MODERN
r SEWAGE
TREATMENT
PLANTS
How much do
. they cost ?
U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE
PUBLIC HEALTH SERVICE
-------
Modern Sewage Treatment Plants-
How Much Do They Cost?
A Practical Guide to Estimating Municipal Sewage
Treatment Plant Construction Costs
(Based on Studies of Projects Built With PHS Grants—1956-63)
U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE
Public Health Service
Division oi Water Supply and Pollution Control
Washington, D.C. 20201
-------
Public Health Service Publication No. 1229
U.S. GOVERNMENT PRINTING OFFICE, WASHINGTON, D.C. : 1964
-------
Contents
Page
Foreword - v
Abbreviations vi
Definition of Terms (Glossary) vi
An lntroductoryStatement 1
Earlier Construction Cost Studies 2
Current Cost Study 4
Proceduresof Analysis 4
Treatment Plant Costs per Capita 11
Treatment Plant Costs per Population Equivalent 21
Treatment Plant Costs per Unit Flow 27
Limitations of Cost Estimating Data 36
Recapitulation of the Studies 37
References 37
Figures
1. Map of 20 Index Cities and Their Assigned Areas of Cost In-
fluence 8
2. Typical Unit Cost Estimating Curve 9
Construction Cost per Capita—
3. Activated Sludge Plants 10
4. Imhoff Tank Plants 12
5. Imhoff-Type Plants 13
6. Primary Treatment—Separate Sludge Digestion Plants 14
7. Stabilization Ponds 15
8. Trickling Filter—Separate Sludge Digestion Plants 16
9. Trickling Filter—Imhoff-Type Plants 17
10. Cost Comparison by Types of Treatment 18
11. Primary Types of Treatment 19
12. Secondary Types of Treatment 20
Construction Cost per Population Equivalent—
13. Activated Sludge Plants 22
14. Trickling Filter—Separate Sludge Digestion Plants 23
15. Trickling Filter—Imhoff-Type Plants 24
16. Cost Comparison by Types of Treatment 25
Construction Cost per Unit Flow—
17. Imhoff-Type Plants 26
18. Primary Treatment—Separate Sludge Digestion Plants 28
19. Stabilization Ponds 29
20. Activated Sludge Plants 30
21. Trickling Filter—Separate Sludge Digestion Plants 31
22. Trickling Filter—Imhoff-Type Plants 32
23. Cost Comparison by Types of Treatment 33
24. Primary Typesof Treatment 34
25. Secondary Types of Treatment 35
in
-------
Foreword
To stimulate the construction of sewage treatment works in the United
States and to clean up the municipal wastes which are being discharged into
the Nation’s watercourses, lakes, and coastal waters, the Federal Government
offers grants-in-aid incentives to municipalities. The program was initiated
under the Federal Water Pollution Control Act of 1956, and has continued with
increased financial assistance authorized by the 1961 amendments to the Act.
Basically, this legislation was planned to encourage communities to invest
their own funds to augment Federal contributions. The program has worked.
For every dollar of Federal funds, municipalities have invested from $4 to $5
of their own money in the pollution control effort.
Over and above its value as a pollution control catalyst the Federal grants
program has made it possible for the Public Health Service to document
important details in sewage works design, construction, and financing. Infor-
mation on types of treatment used, bases of design, contract bids, and other
basic data has been processed by modern business techniques and evaluated
for guideline purposes.
One of the most useful guidelines derived from the PHS data reflects
the varying costs of sewage treatment plant construction as influenced by size
of plant, type of treatment, regional differences in wage scales and building
material prices, and other factors.
These data have produced valid answers to the important question: How
much does a modern sewage treatment plant cost? The answers provide a
dependable base for future financing practices and will enable municipalities
to plan their treatment facilities and fiscal operations with proper preliminary
accuracy.
