in channel cross-
sectional area. Simmons and Reyonold (1982) noted that dry weather flows in urban watersheds of Long
Island, New York, were reduced by 20 to 85% after development. With less base flow, the wetted

31
perimeter of the stream during dry periods is reduced, reducing the available area for aquatic
macroinvertebrates that reside in the sediments. The reduced base flow spread over the wider stream
channel results in shallower water depth, reducing habitat for fish species that reside in pool
environments. Many perennial headwater streams become intermittent (ephemeral) lafter watershed
urbanization (Walesh, 1989), further reducing available aquatic habitat.

Large woody debris is an important structural component of many headwater streams. The debris creates
habitat structure for aquatic invertebrates and refuge for fish. In urban streams, the quantity of woody
debris is reduced due to loss of riparian forest cover, storm washout, and channel maintenance (Booth
et al., 1996; May et al., 1997).

As urban areas develop, there is generally a reduction in trees and shrubs along the stream corridor. As
fewer leaves are washed into urban streams, there is a shift from external organic matter to internal forms
such as algae. This shift in available food impacts on the makeup of macroinvertebrates in the stream.

Impact of Urbanization on Biotic Communities

Karr and Chu (1997) noted that diverse human activities during urbanization interact to affect conditions
in watersheds, water bodies, or stream reaches. These activities and inmpacts on the stream biota can be
ranked and eventually assessed according to some additive index of the urbanization impact. Biological
impacts at different sites be grouped according to:

! a single dominant variable such as percent of imperviousness

! several qualitative disturbance categories (Paterson, 1996) such as

" little or no human activity
" little recreational use
" heavy recreational use
" heavy use for urbanization, agriculture, etc.

! some measure of adverse impact on aquatic biota such as a numeric risk (Novotny and
Witte, 1997; Parkhurst et al., 1996)

The effect of these factors on biota can be measured in terms of Indices of Biotic Intergity - IBI (Plafkin
et al., 1989) that can be expressed either as the benthic macroinvertebrate index (B-IBI), or fish based
index (F-IBI). Many studies have confirmed patterns and extent of human influences on the composition
and densities of aquatic organisms residing in the streams affected by urbanization.

In a study of 208 Ontario streams, it was found that benthic diversity shifted from fair to poor when urban
development in a watershed reached 35% (Steedman, 1988). The Ontario study found that urban streams
with intact riparian forests had higher diversity than those that did not, for the same level of urbanization.
Figure 4.1 illustrates the result of a study of macroinvertebrate diversity as it relates to percent watershed

32
imperviousness conducted for 23 sampling stations in the headwater streams of the Anacostia watershed

Figure 4.1 Relationship of macroinvertebrate (B-IBI) diversity to

watershed imperviousness (from Schueler and Galli, 1992)
in Maryland (Schueler and Galli, 1992).

Maxted (1996) in a study of 38 streams in Piedmont region of Delaware using community and sensitive
species diversity indexes, also found macroinvertebrate diversity to decline with watershed
imperviousness. The study found that macroinvertebrate species such as stoneflies, mayflies, and
caddisflies, declined and were replaced with species in the chironomidae family. The study found that
a threshold appeared to exist once watershed impervious cover reached 10 to 15%. Increased
urbanization (>15% impervious cover) did not significantly degrade biological communities further. It
was concluded that first and second order (headwater) streams were more sensitive to urbanization.
Similarly, Garie and McInntosh (1986) reported that the number of benthic invertebrate taxa declined
from 13 in relatively undeveloped areas to 4 in heavily urbanized areas for a Trenton, New Jersey stream.

As urbanization increases, fish species diversity also declines (Schueler and Galli, 1992; Wang et al.
1997; Dreher, 1997). Several authors have used Fish Indexes Biotic Integrity (F- IBI) to define the
impacts of urbanization on fisheries. The IBI is an index that utilizes several parameters that reflect
species richness and composition, trophic composition, and fish abundance. In a study of 40 stream in
Northeastern Illinois, Dreher (1997) compared human population density to IBI. Human population was
used in the study as an indicator of urban development. The results of the Illinois study are illustrated in
Figure 4.2 and show a direct decline in IBI as urban density increases.

