&EPA
United States
Environmental Protection
Agency
2012 GREEN INFRASTRUCTURE TECHNICAL ASSISTANCE PROGRAM

                            Pittsburgh UNITED

                         Pittsburgh, Pennsylvania
   Addressing Green Infrastructure Design Challenges

   in the  Pittsburgh  Region

   Abundant and Frequent Rainfall
   Photo: Rain Garden at Biddle Building
   Source: Nine Mile Run Watershed Association
                                                   January 2014
                                                 EPA 800-R-14-003

-------
About the Green Infrastructure Technical Assistance Program

Stormwater runoff is a major cause of water pollution in urban areas. When rain falls in undeveloped
areas, the water is absorbed and filtered by soil and plants. When rain falls on our roofs, streets, and
parking lots, however, the water cannot soak into the ground. In most urban areas, stormwater is
drained through engineered collection systems and discharged into nearby waterbodies. The
stormwater carries trash, bacteria, heavy metals, and other pollutants from the urban landscape,
polluting the receiving waters. Higher flows also can cause erosion and flooding in urban streams,
damaging habitat, property, and infrastructure.

Green infrastructure uses vegetation, soils, and natural processes to manage water and create healthier
urban environments. At the scale of a city or county,  green infrastructure refers to the patchwork of
natural areas that provides habitat, flood protection, cleaner air, and cleaner water. At the scale of a
neighborhood or site, green infrastructure refers to stormwater management systems that mimic
nature by soaking up and storing water. These neighborhood or site-scale green infrastructure
approaches are often referred to as  low impact development.

EPA encourages the  use of green infrastructure to help manage stormwater runoff. In April 2011, EPA
renewed its commitment to green infrastructure with the release of the Strategic Agenda to Protect
Waters and Build More Livable Communities through Green Infrastructure. The agenda identifies
technical assistance  as a  key activity that EPA will pursue to accelerate the implementation of green
infrastructure.

In February 2012,  EPA announced the availability of $950,000 in technical assistance to communities
working to  overcome common barriers to green infrastructure. EPA received letters of interest from
over 150 communities across the country, and selected 17 of these communities to receive technical
assistance.  Selected  communities received assistance with a range of projects aimed at addressing
common barriers to  green infrastructure, including code review, green infrastructure design, and cost-
benefit assessments. Pittsburgh UNITED was selected to receive assistance developing fact sheets and
technical papers to provide solutions for site conditions that are perceived to limit green infrastructure
applicability.

For more information, visit http://water.epa.gov/infrastructure/greeninfrastructure/gi  support.cfm.

-------
Acknowledgements
Principal EPA Staff
Kenneth Hendrickson, USEPA Region 3
Dominique Lueckenhoff, USEPA Region 3
Christopher Kloss, USEPA
Tamara Mittman, USEPA

Community Team
Jennifer Rafanan Kennedy, Clean Rivers Campaign
Sara Powell, Nine Mile Run Watershed Association

Consultant Team
Dan Christian, Tetra Tech
Valerie Novaes, Tetra Tech
Anne Thomas, Tetra Tech

Technical Review Team
Beth Dutton, 3 Rivers Wet Weather
Kari Mackenbach, URS Corporation
Jim Pillsbury, Westmoreland Conservation District
This report was developed under EPA Contract No. EP-C-11-009 as part of the 2012 EPA Green
Infrastructure Technical Assistance Program.

-------
Contents
Introduction	1
Rainfall and Stormwater Management Overview	1
  Stormwater Design Criteria	2
  Rainfall-Runoff Definitions	3
Rainfall in the Greater Pittsburgh Area	4
Methods to Address Abundant and Frequent Rainfall	8
  Design for Specific Criteria	8
    1.  Overflow and Bypass Systems	9
    2.  Underdrains	10
    3.  Soil Investigation	12
  Modeling Tools	12
Examples of Implemented Projects	13
  Sterncrest Drive Bioswale and Rain Gardens, built 2007, Cuyahoga County, OH (Darner and
  Dumouchelle, 2011)	13
    1.  Design Summary	13
    2.  Results Summary	15
    3.  Lessons Learned	15
  Michigan Avenue Bioretention Facilities, Lansing, Ml	16
    1.  Design Summary	16
    2.  Results Summary	17
    3.  Lessons Learned	18
References	19
  Manuals, Articles, and Books	19
  Websites	20
                                              IV

-------
Tables
Table 1. Design Storm Event Depths (inches) for Pittsburgh, PA	7

Figures
Figure 1. Dimensionless Unit Hydrograph and Equivalent Triangular Hydrograph	4
Figure 2. Average Annual Precipitation in Pennsylvania	5
Figure 3. Southwest Pennsylvania Annual Rainfall Grouped by Storm Depth	5
Figure 4. Distribution of Total Annual Precipitation by Month for Pittsburgh, PA	6
Figure 5. Average Annual Precipitation Depths for the United States	6
Figure 6. Rainfall Depth Comparisons for Four Storm Events in Selected Cities	7
Figure 7. Off-line and On-Line Bioretention Systems	9
Figure 8. Example Bioretention System with an Elevated Underdrain	10
Figure 9. Typical Cross-Sections of Permeable Paver System with Underdrains	11
Figure 10. Roadside Rain Garden in Cuyahoga County	13
Figure 11. Rain Garden and Bioswale Typical Cross-Sections	14
Figure 12. Typical Bioretention Cross-Section on Michigan Avenue	16
Figure 13. Photo Sequence of Bioretention on Michigan Avenue	17
Figure 14. Runoff Comparison 1-year 1-hour	18

