v>EPA
United States
Environmental Protection
Agency
Office of Water
Washington, D.C.
EPA 832-F-99-041
September 1999
Combined Sewer Overflow
Management Fact Sheet
Sewer Separation
DESCRIPTION
Sewer systems that convey both sanitary sewage
and storm water through a single pipe are referred
to as combined sewer systems (CSSs). In dry
weather and during light to moderate rainfall, the
CSS is able to convey all flows to the wastewater
treatment facility. During periods of heavy rainfall,
however, the capacity of the CSS may be exceeded,
often causing untreated combined sewage and storm
water to back up into basements and to overflow
from manholes onto surface streets. Traditionally,
CSS outfalls were designed to discharge directly
into receiving waters during combined sewer
overflows (CSOs). This was done to prevent the
excessive combined flows from directly impacting
public health via basement and street flooding.
In addition to flooding problems, CSOs can cause
problems in receiving water bodies. CSOs can
contain untreated domestic, industrial, and
commercial wastes, as well as storm water runoff.
Contaminants contributed by these sources include
potentially high concentrations of suspended solids,
biochemical oxygen demand (BOD), oils and grease,
toxics, nutrients, floatables, pathogenic
microorganisms, and other pollutants. CSO
pollution has caused many receiving waters to
exceed water quality standards, resulting in threats
to public health, aquatic species, or aquatic habitat.
CSO pollutants have impaired receiving water body
uses and have contributed to restrictions on shellfish
harvesting, occasional fish kills, and numerous
beach closures. Potential odors and solids deposits
in the receiving water body can also compromise
aesthetics and limit recreational uses of the water
body.
Many communities have studied and evaluated CSO
control strategies that would effectively reduce, if
not necessarily eliminate, CSOs and their associated
health and ecological risks. One of the strategies
often considered is sewer separation.
Sewer separation is the practice of separating the
combined, single pipe system into separate sewers
for sanitary and storm water flows. In a separate
system, storm water is conveyed to a storm water
outfall for discharge directly into the receiving
water. Based on a comprehensive review of a
community's sewer system, separating part or all of
its combined systems into distinct storm and sanitary
sewer systems may be feasible. Communities that
elect for partial separation typically use other CSO
controls in the areas that are not separated.
APPLICABILITY
Sewer separation can be considered wherever there
is a CSS. However, an evaluation of the most
appropriate CSO control should be performed prior
to selecting sewer separation or any other measure.
Sewer separation has often been the appropriate
technology in areas where one or more of the
following conditions exist:
• Most sewers are already separated;
• Siting constraints and costs prohibit the use
of other structural measures;
The uses and the assimilative capacities of
receiving waters prohibit the use of other
CSO controls;
-------
• Other CSO strategies are not publicly
acceptable;
Additional infrastructure improvements,
such as road repaying, are also required;
• The combined system is undersized;
• Elimination of CSOs is desired; and/or
• Other CSO measures are not able to achieve
the community's goals.
Sewer separation has been used effectively in many
communities. Most of the approximately 1,000
communities that are served by CSSs are located in
the Northeast and the Great Lakes region.
Complete or partial separation of CSSs has
occurred in many of these areas, as well as in several
locations in the West. Cities that have completely
or partially separated CSSs include: Minneapolis,
St. Paul, and South St. Paul, MN; the metro
Detroit, MI, area; the metro Boston, MA, area;
Salem and Portland, OR; the metro Seattle, WA,
area; Lynchburg, VA; Bangor, ME; Hartford and
Norwich, CT; and Remington, IN. Columbus, OH,
has recently elected to separate its CSS as well.
One of the largest sewer separation projects
occurred in Minneapolis, St. Paul, and South St.
Paul, MN. The project involved pipe separation in
more than 21,000 acres of drainage area. By
December 1996, 189 miles of storm sewers and
11.9 miles of sanitary sewers had been installed.
This program was needed to reduce the number of
overflows that were estimated to occur an average
of once every three days (American City and
County., 1996). Overflows have been significantly
reduced by this separation project.