33
Impacts of land use on stream IBI has also been studied in 103 streams located in Wisconsin (Wang et
al. 1997). In the Wisconsin study, both the IBI and a physical habitat index were used. The habitat index
assesses the physical conditions of the stream. The results showed that as the percent urban cover
increased from 0 to 20%, the IBI dramatically decreased. At 20% urban land cover, the effects of
urbanization reached a threshold. The study found that habitat indexes did not decline as percent urban
land cover increased, leading the authors to conclude that other factors such as water quality or
hydrological impacts of urbanization may be more important than habitat degradation. Steedman (1988)
in a study of Toronto, Ontario streams found similar relationships between IBI and the amount of urban

Figure 4.2 Relationship of population density to the stream index of biotic

integrity (F-IBI) (Source Dreher, 1997)
land.

Water Quality Impacts of Urban Runoff

Monitoring conducted as part of the National Urban Runoff Program (NURP) has shown urban surface
water runoff to include problem pollutants such as sediment, nutrients, chlorides, bacteria, oil and grease,
heavy metals, pesticides and volatile organic compounds (VOCs) (USEPA, 1984).

34
Pollution caused by surface water runoff, also called nonpoint source pollution, is the result of human
activity on the land surface. There are two main reasons why urbanization increases pollutant loads in
runoff. First, the volume and rate of runoff are increased as an area is developed. Changes in the
hydrologic budget caused by modifications to soils and covering of the land with impervious surfaces
results in reduced rates of infiltration, loss of surface storage, and increases in surface water runoff.
Increases in surface water runoff provide a mechanism to transport pollutants that are deposited on the
land surface. As the density of urbanization increases, the deposition of urban pollutants also increases.
Increased volume of automobile traffic, for example, has been demonstrated to increase pollution
deposition on road surfaces resulting in higher event mean concentrations of heavy metals in surface
wash-off (Bannerman et al., 1992, 1993).

Studies of urban runoff have shown that the amount of pollution generated off the land surface is directly
proportional to the quantity of runoff (Pitt, 1991). The amount of pollutants that come off the land
surface is a factor of land use, the amount of imperviousness, and automobile traffic. Table 4.1
summarizes the Water quality characteristics found by the NURP study. The event mean concentration
used in Table 4.1 is the total mass of the constituent in the runoff event divided by the total volume of
runoff during the event.

Since the NURP studies in the early 1980’s, several investigators have identified additional pollutants in
urban stormwater (Hoffman et al., 1984; Fam et al., 1987; Tanizaki et al., 1992). In a study by Thon
(1992), involving monitoring of 120 “priority pollutants”, 77 were detected in stormwater discharges.
The study found 14 inorganic and 63 organic compounds. Bannerman et al. (1992, 1993) in a similar
study of priority pollutants monitored in Milwaukee and Madison, Wisconsin, found 11 heavy metals and
inorganic compounds, 14 pesticides, and 8 other organic compounds that were detected in more than 10%
of the samples collected. The results of the Wisconsin study are summarized in Table 4.2.

TABLE 4.1
Water quality characteristics of urban runoff identified
in the National Urban Runoff Program study

Site Median Event Mean Concentration

Constituent Typical Coefficient
of Variance
90th Percentile
Median Urban Site Urban Site
TSS, mg/l 1-2 100 300
BOD, mg/l 0.5-1 9 15
COD, mg/l 0.5-1 65 140
Total P, mg/l 0.5-1 0.33 0.7
Soluble P, mg/l 0.5-1 0.12 0.21
TKN, mg/l 0.5-1 1.50 3.30
NO2+3, mg/l 0.5-1 0.68 1.75

35
 Total Cu, ug/l 0.5-1 34 93
Total Pb, ug/l 0.5-1 144 350
Total Zn, ug/l 0.5-1 160 500
Source: U.S. EPA, 1983

TABLE 4.2
Percent of occurrence of Priority Pollutants in urban runoff
in Milwaukee and Madison, Wisconsin