-------
Introduction
In many areas of the United States, urban stormwater impacts are largely driven by the small, frequent
storms that account for most of the annual precipitation. While these storms generate little runoff in
natural areas, they generate significant runoff in urban environments, where paved surfaces and dense
drainage networks prevent water from following natural pathways. Stormwater generated by these
small storm events carries the first flush of pollutants into local waterbodies and leads to higher flows in
local streams. The cumulative impact of these frequent events can drive many of the physical, chemical,
and biological impacts of urban stormwater on local waterbodies.

Green infrastructure is the practice of mimicking and restoring natural hydrologic processes within the
built environment to mitigate the impacts of urban stormwater. Green infrastructure is an important
design strategy for protecting water quality while also providing multiple community benefits. Common
green infrastructure practices include permeable pavement, bioretention facilities, rain barrels, tree
boxes, and green roofs.  These practices can complement conventional stormwater management
practices by enhancing infiltration, storage, and evapotranspiration and managing runoff at its source.

This paper will address the concern that green infrastructure is not appropriate for the Pittsburgh area's
humid climate, which is characterized by abundant rainfall and frequent storm events. The paper will
analyze the greater Pittsburgh area's  typical rainfall pattern; provide  design guidance to help size
effective green infrastructure practices; and describe effective projects from areas around the country
with similar rainfall patterns. The goal of this paper is to  provide recommendations for green
infrastructure design that are based on facts,  research, and engineering in order to help practitioners
make informed decisions regarding the use of green infrastructure to manage stormwater in areas with
abundant rainfall.
Rainfall and Stormwater Management Overview

Among the impacts of urban stormwater in the greater Pittsburgh area are increased flooding and
combined sewer overflows (CSOs). Pennsylvania is one of the most flood-prone states in the country (PA
SW BMP Manual, 2006). The state experiences flooding problems not only from large tropical storms
but from smaller storms as well. These flooding problems are exacerbated by stormwater runoff from
urban areas, where large amounts of impervious cover generate increased runoff volumes and rates.
Urban stormwater also contributes to CSOs. The greater Pittsburgh area is served by a combined sewer
system that carries both sewerage and stormwater to wastewater treatment plants. When too much
stormwater enters the system, the wastewater treatment plants cannot treat all of the flow, and some
untreated wastewater must be diverted into local waterbodies.

Green infrastructure can  help mitigate these urban stormwater impacts by managing small storms on
site. Retaining small events on site  reduces runoff volumes and rates, protects stream channels, and
effectively adds capacity to combined sewer systems. Many communities across the country are
integrating green infrastructure into their CSO control programs, including Chicago, IL; Kansas City, MO;
New York City, NY; Philadelphia, PA; Syracuse, NY; Portland, OR; and  Milwaukee, Wl. These communities
are using green infrastructure to supplement the storage and treatment capacity provided by more
traditional, "gray infrastructure" approaches.

-------
One of the challenges to the use of green infrastructure in the greater Pittsburgh area is the perception
that the humid climate, characterized by abundant and frequent rainfall, is inappropriate for green
infrastructure. Some stormwater professionals express
concern that large rainfall events or smaller, more frequent
events will overwhelm green infrastructure systems.           Green Infrastructure Performance
Experience demonstrates, however, that even in areas with    under Extreme Conditions - An
abundant rainfall, green infrastructure can consistently        instrumented bioretention practice
retain design storms on site, preventing a significant portion    installed in Cambria, Queens
of the annual  runoff from entering the sewer system. The      provided the opportunity to assess
design of effective green infrastructure practices requires      sYstem Performance during
that practices are sized to manage the runoff volume from a    Superstorm Sandy (October 2012)
range of design storm events. Once the appropriate sizing is    and Hurricane Irene (August 2011).
determined, simple design features can be incorporated to     Analysis of the data collected
optimize system performance, including overflow structures    indicated that the practice retained
and underdrain pipes. Overflow structures provide a           most of the runoff from its drainage
pathway for runoff volumes that exceed the design capacity    area durinS both extreme events.
of the green infrastructure practice, while underdrain pipes    The Practice retained 79% of the of
allow the green infrastructure practice to dewater in time      the 6A inches of rainfa" associated
for the next storm event.                                   with Hurricane  Irene, and 100% of
                                                         the 1.3 inches of rainfall associated
Additional design considerations include landscape water      with Superstorm Sandy. (Montalto
requirements and scouring. Thought must be given to         et a'v 2013)
vegetated systems to ensure that  plants are provided a
suitable amount of water to survive - not too much and not
too little. In addition, inflow velocities and energy dissipation at the inlet must be considered to prevent
scour of the soil.