The Public Health Service and others in the sanitary engineering profes-
sion, have made previous studies and published their findings as guides to
sewage treatment plant cost-estimating. The information here reported is be-
lieved to be more truly authentic and defines more sharply the treatment
process categories. If it sheds new light on the question of sewage treatment
plant costs and the factors influencing construction costs, it will have achieved
it purpose.
This report was prepared by Assistant Sanitary Engineer Donald R.
Kaiser under the supervision of Mr. Peter P. Rowan, Chief, Evaluation Sec-
tion, Construction Grants Branch, Division of Water Supply and Pollution
Control, Public Health Service. Editorial advice and assistance was provided
by Dr. Morris M. Cohn, Special Consultant to the Public Health Service.
DAVID H. HowEu s,
(]hief, Construction Grants Branch,
Division of Water F pply and Pollution Control
V
-------
Abbreviations
AS
ENR-
ENR—C
IT
I’rT
MGD
cc
FE
FHS
Activated Sludge
Engineering News-Record
Magazine
Engineering News-Record
Construction Cost Index
Inthoff Tank Treatment
Imhoff-Type Treatment
Million Gallons per Day
Cost per Capita
Population Equivalent
Public Health Service
PHS—STP
P-SD
SE
SF
TF-P
TF-ITT
Public Health Service Sewage
Treatment Plant Construc-
tion Cost Index
Primary Treatment With Sep-
arate Sludge Digestion
Correlation Coefficient
One Standard Error of Estimate
Stabilization Pond
Trickling Filter Plus Primary
Treatment
Trickling Filter Plus Imhoff-
Type Treatment
Definition of Terms
Sewage: Liquid carried wastes of a commu-
nity from domestic, commercial, and industrial
sources..
Interceptor sewer: Pipe or conduit used to
collect and transport sewage from points of final
collection to point of treatment.
Outfall sewer: Pipe or conduit used to trans-
port the effluent from the treatment facility to the
point of final discharge.
Sewage treatment plant: Manmade structures
which subject sewage to treatment by physical,
chemical, or biological processes for the purpose of
removing or altering its jectionable constituents
and rendering it less offensive or dangerous.
Sewage treatment construction cost: Con-
struction contract cost plus an allowance of 20
percent for administrative, engineering, legal, and
fiscal costs (cost of land is excluded).
Design population: Number of people a facil-
ity is designed to serve.
Design flow: Hydraulic load for which facility
is designed.
Population equivalent: Sewage including in-
dustrial wastes, converted on a strength equiva-
lent basis, to a do nestic population value. (For
BOD, the strength equivalent is taken as 0.20
pound of BOD per day= 1 PE.)
Area of influence: One of the 20 commercial
marketing areas into which the country has been
divided for statistical purposes.
Least squares method: Statistical procedure
of fitting a line or curve to a distribution, of
points.
Simple correlation: Statistical comparison be-
tween two variables to determine their inter-
dependence.
Multiple correlation: Statistical comparison
between more than two variables to determine
their interdependence.
Correlation coefficient: A measure of the in-
terrelationship of one variable with respect to
another.
Standard error: The minimal sum of squares
of the differences between the actual values and
the estimated values for a. distribution of points.
Expected cost curve: An estimating curve es-
tablished on the basis of actual construction cost
data by the method of least squares.
Upper limit cost curve: Curve calculated at
a factor of one standard error of estimate above
the expected cost curve.
Lower limit cost curve: Curve calculated at
a factor of one standard error of estimate below
the expected cost curve.
V I
-------
Cost interval: Range of costs falling between
the upper and lower limit cost curves.
Ratio of the upper limit: Ratio of the upper
limit cost to the expected cost at the mean of the
distribution.
Ratio of the lower limit: Ratio of the lower
limit cost to the expected cost at the mean of the
distribution.
Valid size range: Interval of reliable data for
the dependent variable which marks the limits of
the expected cost curve.
Primary treatment: The removal of settleable
organic and inorganic solids by the process of
sedimentation.