Parameter Percent Parameter Percent Parameter Percent

Occurrence Occurrence Occurrence
Metals and Inorganics Pesticides Atrazine 59%
Cyanide 12% Aldrin 15% 2,4-D 67%
Silver 31% Heptachlor 15% Diazinon 82%
Mercury 34% Cyanizine 22% PAH’s PCB’s & Plasticizers
Cadmium 43% Metolachlor 26% Chrysene 12%
Antimony 51% Lindane 29% Benzo(ghi)Perylene 12%
Arsenic 81% Chlordane 29% PCB’s 17%
Chromium 98% DDT 36% Phenanthene 19%
Copper 98% Methoxychlor 44% Pyrene 23%
Lead 98% Malathion 49% Fluoranthene 31%
Nickel 98% Alachlor 52% Bis Phthalate 48%
Zinc 100% Dicamba 56% Benzo A Pyrene 98%
Source: Bannerman, 1992

TABLE 4.3
Pollutants found in the bottom sediments of three Milwaukee streams (mg/kg)

Rural
Parameter Reference Kinnickinnic Wilson Park Oak Creek
Stream River Creek
Percent Urban Land Use in Watershed 0% 100% 100% 50%
Cadmium 0.25 0.36 0.4 0.2
Chromium 13 17.2 13.1 9.1
Copper 24.4 11.9 11.2 8
Lead 8.45 68 92 12
Mercury 0.08 0.03 ND ND
Zinc 32.5 71 67 82
PCBs (total) ND1 0.05 ND ND
DDT (total) ND 0.09 ND ND
PAHs (total) 0.37 33.2 83.2 116
Oil and Grease 425 1,200 2,000 3,200

36
 Total Phosphorus 645 370 240 200
Total Organic Carbon 50,550 17,600 6,070 16,600
Source: Masterson and Bannerman, 1994
1
ND means not detected
Chemicals in stormwater runoff, once they reach the receiving body of water partition into either
dissolved compounds in the water, attached to bottom sediments, or are bio-accumulated by aquatic
organisms (Pitt and Baron, 1989; Novotny and Olem, 1994). Pollutants found in the bottom sediments
of three Milwaukee, Wisconsin urban streams are summarized in Table 4.3 (Masterson and Bannerman,
1994).

Nutrients . Nutrients, such as nitrogen and phosphorus, in urban runoff, in excess concentrations, can
cause enrichment to downstream resources. Increased plankton and periphyton growth have been
documented downstream of stormwater discharges (Welch, 1992). In excess quantities algae can cause
high levels of turbidity, shade rooted aquatic plants, effect sight feeding fish, and cause evening dissolved
oxygen depletion.

Toxic Compounds. Pollutants detected in water and stored in the sediments can bioaccumulate in aquatic
organisms (Fitchko, 1986). In the Milwaukee study by Masterson and Bannerman (1994), whole fish
tissue analysis found elevated levels of PCBs, DDT, cadmium, and chromium, when compared to fish
collected from a rural reference site. The same study found elevated levels of cadmium, chromium,
copper, zinc, and total PAHs in whole crayfish samples. Pitt and Barron (1989) discuss the
bioaccumulation of toxic urban runoff pollutants in aquatic organisms in detail in a report produced for
USEPA. Bioavialability is also discussed in detail in a book by Hamelink et al (1994).

While many studies have documented the presence of pollutants in urban stream environments, the direct
cause and effect relationships of urban runoff on aquatic organisms is not well documented (Heaney and
Huber, 1984). Field and Pitt (1990) have concluded that the effects of urban runoff on receiving water
biota are very site specific.

Many authors have attempted to document the biological impacts of urban runoff on aquatic life. Heaney
and Huber (1984) reviewed fish kill information reported to government agencies for the period of 1970
to 1979. They found that 3% of the 10,000 reported fish kills were reported to be caused by urban runoff.
However, a large number of the 10,000 fish kills were not identified as having a direct cause and may
have been caused or worsened by urban runoff. Since most effects of urban runoff are related to chronic
toxicity and not acute toxicity fish kill data is not a good indicator of receiving water impacts (Pitt, 1995).
Fish disease surveys have been used as indicators of water quality problems (Scott et al. 1982; Pitt and
Bissonnete, 1983; Murchelano, 1988). Bioassays using various fish and invertebrate species (zooplankton
and macroinvertebrates) have also been used (Burton and Stemmer, 1988; Bascombe et al., 1990;
Bannerman et al., 1992, 1993). The results of whole effluent bioassays, using daphnia, on the discharges
of urban runoff from four sites in Milwaukee, Wisconsin found mortality to range from 90 to 100%
(Bannerman et al., 1992, 1993).