Stormwater Design Criteria

Stormwater design criteria typically require the management of a range of design storm events to
mitigate a range of receiving water impacts. Management of larger storm events is required to mitigate
flooding, while management of smaller storm events is required to mitigate channel erosion  and water
quality degradation. Green infrastructure practices are generally designed to meet water quality criteria
and to help meet channel protection criteria. Specific design criteria vary for each municipality across
the country, but some  general  patterns may be observed.

For water quality criteria, common requirements include treatment or retention of a certain  volume of
runoff. Many water quality criteria address small storm events that are exceeded once to several times
per year. For  example, Nashville, TN requires treatment of the runoff from the first 1.1 inches of rainfall
to remove 80  percent of total suspended solids, while Albany, NY requires treatment of 90 percent of
the average annual runoff volume.

For channel protection and flood mitigation criteria, common requirements include control of the post-
development  peak runoff rate to match the pre-development peak runoff rate, and control of the post-
development  runoff volume to match the pre-development runoff volume.  Many channel protection
criteria address the one- to two-year 24-hour storm events while flood mitigation criteria commonly
address  the 10- to 100-year 24-hour storm events.  For example, Washington, DC requires post-
development  control of both the peak runoff rate and runoff volume to match pre-development
conditions for the 1-, 2-, 10-, and 100-year, 24-hour storm events.

-------
Rainfall-Runoff Definitions
The following are important definitions related to rainfall and stormwater runoff.

Contributing Drainage Area (CDA) - The drainage area that contributes to a green infrastructure
practice.

Depth-Duration-Frequency Curve - Describes rainfall depth as a function of duration for given return
periods.

Inter-Event Time - The minimum number of dry hours between separate storm events. Typically values
ranging from 3 to 30 hours are used to separate rainfall events. A 6-hour inter-event time is commonly
assumed. This is used when evaluating the rainfall statistics of an area. For example, the annual 90th
percentile non-exceedance rainfall event will depend on the inter-event time.

Peak Discharge Rate-The maximum instantaneous rate of flow (volume of water passing a given point
over a specific duration, such as cubic feet per second) during a storm, usually in reference to a specific
design storm event.

Rainfall Intensity - A measure of the amount of rain that falls over a given time period.

Rainfall Duration - The amount of time over which  a rainfall event occurs. Typically presented from 30
minutes to 24 hours.

Rainfall Distribution -The variation of rainfall intensity overtime. An SCSType II rainfall distribution is
commonly used for the greater Pittsburgh area. It is also referred to as a rainfall hyetograph.

Runoff Coefficient - A dimensionless coefficient relating the amount of runoff to the amount of
precipitation. The value is higher for areas with low infiltration (pavement, steep gradients) and lower
for permeable, well vegetated areas.

Runoff Hydrograph - A plot of discharge versus time. The area under the hydrograph represents volume
of water. Figure 1 shows a unit hydrograph.

Stormwater - Water consisting of precipitation runoff or snowmelt.

Stormwater Pollutants - Typical stormwater pollutants include sediment, nutrients, temperature,
bacteria, trash/debris, and toxic contaminants/heavy metals. The specific pollutants generated will vary
depending on the land use.

Storm Recurrence Interval - The probability that a give storm event will be equaled or exceeded in any
given year. For example, a two-year storm  event occurs on average once every two years  or statistically
has a 50 percent change  of occurring in a given year. Green infrastructure practices are sized for specific
storm recurrence intervals.

Unit Hydrograph - Defined as the runoff hydrograph that results from 1 inch of effective rainfall
generated uniformly over a watershed, at a constant rate, and over a specified period of time. Figure 1
shows a dimensionless unit hydrograph and its components.

-------
              Excess rainfall
                                     Source: NEH, Chapter 16
              Figure 1. Dimensionless Unit Hydrograph and Equivalent Triangular Hydrograph
Rainfall in the Greater Pittsburgh Area
The distribution of rainfall events in the Pittsburgh area is well-suited to green infrastructure practices.
The National Oceanic and Atmospheric Association (NOAA) records and publishes long term
precipitation data for the 50 contiguous states. In Pennsylvania, the average annual precipitation
amount ranges from 37 inches to more than 45 inches per year (Figure 2), with nearly all of the annual
rainfall occurring in small storm events. Precipitation of an inch or less is the most common rainfall
event in southwest Pennsylvania and accounts for approximately 75% of annual storm events (Figure 3).
These precipitation events are also distributed evenly throughout the year (Figure 4). As little as one-
tenth of an inch of rain or snowmelt can cause sewage to overflow into  Pittsburgh's rivers and streams
(3 Rivers Wet Weather, 2013). Since green infrastructure practices are designed to manage small
storms on site, these rainfall characteristics are well-suited to green infrastructure controls.

When compared to other regions of the United States, annual precipitation and design storm depths in
the Pittsburgh area are similar to those observed in many areas east of the Mississippi. Figure 5 shows
average annual precipitation values for the contiguous United States. Pittsburgh is not unique in the
amount of annual precipitation it receives, with much of the area east of the Mississippi receiving at
least as much precipitation as Pittsburgh.