ADVANTAGES AND DISADVANTAGES
Positive impacts resulting from sewer separation
include: reduction or elimination of basement and
street flooding; reduction or elimination of sanitary
discharges to receiving waters; decreased impacts to
aquatic species and habitat; decreased contact risk
with pathogens and bacteria from domestic sewage
in the receiving water; and relief from CSO
regulations. In addition, incidental infrastructure
work (e.g., road repaving and the repair or
replacement of miscellaneous utilities, such as water
and cable lines) could be conducted more cost
effectively if it were to coincide with sewer
separation. For example, as a result of the CSO
program in the City of St. Paul, MN, streets were
paved and handicap ramps were added to sidewalks,
gas and water mains were installed, gas services
were renewed or replaced, lead water service
connections were replaced, and street lights were
installed.
Separating CSSs may contribute to improvements to
water quality due to the reduction or elimination of
sanitary discharges to receiving water bodies.
However, the increased storm water discharges
resulting from sewer separation could decrease the
positive impacts of the separation unless storm
water discharges are mitigated. Without mitigation,
increased loads of storm water pollutants, including
heavy metals, sediments, and nutrients, may run off
into local water bodies. For example, in Atlanta,
GA, sewer separation was predicted to increase
pollution to local creeks (AMSA, 1994) as polluted
storm water previously reaching the treatment plants
now is discharged directly into receiving waters.
However, in many cases, separating sewers reduces
pollution to receiving waters, as described above for
St. Paul, MN. A second example of successfully
reducing pollution to receiving water bodies has
occurred in Juneau, AK. It has been reported that
in Juneau, where there is in excess of 70 inches of
precipitation a year, the storm water concentrations
conveyed through the recently separated storm
water sewers are rather dilute. This has also been
attributed to large quantities of clean groundwater
that infiltrate into the storm sewer, relatively clean
activities within the watershed, and several
non-point source pollution control programs within
the City (City of Juneau, 1997). Existing and future
storm water impacts to the receiving water body
should be evaluated prior to implementing sewer
separation.
Negative impacts associated with sewer separation
include extensive construction and construction
related impacts (e.g., noise, dust, erosion),
disruption to residents and businesses, possible
disruptions in sewer service, and the need for storm
water controls or best management practices.
-------
In addition, complete separation of sanitary and
storm water flows can be hard to accomplish
whether the combined sewer is converted to a storm
sewer or to a sanitary sewer. Complete separation
of a CSS would involve disconnection of all storm
water drainage structures, sump pumps, and roof
and footer drains. Disconnection of footer drains is
often not cost effective. Some communities have
offered financial incentives to homeowners and
businesses for voluntarily disconnecting some of
these storm water sources from sanitary sewers.
Many communities have also passed ordinances
requiring the disconnection. Despite these
provisions, there is still potential for some storm
water flows to remain connected to sanitary sewers.
Likewise, complete disconnection of sanitary flows
from a converted storm water sewer may be difficult
to accomplish, but is usually more successful than
eliminating all storm flow connections from
connected sanitary sewers.
KEY PROGRAM COMPONENTS
Decisions for a CSO control strategy should be
made on a site-by-site basis utilizing drainage area
data, receiving water use and water quality data, and
sewer system data. Sewer system information can
be obtained from review of sewer plans, television
inspection, and flooding records. Communities may
consider performing house to house inventories of
house connections to the combined system (i.e.,
sanitary and roof drains). This was successfully done
in parts of the metropolitan Boston area. Modeling
and Geographical Information Systems (GIS) may
be useful data analysis and prediction tools.
Using these data, communities should determine
what CSO controls, or combination of controls, will
meet performance goals established by the
community. Other factors, such as cost
effectiveness, natural and urban topography and soil
types, siting constraints, location of current and
future utilities, land use and cover, existing sewer
capacity, layout, and condition, pump and treatment
plant capacities, and requirement for other
infrastructure work in the same area, should be
considered before finalizing project plans. For
example, sewer separation was selected in
Minneapolis, South St. Paul, and St. Paul, MN, due
to local needs for eliminating sewage backups into
basements, reducing street flooding, and
reconstructing aging portions of the sewer system
(MWCC, 1984).