It has been well documented that urban runoff is contaminated with several known pollutants. When
exposed to these pollutants aquatic life has experienced disease, reproductive problems, and even death.
What is not well understood at this time is the effect individual contaminants. Urban runoff is a soup of

37
many compounds that may be working individually or in combination with other compounds to impact
on the aquatic environment. Aquatic organisms are impacted by several overlapping stresses ranging
from habitat degradation to chemical pollution. Which factors are most limiting to the system are not
well understood and likely vary from individual site to site.

Elimination of Riparian Vegetation

In the effort to increase capacity of the urban streams and to develop surrounding riparian lands, riparian
vegetation of urban stream is commonly removed, leaving channel bare. This may increase the velocities
in the channel or enhance stream views of riparian home owners. However, it reduces or eliminates the
aquatic habitat, deprives the stream of shading and refuge and increases stream bank erosion (Patrick,
1975). Removing bank vegetation and clearing the riparian zones has the following adverse impacts:

! Increases stream bank erosion and cause channel widening

! Increases temperature of the stream by reducing or eliminating shading.
! Eliminates refuge and shelter for aquatic organisms
! Reduces dissolved oxygen levels
! Causes a shift of stream flora to algal domination
! Affects the pool/riffle sequence
! Changes the composition and densities of aquatic organisms and causes a shift to more
tolerant organisms.

Summary of the Ecological Impacts of Urban Stormwater Runoff

Table 4.4 summaries the impacts of urban stormwater runoff on the stream environment.

Ecological impacts of urban runoff and urbanization on streams is wide spread but still not clearly
understood. Models that would link discharges of storm water and physical alteration of the channels and
stream corridors are not available. Linking ecological changes expressed by indices of biotic integrity to
simple determining factors such as degree of imperviousness is persuasive and conclusive, however, the
spread of the data points and the reliability of the simple relationships is not great, especially, in the break
point zone of 10 to 20 percent imperviousness.

Karr and Chu (19997) have pointed out that many water resource managers have turned to two
approaches: multivariate statistics and multimetric indices. Combination of the two leads to usefull
results. Multimetric indexes are designed specifically to document which components of the
biological/ecological system provide strong signals about the impact of humans and to use those signals
to define biological conditions and diagnose the factors likely to have caused degradation when it is
detected.

Karr and Chu also recognize that while the classic ecology studies the systems from the reference
point of natural systems, urbanization is an irreversible change of the system that brings the ecology
of the urban streams to a new level from which a return to the previous natural state is impossible and
unrealistic. Therefore, the basic goal of urban runoff mitigation and restoration of the integrity of the

38
urban stream is the goal of maintaining a balanced, integrated, and adaptive biological system having
the full range of elements and processes expected for the (eco-)region in which the system resides. .
TABLE 4.4
Summary of impacts of urban stormwater runoff on stream ecosystems

39
40
Such new restored systems may not be identical to the original unimpacted state nor to the reference
unimpacted reaches. The degree to which the urban streams should approach such unimpacted states has
not been determined by authorities nor by the public.

41
REFERENCES
Allan, J.D.(1995). Stream Ecology Structure and Function of Running Waters. Chapman & Hall. New
York, NY.

Annable, W.K.(1996). Morphological Relationships of Rural Watercourses in Southern Ontario and

Selected Field Methods in Fluvial Geomorphology. Ontario Ministry of National Resources,
Ontario, Canada.

Arnold, C., P. Boison, and P. Patton. (1983). Evaluation of Catch Basin Performance for Urban
Stormwater Pollution Control: Project Summary, EPA 600/S2-83-043, Municipal Environmental
Research Laboratory, U.S. EPA, Cincinnati, OH.