-------
                       Pennsylvania
                    30 Year (19711-2000)
            Mean Annual Precipitation (inches)
Annual
Precipitation
(Inches)
1 h'.-t
| 31.5-32
1 | 32.1 - 33
^ 33.1 - 34
| 34.1 - 35
| 35.1 - 36
EH«.i
H| 371
^B 33.1
B|40.1
1 141.1
1 142.1
cn«.«
-37
-33
-40
-41
-42
-43
-44
B
44.1
4S.1
46. 1
47.1
48.1
49.1
50.1
51.1
• 45
• 46
-47
-43
-49
• 50
-61
•52
•152.1
r i54.i
1 I5S.1
I 1 67.1
1 158.1
I 1 69.1
-S3
-54
•55
-56
-57
-58
-59
-60
               Source: NOAA National Weather Service
        Figure 2. Average Annual Precipitation in Pennsylvania
                       < 1%
                                           • 0.0-0.01"
                                           • 0.01-0.1"
                                            0.1-0.25"
                                           • 0.25-0.5"
                                           • 0.5-1.0"
                                           = 1.0-1.5"
                                           • 1.5+"
        Adapted from Westmoreland Conservation District, 2013
Figure 3. Southwest Pennsylvania Annual Rainfall Grouped by Storm Depth

-------
                 Precipitation by Month for Pittsburgh, PA
01
.c
u
_g
4-*
as
u
01
        J*
       
-------
Design storms are generally described in terms of depth, duration and frequency (presented in depth-
duration-frequency curves). Design storm depths for much of the United States can be obtained from
NOAA Atlas 14, which provides precipitation frequency estimates for 5-minute through 60-day
durations at average recurrence intervals of 1-year through 1,000-year. Precipitation depths in the
Pittsburgh area are based upon statistical analyses of precipitation data collected at a gauge near the
Pittsburgh International Airport (Table 1).

Comparing the magnitude of four common design storms for several cities in the eastern United States
demonstrates that design storm depths in Pittsburgh are typical of design storm depths in much of the
eastern United States (Figure 6). Indeed, Pittsburgh is observed to have the smallest design  storm
depths of all the cities sampled.
Table 1. Design Storm Event Depths (inches) for Pittsburgh, PA
Avg. Recurrence
Interval
1-year
2-year
5-year
10-year
Rainfall Depth (inches) for Recurrence Intervals
30-min
0.79
0.96
1.19
1.36
60-min
0.97
1.18
1.49
1.73
2-hour
1.11
1.34
1.69
1.96
3-hour
1.17
1.42
1.78
2.07
6-hour
1.41
1.70
2.12
2.46
12-hour
1.67
1.99
2.46
2.84
24-hour
1.96
2.33
2.85
3.27
Source: NOAA Atlas 14
            Rainfall  Depth Comparison for Selected Cities
                      (24-hour Duration Storm Event)
                                                                    ^^DC, Washington
                           2y               5y

                             Storm Event (24-hour duration)
lOy
                                                                        MD, Baltimore
                                                                        NY, New York City
                                                                        OH, Cincinnati
                                                                        OH, Cleveland
                                                                        PA, Philadelphia
                                                                        PA, Pittsburgh
                                                                        VA, Richmond
              Figure 6. Rainfall Depth Comparisons for Four Storm Events in Selected Cities

-------
Methods to Address Abundant and Frequent Rainfall
The key to the effectiveness of any stormwater control, green or gray, is the performance criteria to
which the control is designed.  Determining appropriate design criteria is beyond the scope of this
document. Once the design criteria are known, however, sizing and designing green infrastructure
practices to meet those criteria is relatively straightforward. The Pennsylvania Department of
Environmental Protection Stormwater Best Management Practices Manual is available on-line as
guidance for designing green infrastructure practices. The remainder of this section reviews important
design considerations for areas with abundant and frequent rainfall, and discusses the use of modeling
tools to meet design criteria.

Design  for Specific Criteria

Prior to the design of any green infrastructure practice, the designer should identify the applicable
stormwater criteria.  This is often a local issue, and the designer should check with the appropriate
agencies  to understand the specific requirements. Municipalities may have different requirements
based on whether an area is served by a combined sewer system or a municipal separate storm sewer
system (MS4).

Much of the greater  Pittsburgh area is served by combined
sewers. For green infrastructure design in these areas, the        Combined Sewer Overflow - A
design objective is to reduce the frequency and volume of        combined sewer overflow (CSO)
combined sewer overflows. 3  Rivers Wet Weather has           occurs when a combined sewer
developed the "RainWays" tool to allow designers to            system, a system that carries both
determine the impact of green infrastructure practices on        stormwater and wastewater in the
overflow volume and frequency. The tool can be accessed        same pipe; becomes overwhelmed
from the 3 Rivers Wet Weather website,                        and overflows into the nearest
www.3riverswetweather.org.                                  waterway. This can occur during
                                                           periods of rain or snowmelt, when
For green infrastructure design in MS4 areas, the Pennsylvania    the vo|ume of water insjde the
Stormwater Best Management Practices Manual provides        sewer pjpe exceeds the capacity of
recommendations for design criteria. The manual recommends    the system to  transport it to the
that stormwater practices be sized to maintain the post-         treatment olant
development runoff  volume for all storms less than or equal to
the 2-year, 24-hour event, or to capture the first two inches
(2") of runoff from all contributing impervious surfaces. At a minimum, the first one inch (1") of runoff
from new impervious surfaces must  be permanently removed from the runoff flow (i.e. shall not be
released  into surface waters).