Sewer separation can be accomplished through
installing new storm or sanitary sewers to be used in
conjunction with the existing sewer. Economics,
capacity, condition, and layout of the combined
sewer are the typical factors used in deciding the
existing line's post-separation use.
An advantage of converting the combined sewer to
a sanitary sewer (referred to as a converted sanitary
sewer in this document) is that all sanitary flows are
already connected to the converted sanitary sewer.
Using the existing combined sewer as the sanitary
sewer and installing a new storm sewer would likely
require that any overflow weirs, gates, or other
regulating devices remaining in the converted
sanitary system be bulkheaded or otherwise disabled
to eliminate the potential for sewage to overflow.
In addition, storm water drainage structures, sump
pumps, and roof drains must be disconnected from
the converted sanitary system and connected to the
new storm water sewer. This will provide more
capacity in the converted sanitary sewer and will
reduce the possibility of overflows. Building footer
drains, however, are often left connected to the
existing combined system and do consume some of
the converted sanitary sewer capacity.
Rehabilitation or relining of the converted sanitary
system, storage tanks, and/or equalization basins
may be required if infiltration is a significant
problem due to cracks or inadequate construction
materials (e.g., brick sewers). In some cases, such
as in Juneau, AK, the existing combined sewer may
be in such poor condition that new sanitary sewers,
as well as new storm sewers, are constructed.
There are some circumstances that may make the
conversion of the combined sewer to a storm sewer
(referred to as a converted storm sewer in this
document) more appropriate. For instance,
combined sewers that have a large diameter and
have little slope (less than 3 percent) would not
have the flushing velocity required of a sanitary
sewer. In cases such as this, the existing CSS may
be more appropriately converted to a storm sewer,
provided that the sewer has sufficient capacity for
safe conveyance of the local design storm. A
-------
smaller sewer should be appropriately designed,
sized, and constructed to convey the sanitary flows.
Storm, roof, and footer drains, as well as
catchbasins could remain connected to the
converted storm sewer. Sanitary connections,
however, would need to be disconnected and
conveyed to the new sanitary line. Any remaining
sanitary lines connected to the converted storm
sewer will cause direct discharges of sanitary flows
to the receiving water body. Post-separation
sampling and monitoring of the converted storm
sewer is typically required to confirm that all
sanitary flows have been removed from the
converted storm sewer and redirected into the
sanitary sewer. Conversion of the combined sewer
to a storm sewer would also require that the
interceptor connection at the regulating device (e.g.,
weir or gate) be plugged, and may potentially
require modifications to prevent water from
stagnating upstream of the regulator.
Consideration should be given to coordinating
sewer separation with improvements to other
utilities, as this enhances the cost-effectiveness of
both/all projects and minimizes disruption to the
public.
IMPLEMENTATION
Sewer separation reduces and often eliminates
untreated sanitary discharges from discharging into
receiving water bodies, and therefore positively
impacts receiving water quality. Sewer separation,
however, greatly increases untreated storm water
discharges to the receiving water body. In a CSS,
at least some of the storm water flows are treated at
the treatment plant. The performance achieved with
sewer separation will vary depending on the existing
storm water pollutant loading and the existing
sanitary pollutant loading. For example, a study
performed for North Dorchester Bay, MA,
estimated that the overall fecal coliform removal
potentially achieved by sewer separation was only
45 percent (Metcalf & Eddy, 1994). This was
attributed to the increase in storm water discharges
to the receiving water body, and the corresponding
increase in non-point runoff pollutants.