Aruthurai, R., and K. Arulanandan (1978) “Erosion rates of cohesive soils,“ Journal Hydraul. Div.,
ASCE; 104(HY2):279-283

Bannerman, R. T., R. Dodds, D. Owens, and P. Hughes (1992) Sources of Pollutants in Wisconsin
Stormwater. Grant No. C9995007-01, Wisconsin Dept. of Natural Resources, Madison, WI

Bannerman, R. T., D. W. Owens, and R. B. Dobbs. (1993). Sources of Pollution in Wisconsin

Stormwater. Water Science & Technology. 28, (3/5): 241-259.

Bascombe, A. D., J. B. Ellis, D. M. Revitt, and R. B. Shutes (1990). Development of ecotoxicological

criteria in urban catchments. Water Science and Technology, 22 (10/11): 173-179.

Booth D. and L. Reinelt. (1993). Consequences of urbanization on aquatic system- measured effects,
degradation thresholds and corrective strategies. Proceedings Watershed 93 A National
Conference on Watershed Management, Alexandria, VA. March 21-24, 1993.

Booth, D., and C. Jackson. (1997). Urbanization of aquatic systems: Degradation threshold, storm
detention and limits of mitigation. Journal AWRA, 33(5): 1077-1089.

Booth, P.(1991). Urbanization and natural drainage system-impacts, solutions and prognoses. Northwest
Environmental Journal, 7(1): 93.

Brinson, M.M., B. L. Swift, R. C. Plantico, and J. S. Barclay. (1981). Riparian Ecosystems: Their
Ecology and Status. FWS/OBS-81/17. U.S. Fish and Wildlife Service, Office of Biological
Services, Washington, DC.

Brookes, A. (1988) Restoration and enhancement of engineered river channels: some European
experience. Regulated Rivers: Research & Management 5(1):45-56

Burton, G. A., and B. L. Stemmer. (1988) Evaluation of surrogate test in toxicant impact assessment.
Toxicity Assessment: An International Journal, 3:255-269.

42
Carothers, S. W., and R. R. Johnson. (1971). A Summary of the Verde Valley Breeding Bird Survey,
1970. Completion Report FW/6-10:46-64. Arizona Game and Fish Department. Phoenix, AZ.

Carson, C. W. (1965). The relation of free meander geometry to stream discharge and its geomorphic
implications. American Journal of Science, Vol. 263: 864-885.

Collier, R.H. and J.C. Schmidt. (1996). Dams and Rivers: A Primer on the Downstream Effects of Dams.
U. S. Geological Survey (USGS) Professional Paper Circular Vol. 1126 (June).

Cooper, J. R., J. W. Gilliam, R. B. Daniels, and W. P. Robarge. (1987). Riparian areas as filters for
agricultural sediment. Soil Science Society of American Journal. 51:416-420.

Copper, J. R. and J. W. Gilliam. (1987). Phosphorus redistribution from cultivated fields into riparian
areas. Soil Science Society of America Journal. 51:1600-1604

Cowx, I. G., and R. L. Welcomme. (1998). Rehabilitation of Rivers For Fish. Fishing News Books.
London, UK.

Crawford, J., and D. Lenat. (1989). Effects of Land Use on Water Quality and Biota of Three Streams
in the Piedmont Province of North Carolina. U. S. Geological Survey Water Resources
Investigation Report 89-4007, Raleigh, NC.

Dreher, D. W. (1997). Watershed impacts on stream quyality indicators in Northeastern Illinois. In

Proceedings of A National Symposium: Assessing the Cumulative Impacts of Watershed
Development on Aquatic Ecosystems and Water Quality, Chicago, IL, March 19-21, 1996

Dury, G. H. (1973). Bed width and wave length in meandering valleys. Nature. Vol. 176, No.
31.

Dunne, T., and L. B. Leopold. (1978). Water in Environmental Planning. W. H. Freeman Co. San
Francisco, CA.

Fam, S., M. K. Stenstrom, and G. Silverman (1987) Hydrocarbons in urban runoff, J. Environ. Eng. Div.,
ASCE, 100(EE2):459-477

Federal Emergency Management Agency (FEMA). (1998). Testimony of James L. Witt, Director FEMA,
before the subcommittee on House Water Resources and Environment Committee on
Transportation and Infrastructure, March 26, 1998.