Sizing green infrastructure practices  to meet volume-based criteria is relatively straightforward. The
practice should be sized such that the storage capacity equals the volume of runoff from the tributary
drainage area. The storage capacity should include ponding area, storage within the soil and aggregate
layers, and infiltration rate. Note that different practices function best with different tributary areas.
Bioretention/infiltration practices work best with a tributary area to infiltration area ratio of no more
than 5:1. Permeable  pavement works best when the tributary drainage area is limited to paved surfaces
and when the tributary area to permeable area ratio does not exceed 2:1. Many models are available to
help size  green infrastructure  practices. While a discussion of available models is beyond the scope of
this document, some considerations for selecting an appropriate model are discussed in Section
"Modeling Tools."

-------
I.  Overflow and Bypass Systems
Green infrastructure practices should always be designed with an overflow or bypass system, regardless
of the specific design criteria used for sizing the facility. These systems provide a pathway for runoff
volumes that exceed the design capacity of the green infrastructure practice, maintaining the practice
integrity and preventing excess  ponding. This is particularly important in areas with frequent rainfall to
ensure that stormwater is redirected safely if the green infrastructure practice has not completely
drained from a previous rainfall event.

The type of system selected depends on whether the practice is on-line or off-line. On-line practices
provide stormwater control within the runoff flowpath, and receive runoff from all storms. Off-line
practices, in contrast, provide stormwater control away from the runoff flowpath, and only receive
runoff until their design capacity is exceeded (Figure 7).  Overflow systems are included in on-line
practices, while bypass systems are included in off-line practices.   Note that off-line practices are better
protected than on-line practices from erosive velocities and water damage caused by larger storm
events.  On-line practices can also work very well, however, care must be taken to minimize the risk of
excessive velocities and transport of mulch downstream.

To ensure that the overflow or bypass system protects property and maintains safety, the full build-out
100-year, 24-hour design storm should be routed through the green infrastructure practice and the
effects on the system, adjacent  property, and downstream areas should be assessed. Even though the
green infrastructure practice may be designed for a smaller storm event, the overall site should be
designed to safely pass the flows resulting from the full build-out 100-year storm event as much as
practicable. Refer to local standards for variations in this standard.
Source: Tetra Tech
                 Off-Line System
Water enters the bioretention area from a curb cut.
Once the ponding area is full to the level of the gutter,
stormwater will not enter the area but will be conveyed
down the gutter to a catch basin.
Source: Tetra Tech
                 On-Line System
Water enters the bioretention area from a curb cut. An
overflow structure is placed within the bioretention
area to convey flows in excess of the design flow.
                          Figure 7. Off-Line and On-Line Bioretention Systems

-------
2.  Underdrains
In addition to an overflow or bypass system,
installation of an underdrain can be important to
meeting acceptable dewatering times and in preparing
for the next storm event.  Underdrains may be
necessary if the measured permeability of the
underlying soils is less than 0.5 inches/hour,
particularly in areas subject to frequent storm events.
For a green infrastructure practice to be successful in
treating stormwater there must be capacity for runoff
from the next storm event. When designing a green
infrastructure practice, designers should verify soil
permeability (see "Soil Investigation" section  below). The underdrain may outlet to a suitable location
such as a common space area, stream valley,  drainage swale, roadside open channel, or an existing
enclosed drainage system. Typically a 4-inch or 6-inch perforated underdrain is placed at an elevation to
ensure the required dewater time is met while still promoting infiltration through the bottom of the
system. Figure 8 shows an elevated underdrain pipe configuration in a bioretention system and Figure 9
shows permeable paver designs with an underdrain.
Dewater Time - The dewater time is
defined as the time it takes to drain
the practice. It can be divided into
surface dewater time and complete
cross-section dewater time. Refer to
the PA BMP Manual for
recommendations on dewatering
times.
                    PONDING ELEVATION
         NATIVE GRASSES
            AND SHRUBS
         PRETREATMENT
           *67 STONE
       DEPTH = ? MIN.
       INFILTRATION SUMP
      2-3' MULCH
      1n i—in—i ii—ni-
      FILTER MEDIA
              24' MINI
                STABILIZED OUTFALL

      2" CHOKER STONE
                                                          4-ff PERFORATED
                                                          LJNDERDRAIN
                  Source: Draft District of Columbia Stormwater Management Guidebook
                   Figure 8. Example Bioretention System with an Elevated Underdrain
                                              10

-------
                          2 1/2" PERVIOUS PAVING SURFACE COURSE

                           1" CHOKER COURSE (AASHTO NO. 57 - WASHED)
CLEAN WASHED
UNIFORMLY GRADED
COARSE AGGREGATE
AASHTO NO. 3
PERF. HOPE
ALONG BOTTOM
                   \  I
                       NONWOVEN
        UNCOMPACTED  GEOTEXTILE
        BED BOTTOM
                                           \
                                            6" WIDE CONCRETE
                                         7;;,WEIR, HEIGHT VARIES
 PROVIDE 12" SEDIMENT TRAP
 {BELOW LOW-FLOW ORIFICES) W/
 1" WEEP HOLES, THREE PER SIDE
PLACE CB ON 6"
COMPACTED 2AW/
COMPACTED SUBGRADE
                                         Concrete Pavers