Actual fecal coliform removal rates have been
determined for several sites where sewer separation
has been implemented. Water quality monitoring
data collected in St. Paul and Minneapolis from
1976 to 1996 indicated a fecal coliform
concentration reduction of 70 percent. One of the
four sites where data was collected reduced fecal
coliform concentrations from an average of 500
organisms per 100 mL to 150 organisms per 100
mL. At another site, fecal coliforms were reduced
from 489 organisms per 100 mL to 143 organisms
per 100 mL (Richman, 1996). This reduction has
been attributed to sewer separation and to the
reduction in the number of overflows occurring
every year.
Sewer separation may also result in other related
improvements to water quality. In stretches of the
Mississippi, water quality improvements attributed
to sewer separation projects have resulted in the
return of the pollution-sensitive Hexagenia mayfly
after a 30 year absence; the return of Bald Eagles to
the area; and the recovery of fish populations and
diversity from 3 species to over 25 species (Cities of
Minneapolis, et. al., 1996).
Monitoring the performance of CSO control
strategies at the Rouge River Demonstration
Program has been underway since the summer of
1997. Part of the monitoring program will identify
the effectiveness of sewer separation in terms of
improvements to water quality. Instream
monitoring is also occurring in Portland, OR. The
supplemental data will add to the performance data
collected in Minnesota (70 percent fecal coliform
reduction) and estimated for Massachusetts (45
percent fecal coliform reduction).
OPERATIONS AND MAINTENANCE
The Operations and Maintenance (O&M
requirements of separated sewers are generally the
same as those of other sewer systems. Maintenance
must be conducted on pump stations (including
routinely cleaning wet wells, testing for adequate
pumping capacity, and exercising pumps and
stand-by generators), sewer lines, and catchbasins
and grit chambers. Catchbasins and grit chambers
located in the sanitary or storm sewer system will
require routine cleaning to prevent accumulation of
sediment. Jet spray cleaning, pumping, and
-------
vacuuming are common methods for cleaning
catchbasins and grit chambers.
In addition, all sewer lines, and in particular sewers
that were previously combined, need to be
monitored to verify hydraulic capacity and to
identify infiltration and inflow. Basement or street
flooding is a likely indication of hydraulic capacity
or gradient problems in the sewer and may require
maj or repairs. Excessive infiltration into a converted
sanitary sewer may require rehabilitation of the
sewer system. Methods for assessing the condition
of the sewers include modeling, smoke testing, and
television inspection. Monitoring will identify
cracked and collapsed sewers that will need to be
repaired. In addition, monitoring can identify the
location and cause of sewer blockages. To prevent
blockages, lodged debris, sediment, and grit must be
removed on a regular basis.
Post-separation monitoring and sampling may be
required to ensure that no sanitary flows are still
connected to the storm sewer and being directly
discharged to the receiving water body.
Alternatively, simple dye studies can be employed to
verify separation.
COSTS
Separation costs vary considerably due to the
location and layout of existing sewers; the location
of other utilities that will have to be avoided during
construction; other infrastructure work that may be
required; land uses and costs; and the construction
method used (e.g., open cut verses microtunneling).
Communities that have other infrastructure
requirements (such as road repairs) in addition to
sewer separation may find that upgrading the
facilities simultaneously can result in a much lower
cost relative to upgrading them independently.
Construction occurring in existing right-of-ways
would probably not require land acquisition, and
thus would not add to the project cost. Project
costs could increase depending on the land use. For
example, project construction occurring in an
industrial area that contained hazardous materials or
wastes would likely increase the project cost.
Methods of construction, such as the need to tunnel
or bore versus open cutting, can also add to the
cost. Project costs could also increase if sanitary
equalization basins are required as part of the
separation project or if storm water best
management practices are required to control the
increased storm water discharges to the receiving
water body.
Actual construction costs are available from the St.
Paul sewer separation project. For that project,
sewer separation costs ranged from $8,350/acre
to$40,060/acre, with an average cost of
$15,400/acre (all costs are in 1984 dollars).
Estimates from the City of Portland and Detroit are
$18,000/acre and $67,800/acre, respectively.