Federal Interagency Stream Restoration Workgroup (1998) Stream Corridor Restoration: Principles,
Processes, and Practices, Washington, DC

Ferguson, B. K. (1998). Introduction to Stormwater; Concept, Purpose, Design. John Wiley & Son,
Inc., New York, NY.

43
Field, R. and R. Pitt. (1990), Urban storm-induced discharge impacts: U.S. Environmental Protection
Agency research program review. Water Science and Technology, 20 (10/11): 1-14.

Fitchko, J. (1986). Literature Review of the Effects of Persistent Toxic Substances on Great Lakes Biota.
Report to the Great Lakes Advisory Board.

Galli, J. (1991) Thermal Impacts Associated with Urbanization and Stormwater Best Management
Practices. Metropolitan Washington Council of Governments, Maryland Department of
Environment, Washington, DC

Garie, H. L. and A. Mcintosh. (1986). Distribution of Benthic macroinvertebrates in a stream exposed

to urban runoff. Water Resources Bulletin, 22: 447-455.

Gregory, K., R. Davis, and P. Downs. (1992). Identification of River change due to urbanization.
Applied Geography, 12:299-318.

Hamelink, J. L., P. F. Lanndrum, H. L. Bergman, and W. H. Benson. (1994). Bioavailability: Physical,

Chemical and Biological Interactions. Lewis Publishers, New York, NY.

Harris, L. D. (1984). The Fragmented Forest. University of Chicago Press, Chicago Il.

Heaney, J.P. and W. C. Huber (1984) Nationwide assessment of urban runoff impacts on receiving water
quality. Water Res. Bull. 20(1):35-42

Hoffman, E. J., G. L. Mills, J. S. Latimer, and J. G. Quinn. (1984). Urban runoff a source of polycyclic
aromatic hydrocarbons to coastal water. Environmental Science & Technology, 18: 580-587

Hynes, H.B.N. (1970) The Ecology of Running Waters. University of Liverpool Press, UK

Johnson, C. D., D. R. Patton, P. F. Folliot, and R. H. Harme. (1977). Riparian Ecosystems and their
Management: Reconciling Conflicting Uses. General Technical Report RM-120. U. S. Forest
Services, Rocky Mountain Forest and Ranger Experimental Station. Fort Collins CO.

Karr, J.R. and E. W. Chu (1997) Biological Monitoring and Assessment: Using Multimetric Indexes
Effectively. EPA 235-R97-001, University of Washingron, Seattle, Washington

Linder, M. W. (1976). Designing for sediment transport. Water Spectrum, Spring-summer: 36-43.

Leopold, L.B. (1968) “Hydrology for urban land planning - A guidebook on the hydrologic effects of
urban land use,” U.S. Geol. Surv. Circ. 554, USGS, Washington, DC

Leopold, L. B. (1994). A View of the River. Harvard University Press. Cambridge, MA.

44
Leopold, L. B., M. G. Wolman, and J. P. Miller. (1992). Fluvial Processes in Geomorphology.
W. H. Freeman and Company, San Francisco, CA.

Macrae, C. (1996). Experience from morphological research on Canadian stream: is control of the two
year frequency storm the best basis for stream channel protection? In Effects of Foundation
Conference Proceedings, Snowbird UT, August 4-9, pp144-160.

Masterson, J. P., and R. T. Bannerman. (1994). Impacts of stormwater runoff on urban streams in
Milwaukee, Wisconsin. Proceedings National Symposium on Water Quality, American Water
Resources Associations, November 1994.

Maxted, J.R. (1996) The use of percent impervious cover to predict the ecological condition of wadeable
nontidal streams in Delaware. In Proc. of a National Symposium - Assessing the Cumulative
Impacts of Watershed development on Aquatic Ecosystems and Water Quality, Chicago, IL,
March 19-21,1996

May, C. R. Horner, J. Karr, B. Mar, and E. Welch. (1997). Effects of urbanization on small streams in
the Puget Sound ecoregion. Watershed Protection Techniques, 2(4): 483-494.

Mehta, A.J. et al. (1989) “Cohesive sediment transport. I . Process description,” Journal Hydraulic
Engng. 115(8):1076-1093

Murchelano, R. A. (1988). Fish as Sentinels of Environmental Health. PB89-139737, National Marine

Fisheries Service, Woods Hole, MA.