                                                   Permeable Joint Material
                                                      Open-graded
                                                      Bedding Course

                                                        Open-graded
                                                        Base Reservoir

                                                           Open-graded
                                                           Subbase
                                                           Reservoir

                                                             Under drain
                                                             (as required)

                                                      Optional Geotextile
                                                      Under Subbase

                                              Uncompacled Subgrade Soil
      Source: Pennsylvania Stormwater BMP Manual, Chapter 6.4.1 BMP Pervious Pavement with
      Infiltration Bed, and Virginia OCR Stormwater Design Specification No. 1 Permeable Pavement
           Figure 9. Typical Cross-Sections of Permeable Paver System with Underdrains
                                     11

-------
3.  Soil Investigation
Site evaluation and soil infiltration testing should be completed early in the site planning and design
process.  Prescreening may be conducted to identify preliminary sites for green infrastructure practices.
Once preliminary sites are proposed, further investigation at the location of each proposed practice is
recommended. Even if the soil is expected to have a low capacity for infiltration, accounting for the
removal of runoff through infiltration may decrease the required size of the practice.  Suggested
methods to obtain information include conducting test pits and soil infiltration tests per Appendix C of
the Pennsylvania Stormwater Best Management Practices Manual.

Modeling Tools

A range of models are available to assess the water quality and water quantity impacts of green
infrastructure practices. Models range from very simple tools with minimal data input required, to more
complex tools that require expertise and time. In general, complex models have many variables and
parameters that need to be defined.

Models are a simulation of a real or theoretical situation which has parameters a user can alter.
Modeling can simulate scenarios to  predict outcomes of different design conditions. In the context of
green infrastructure practice design, models can calculate the expected runoff from a user input rainfall
record and compare results before and after construction of green infrastructure.

It is important to choose the simplest model that will satisfy the project/design objectives. Some models
perform calculations only for discrete storm events and others allow for continuous long-term
simulations. Continuous long-term simulations allow the designer to account for the effects of back to
back storm events (antecedent conditions) on the green infrastructure practice and evaluate long term
conditions. Discrete storm event simulations, in contrast, model one event and do not account for
changes in soil moisture at the start of an event.

Some considerations when choosing a model include:

    •  Scale of project (e.g. site-level,  neighborhood, community, watershed)

    •  Design criteria (e.g. peak flow attenuation and/or volume reduction)

    •  Availability of various model input parameters for a  project (inflow hydrograph, soil type,
       topographic info.)

    •  Desired outputs (outflow hydrograph, volume reduction, pollutant removal, evapotranspiration
       loss, infiltration loss)

    •  Level of expertise required to perform  modeling

Refer to the US EPA's website for more information on modeling tools available,
http://water.epa.gov/infrastructure/greeninfrastructure/gi  modelingtools.cfm.
                                              12

-------
Examples of Implemented Projects
Sterncrest Drive Bioswale and Rain Gardens, built 2007, Cuyahoga County, OH
(Darner and Dumouchelle, 2011)

In 2007, the Chagrin River Watershed Partners with a grant from the U.S. EPA replaced 1,400 feet of
roadside ditch with grassed bioswale and nine rain gardens on Sterncrest Drive near Cleveland, Ohio to
alleviate flooding problems in the road and yards. The U. S. Geological Survey (USGS) then monitored
the site from 2008-2010 to better define the effect of green infrastructure on stormwater runoff. This
project demonstrates the ability of green
infrastructure to retain stormwater in an area
with rainfall patterns similar to Pittsburgh's.

I.  Design Summary
To address historical flooding at the site, the
ditch was replaced with a series of bioswales
interspersed with rain gardens (Figure 10) and
overflow structures. The  bioswale and rain
garden system was designed to capture the
runoff from the drainage area for a 0.75-inch
rainfall event. The drainage area  is made up of
roadway and a single-family residential area,
which discharges directly to the on-line green
infrastructure system.
                                                     Source: Darner and Dumouchelle, 2011
                                               Figure 10. Roadside Rain Garden in Cuyahoga County
The soil at the Sterncrest Drive site is
predominately a clay-rich till with low
permeability. The water table is near the ground
surface during wet periods of the year. With clay soil and a high water table, the soil provides little
capacity to store infiltrated water so overflow structures and perforated underdrains were included in
the design. The underdrains and overflow structures discharge to a downstream stormwater pipe.  Both
the bioswale and rain gardens were excavated and backfilled with a gravel storage layer and engineered
soil mixture for better performance (Figure 11).
                                             13

-------
                                           4-in. plastic edging

                                              Sod
 Native materi

            3ir
                       -.;,
   tv '
   . '\.',VA''<
   ?:• ">.:,'!:
                       /•":''»":'' '•>'."'' .;c
$.$*•%'£&
 *^S&
* •••• ••••'"• (t>*.
 •, ,•• i'
~' .<• , ••. '•. * y , *i
^ ;iSlS:
' '•'  «" L ..*! •)*«  ^
•' "••,'• >.' •"..";.
„.:•,...:»:.'..i-'v  4
'i '•.'_jJl'..''-!il|i''' —
           4 in. of sand
                              a	
                               18 in.