The Rouge River project has also generated good
cost data for sewer separation. Costs were
approximately $377,000 for separating
approximately 600 meters of pipe on a small
residential street (CSO Area 42, Windsor Avenue),
which included costs for removing existing
pavement, laying a new sewer line, and re-paving
and re-sodding. A second project (CSO Area uing
cost $1.3 million to separate approximately 2,600
meters of pipe. All costs are presented in 1995
dollars.
The cost of operation and maintenance (O&M) of
the separated sewer system is difficult to predict.
Factors contributing to the O&M costs include the
age and the condition of the previously combined
sewer, the length and diameter of the sewers, the
frequency and the amount of sand and grit removed,
and the size of drainage areas.
Sewer separation can reduce treatment and O&M
costs at the receiving treatment plant by potentially
eliminating storm water flows to the plant. Energy
costs for transporting flows to the treatment plant
could also be reduced due to the reduced flow
volume.
REFERENCES
1. American City and County, 1996. "CSO
Control Revitalizes Stretch of the
Mississippi," pp. 31-32.
-------
Association of Metropolitan Sewerage
Agencies (AMSA), 1994. Approaches to
Combined Sewer Overflow Program
Development. Washington, D.C.
City and Borough of Juneau, Alaska, 1997.
Personal Communication with Ernie Mueller
with the City and Borough of Juneau, Public
Works.
Metcalf & Eddy, 1994. Final CSO
Conceptual Plan and System Master Plan:
Part II CSO Strategies. Prepared for the
Massachusetts Water Resources Authority.
Wakefield, MA.
Metropolitan Waste Control Commission
(MWCC), 1984. MWCC Metro 201
Facilities Plan. St. Paul, MN.
Cities of Minneapolis, Saint Paul, South St.
Paul, and Metropolitan Council
Environmental Services, 1996. Separating
Combined Sewers to Improve and Protect
Mississippi River Water Quality: A
Ten-Year Commitment. Annual Progress
Report to the Public - Year Ten.
Minneapolis, St. Paul, South St. Paul, MN.
City of Portland, Oregon, 1997. Personal
communication with Lester Lee, City of
Portland.
Richman, M., 1996. "Sewer Separation
Lowers Fecal Coliform Levels in the
Mississippi River". Water Environment &
Technology, Vol. 8, No. 11, pp 20-22.
Wade-Trim, 1997. Personal communication
with Laura Crane, Wade-Trim.
Borough and City of Juneau, Alaska
Ernie Mueller
Department of Public Works
5433 Shaune Dr.
Juneau, AK 99801
Michigan City, Indiana, Sanitary District
Tim Haus
532 Franklin Street
Michigan City, IN 463 61
City of New Haven, Connecticut
Raymond Smedberg
Water Pollution Control Authority
345 East Shore Parkway
New Haven, CT 06510
City of Portland, Oregon
Lester Lee
City of Portland Bureau of Environmental Services
1211 Southwest 5th Avenue, Suite 800
Portland, OR 97204
Rouge River Demonstration Project
Vyto Kaunelis
Wayne County Department of Environment
415 Clifford Street, 7th Floor
Detroit, MI 48226
City of Saco, Maine
Larry Nadeau
Department of Public Works
300 Main Street
Saco, ME 04072
City of St. Paul, Minnesota
Mike Kassan
Sewer Utility, Department of Public Works
1000 City Hall Annex
St. Paul, MN 55102
ADDITIONAL INFORMATION
City of Columbus, Ohio
Laurie Mehl
Public Utilities, Division of Sewerage and Drainage
910 Dublin Road, Room 32
Columbus, OH 43215
-------
The mention of trade names or commercial products
does not constitute endorsement or recommendation
for the use by the U.S. Environmental Protection
Agency.
For more information contact:
Municipal Technology Branch
U.S. EPA
Mail Code 4204
401 M St., S.W.
IMTB
Excelence fh compliance through optimal technical soLtrons^
MUNICIPAL TECHNOLOGY
------- |