Novotny, V., and G. Chesters, (1981). Handbook of Nonpoint Pollution: Source and Management, Van
Nostrand Reinhold Company, New York, NY.

Novotny, V. and H. Olem. (1994). Water Quality: Prevention, Identification, and Management of
Diffuse Pollution, Van Nostrand Reinhold Company, New York, NY.

Novotny, V. and J. Witte (1997) “Ascertaining aquatic ecological risks of urban stormwater
discharges,” Water Research 31(10):2573-2585

Ohmart, R.D. and B.W. Anderson (1986) Riparian habitat. In Inventory and Monitoring of Wildlife
Habitat (A.Y. Cooperrrider, R.J. Boyt, and H.R. Duart, eds.) Pp. 169-201, Bureau of Land
Reclamation, U.S. Dept. of Interior, Denver, CO

Parker, G. and D. Anders. (1976). Detrimental effects of river channelization. Proceedings of Conference
Rivers 76, American Society of Civil Engineers: 1248-1266.

Parkhurst, B.R., W. Warren-Hicks, R.C. Cardwell, J. Volosin, T. Etchison, J. B. Butcher and S. M.

Covington (1996) Aquatic Ecological Risk Assessment: A Multi-tiered Approach. Project 91-
AER-1, Water Environment Research Foundation, Alexandria, VA

45
Partheniades, E. (1972) “Results of recent investigations on erosion and deposition of cohesive
sediments,
In Sedimentation (H.W. Shen, ed.), pp 20-1 to 20-39, Fort Collins, CO

Paterson, A.J. (1996) The Effect of Recreation on Biotic Integrity of Small Streams in Grand Teton
National Park. Msc Thesis, University of Washington, Seattle, WA

Patrick, R. (1975) “Some thoughts concerning correct management of water quality,” In: Urbanization
and Water Quality Control (W. Whiple, Jr., ed.), AWRA, Minneapolis, MN

Pitt, R. E. and P. Bissonnette. (1983). Bellevue Urban Runoff Program. Summary Report PB84-237213,
U. S. Environmental Protection Agency, Water Planning Division, Washington, DC.

Pitt, R. E., and P. Barron. (1989). Assessment of Urban and Industrial Stormwater Runoff Toxicity and
Evaluation/Development of Treatment Runoff Toxicity Abatement Phase-1. U.S. Environmental
Protection Agency, Office of Research and Development, Edison, N.J.

Pitt, R. E., (1991). Small Storm Hydrology: The Integration of Flow With Water Quality Management
Practices. University of Wisconsin - Madison Dept. of Engineering Professional Development
short course Designing Storm Water Management Practices Madison, WI.

Pitt, R. e. (1995). Biological effects of urban runoff discharges. Stormwater Runoff and Receiving
Systems: Impacts, Monitoring, and Assessment. Lewis Publisher, New York, NY.

Plafkin, J. L., M. T. Barbour, K. D. Poter, S. K. Gross, and R. M. Hughes. (1989). Rapid Bioassessment
Protocols for Use in Streams and Rivers: EPA/144/4/89-001. U. S. Environmental Protection
Agency (EPA), Washington, DC.

Platts, W.S. (1974) Geomorphic and Aquatic Conditions Influencing Salmonids and Stream
Classifications - with Application to Ecosystem management. U.S. Department of Agriculture,
SEAM Program, Billings, Montana

Riley, A. L. (1998). Restoring Streams in Cities: A Guide for Planners, Policy-Makers and Citizens,
Ireland Press.

Robinson, A. M. (1976). Effects of urbanization on stream morphology. Proceedings National

Symposium Urban Hydrology, Hydraulics, Sediment Control, B-27-B-39.

Ryckborst, H. (1980). Geomorphological changes after river meander surgery. Geologie en Mijboun,
59:121-128.

Schamberger, M., A. H. Farmer, and J. W. Terrel. (1982). Habitat Suitability Index Models:
Introduction. FWS/OBS-82/10. U. S. Fish and Wildlife Service, U. S. Department of the Interior.
Washington, DC.