          Typical rain garden  detail
   Backfill,70% sand/
           30% compost

   Filter fabric

8-15 in. perforated drain

   Limestone gravel

   9 in. of tilled soil
                               12 in.
  Asphalt pavement
2ft 6 in. minimum depth
     Native material
             Sin
             3 in

           4 in. of sand
                               18 in.

            Typical swale detail
                        Backfill,70% sand/
                                30% compost

                        Filter fabric
                     8 or 15 in. perforated drain

                        Limestone gravel
                        9 in. of tilled soil
             Figure 11. Rain Garden and Bioswale Typical Cross-Sections
                                 14

-------
2.  Results Summary
Rainfall and runoff data were collected along with overflow data to determine the frequency of
overflows for rainfall events exceeding the design storm depth of 0.75-inch.

    •   Numerous rainfall events greater than 0.75-inch were retained and infiltrated by the bioswales
       and rain gardens.

    •   Over the three years of monitoring, the system only overflowed 22 times. Of the 22 recorded
       overflow events, 13 occurred when rainfall in the previous 24 hours exceeded the design storm
       depth, seven occurred when there was rainfall within the previous 96 hours, and two were
       unexplained overflow events occurring six hours apart.

    •   The bioswales and rain gardens performed better than expected in that there were more rainfall
       events greater than 0.75-inch that did not cause an overflow than events that caused an
       overflow.

3.  Lessons Learned
Although the project site near Cleveland, Ohio was subject to abundant and frequent rainfall similar to
that in the Pittsburgh area, the performance of a  green infrastructure system within a road right-of-way
out-performed its design capacity.  This was likely due to the conservative assumptions on the storage
and infiltration capabilities of the existing soil.
                                              15

-------
Michigan Avenue Bioretention Facilities, Lansing, Ml
The Michigan Avenue Bioretention project in Lansing, Michigan was completed in 2007 and is an
example of an off-line planter box style bioretention system in an ultra-urban environment. The system
was designed for abundant and frequent rainfall typical of Lansing.  Similar to Pittsburgh, most of its
rainfall events are 1 inch or less. The project encompasses four blocks along a very busy five-lane street
with parallel parking and extra wide sidewalks lined with stores, restaurants, and other businesses.

I.  Design Summary
The project includes approximately 20 individual planter box-style bioretention cells dispersed behind
the curb along four city blocks. The primary design goal was to include bioretention wherever possible
along the project corridor resulting in a different treatment capacity for each of the four blocks. On
average, the bioretention cells treat nearly 1 inch of rainfall that drains from the adjacent road and
sidewalk. One of the blocks is capable of treating nearly 4 inches of rainfall from its drainage area.
      Typical Cross-Section
                    CWB
  MICHIGAN AVENUE
                                                                                     BUIICHMG
                                                                    UTAt.'t, ,
                                                                   MOCKWUL
                    Figure 12. Typical Bioretention Cross-Section on Michigan Avenue

Stormwater enters the bioretention cells from the road through curb cuts. The curb cuts direct runoff to
a sediment forebay before spilling into the bioretention cells (Figure 12). Characteristic of an off-line
system, once the cell has reached capacity the water backs up onto the road and continues along the
curb line until reaching a catch basin or another bioretention cell.  Captured stormwater moves through
the soil matrix and leaves the cell either through a perforated underdrain or through infiltration. The
existing soils are slowly-permeable so an underdrain was provided to help dewater the system within 24
hours and prepare for the next storm event.

Specifically chosen trees and native plantings within the bioretention cells provide water uptake and
also aid infiltration with their root system while offering natural beauty (Figure 13). To determine
                                               16

-------
project effectiveness, monitoring was performed including infiltration testing, flow monitoring, plant
health assessment, and porosity and field capacity testing.
Photos: Top left: 600 Block Michigan Avenue Before. This corridor was 100% impervious, receiving
runoff from the roadway, sidewalk and buildings. Bottom left: during construction. Right: A 1,300-
square-foot rain garden was installed with trees, perennial plants, and grasses to capture and treat
stormwater runoff.
                     Figure 13. Photo Sequence of Bioretention on Michigan Avenue
2.  Results Summary
Pre-construction flow monitoring was conducted to gather hydrologic characteristics of the site prior to
installation of the bioretention cells. Post-construction flow monitoring efforts looked at the inflow and
outflow hydrographs from an individual bioretention cell.  The infiltration rate of the engineered soil,
field capacity, porosity, and an assessment of overall plant health were also monitored. The monitoring
information was then used to calibrate a system-wide model using US-EPA SWMM v 5.0.021. With the
calibrated model, discrete design storms were simulated to analyze the effect of the entire project on
the hydrology of the corridor.
                                               17

-------
Results showed that the bioretention system was able to remove a greater volume of stormwater than
expected, likely due to the storage capacity of the engineered soil and the robust vegetation. For a 1-
year, 1-hour design storm, overall volume is reduced by 46%, peak flow rate is reduced by 66%, and time
to peak is reduced by 33% (Figure 14). A long-term continuous simulation using approximately 50 years
of data indicated an average annual reduction of runoff volume of approximately 75%.