46
Schueler, R. L. (1994). The importance of imperviousness. Watershed Protection Techniques,
1(3):100-111.

Schueler, T. and J. Galli. (1992). Environmental impacts of stormwater ponds. Stormwater Runoff and
Receiving Systems: Impacts, Monitoring and Assessment. Lewis Publisher, New York, NY.

Scott, J. B., C. R. Steward, and Q. J. Stober. (1982) Impacts of Urban Runoff on Fish Populations in
Kelsey Creek, Washington. Contract No. R806387020, U. S. Environmental Protection Agency,
Environmental Research Laboratory, Corvallis, OR>

Simmons, D., and R. Reynolds. (1982) Effects of Urbanization on baseflow of selected south shore
streams, Long Island, N. Y. Water Resources Bulletin, 18(5): 797-805

Simons, D. B. and F. Senturk. (1977). Sediment Transport Technology. Water Resources Publications,
Fort Collins, CO.

Sparks, R. (1995) Need for ecosystem management of large riers an their floodplains. Bioscience
45(3):170

Steedman, R. J. (1988). Modification and assessment of a index of biotic integrity to qualify stream
quality in Southern Ontario. Canadian Journal of Fisheries and Aquatic Science, 45:492-501

Sweeney, B.W. (1992) Streamside forests and the physical, chemical, and trophic characteristics of
piedmont streams in eastern North America, Water Science & Technol 26:1-12

Tanizaki, Y., T. Shimokawa, and M. Yamazaki. (1992). Physico-chemical speciation of trace elements
in urban streams by size fractionation. Water Research, 26(1): 55-66.

Thayer, R. L., and T. Westbrook. (1990). Open drainage systems for residential communities: case
studies from California’s Central Valley. In Proceedings of the 1989 Conference of the Council
of Educators in Landscape Architecture. Landscape Architecture Foundation, Washington, DC.

Thomas, J. W. (1979). Wildlife Habitat in Managed Forests: In the Blue Mountains of Oregon and
Washington. U. S. Department of Agricultural, Forest Service. Ag Handbook 553.

Thon, H. M. (1992). Stormwater regulations and CSO strategy in the USA. In Implementation of
Pollution Control Measures for Urban Stormwater Runoff. University Toronto Press, Toronto,
Ont.

Trimble, S. (1997). Contribution of stream channel erosion to sediment yield from an urbanizing
watershed. Science, 278:1442-1444.

U. S. Environmental Protection Agency (EPA). (1984). Report to Congress: Nonpoint Source Pollution
in the U.S. Office of Water Program Operations, U. S. EPA, Washington, DC.

47
Vannote, R. L., G. W. Minshall, K. W. Cummins, J. R. Sedell, and C. E. Cushings. (1980). The river
continuum concept. Canadian Journal of Fishery Science. 37(1): 130-137.

Walesh, S. G. (1989). Urban Surface Water Management. John Wiley & Son, Inc. New York, NY.

Wang, L. J. Lycons, P. Kanehl, and R. Gatti. (1997). Influences of watershed land use on habitat quality
and biotic integrity in Wisconsin. Fisheries, 22(6): 6-12.

Welch, E. B. (1992). Ecological Effects of Wastewater: Applied Limnology and Pollution Effects.
2nd Edition. Chapman & Hall. London, UK.

Williams, G.W. (1978) Bankfull discharge of rivers. Water Resources Research 14:1141-1154

Wolman, M. G. (1964). Problems Posed by Sediments Derived from Construction Activities in

Maryland. Maryland Pollution Control Commission, Baltimore, MD.

Wolman, M. G., and L. B. Leopold. (1957). River Floodplains, Some Observations on Their Formation.
U. S. Geological Survey (USGS) Professional Paper No 283-C: 86-109.

Wotten, R. S. (1994). The Biology of Particles in Aquatic Systems, 2nd Edition. Lewis Publishers. Ann
Arbor, MN.

Yang, C. T. (1996). Sediment Transport theory and Practice. McGraw-Hill, New York, NY.

Ziegler, C.K. and W. Lick (1988) “The transport of fine-grained sediments in shallow waters,” Environ.
Geol. Water Sci., 11(1):123-132

48