The function of the curb cut inlets and gutter bypass are working as an off-line system as anticipated.
The underdrains are working to dewater the system within 24 hours as designed.
   0.6
   0.5 -
   0.4 -•
 •e
 HI
 a 0.3 -•
 tt
   0.2 -
   0.1



\
\Pr




e-(no biore
v_




tention)











Post- (with bioretent

TotalPre
Huff Ang
	 Post


on)
ripitation =
=1 1st Quart
-..; 	 Pre-.



0.95-inches
1e Distribut



              30
                     60
                             90      120     150
                                  Time (minutes)
                                                    180
                                                            210     240
Figure 14. Runoff Comparison 1-year 1-hour
3.  Lessons Learned
Based on modeling results, the bioretention system is performing better than expected with peak flow,
volume, and time-to-peak reductions. The off-line design ensures excess stormwater bypasses the
system and the underdrain ensures an acceptable dewater time.  Lansing's abundant and frequent
rainfall is not a barrier; rather it was addressed as part of the design process.
                                              18

-------
References


Manuals, Articles, and Books
3 Rivers Wet Weather. 2013. Report on the 3 Rivers Wet Weather Project for Municipal Area-Wide
       Assessment of the Feasibility of Green Infrastructure.

Burgess & Niple, Limited. June 1989. City of Cincinnati Stormwater Management Rules and Regulations,
       Part 1, Technical Reference Manual.

Darner, R.A., and Dumouchelle,  D. H. 2011. Hydraulic characteristics of low-impact development
       practices in northeastern Ohio, 2008-2010: U.S. Geological Survey Scientific Investigations
       Report 2011-5165, 19 p.

Draft District of Columbia Stormwater Management Guidebook.

FISRWG. October 1998. Stream  Corridor Restoration: Principles, Processes, and Practices. By the Federal
       Interagency Stream Restoration Working Group. Chapter 2: Stream Corridor Processes,
       Characteristics, and Functions.

Lukes, R., Kloss, C. 2008. Managing Wet Weather with Green Infrastructure, Municipal Handbook, Green
       Streets. Low Impact Development Center. Environmental Protection Agency.

Montalto, F., Smalls-Mantey, L., DiGiovanni, K., Gunther, B., Compton, N., and Shetty N. 2013. The
       Performance of Green Infrastructure Under Extreme Climate Conditions. Drexel/CCRUN, NYC
       Department of Parks &  Recreation.

Pennsylvania Department of Environmental Protection (PADEP). 2006. Pennsylvania Stormwater Best
       Management Practices  Manual.

Prince George's County, Maryland. Bioretention Design  Specifications and Criteria.

U.S. Department of Agriculture.  March 2007. Natural Resources Conservation Service. Part 630
       Hydrology National Engineering Handbook. Chapter 16 Hydrographs.

U.S. Department of Commerce.  National Oceanic and Atmospheric Administration. National Weather
       Service. 2006. NOAA Atlas 14 Precipitation-Frequency Atlas of the United States, Volume 2,
       Version 3.

U.S. Department of Commerce.  Weather Bureau. 1961.  Technical Paper No. 40 Rainfall Frequency Atlas
       of the United States.

U.S. Army Corps of Engineers Hydrologic Engineering Center. March 1979. Flood Hydrograph and Peak
       Flow Frequency Analysis.

Virginia OCR Stormwater Design Specification No. 7 Permeable Pavement. Version 1.8. March 2011

Westmoreland Conservation District. 2013. Primer on Stormwater Management: Bioretention in Clay
       Soils, 3rd Edition, www.wcdpa.com Last Accessed September 2013.
                                             19

-------
Websites
3 Rivers Wet Weather
http://www.3riverswetweather.org/about-wet-weather-issue. Last accessed September 2013.

Clean Rivers Campaign
http://cleanriverscampaign.org/green-solutions/. Last accessed March 2013.

New York State Department of Environmental Conservation
http://www.dec.ny.gov/lands/58930.html. Last accessed March 2013.

NOAA National Weather Service
http://www.erh.noaa.gov/ctp/hydro/index.php?tab=precip#picture. Last accessed March 2013.

Oregon  State University
http://prism.oregonstate.edu/. Last accessed March 2013.

Penn State Climatological Office
http://www.stateclimate.org/state.php?state_id=PA. Last accessed March 2013.

Philadelphia Water Department
http://www.phillywatersheds.org/what_were_doing/green_infrastructure/tools. Last accessed March
2013.

Sustainable City Network
http://www.sustainablecitynetwork.com/topic_channels/water/article_44044ddc-e499-lleO-81dl-
0019bb30f31a.html. Last accessed March 2013.

US EPA Website
http://water.epa.gov/infrastructure/greeninfrastructure/gi_design.cfm. Last Accessed March 2013.
http://water.epa.gov/infrastructure/greeninfrastructure/gi_modelingtools.cfm. Last accessed March
2013.

Yakima County Regional  Stormwater Management Program
http://www.yakimacounty.us/stormwater/lid/overview.htm. Last accessed March 2013.
                                              20

-------