Secondary treatment: Treatment of sewage
by biological methods, following primary treat-
ment.
lmhoff tank treatment: A form of primary
treatment employing the classic linhoff two-story
tank, consisting of an upper sedimentation cham-
ber and a lower digestion chamber, with no
mechanical equipment.
lmhoff -type treatment: A form of primary
treatment employing a two-story tank consisting
of an upper sedimentation chamber and a. lower
digestion chamber, with some type of mechanical
equipment.
Primary treatment—separate sludge diges-
tion: A form of primary treatment which em-
ploys a separate structure for digestion of sludge.
Stabilization pond: A pond designed for the
treatment of sewage by natural aerobic processes,
with or without the addition of supplemental
aeration or chemicals.
Activated sludge treatment: A secondary
treatment process which brings settled sewage into
contact with biologically active sludge in the pres-
ence of excess oxygen.
Trickling filter—separate sludge digestion:
A secondary treatment process, following primary
treatment, using a bed of coarse material over
which the settled sewage is distributed, followed
by final clarification.
Trickling filter—lmhoff -type treatment:
Identical to—trickling filter—separate sludge
digestion, but employing ImhofF-type treatment
for the primary phase.
v i i
-------
Introduction
The desire to own something must be supported
by the ability to own something. Whether it be
the purchase of a home or an automobile; or the
construction of a house, garage, or a sewage treat-
ment plant, the cost of the c.ommodit.y is of great
importance. It is more than that; it is basic
to the ability to own and to care for anything.
It is so basic that knowledge of how much it
will cost. should precede any actions to purchase or
construct the desired facility.
It is the responsibility of municipal officials
faced with the need for sewage treatment facili-
ties; of designers charged with the task of plan-
fling sewage works; of regulatory agencies vested
with the authority to approve, methods and physi-
cal processes to achieve desired degrees of treat-
ment.; of developers and planners required to pro-
vide sewage handling systems; and of investors
faced with bond purchase decisions, to have au-
thentic information on how much the required
installation should cost.
Actual project. costs do not become available
until plans and specifications have been completed
and approved, bids for construction work, mate-
rials and equipment have been received, and con-
tracts have been let. Yet, there is need for pre-
liminary concepts of what the eventual cost will
be long before these finalization steps have been
taken. In short, there is need for valid “measur-
ing sticks” or guidelines which will supply pre-
liminary cost estimates for projects.
There is no substitute for aefual cost informa-
tion, but cost estimates play an important role
in the preliminary stages of sewage treatment
works planning, despite the fact. that decisions
often must be based on water pollution control
needs rather than availability of funds. While
the size of a project may be firmly established
by the population to be served, the. population
equivalent as measured by organic loading on
the proposed treatment plant., or the volume of
flow to be handled, and while the degree of treat-
ment may be dictated by the specific pollution con-
trol needs and the dilution conditions in the ef-
fluent-receiving water resource, knowledge of what
the project may cost will be of great value:
• It may determine, to some extent., the
degree of treatment to be provided.
• It may dictate whether a project should
be phased out in stages rather than a full-scale
works on a one-time basis.
• It may ascertain the future period for
which capacity will be provided, or for which
actual construction will be scheduled on a
long-range, plan.
• It may help determine the use or non-
use of certain equipment and instrumentation
facilities in the design, or their installation at
some. later date.
• It can help municipal officials develop
master plans, not only for sewage works f a-
cilities but for other civic projec.ts, and for
their rational financing on long-range bases.
• It can serve as a guide in judging the
validity of competitive bids when contracts
are to be let.
• It can help guide bond issue referenda
and assure. investors in such bonds of the sta-
bility of the offerings.
These examples of the serviceability of con-
struction cost estimates point up the great respon-
sibilit.y which devolves on developers of such
estimating tools. They demonstrate the need for
using cost statistics of known validity in offering
cost-estimating guidelines, and for lucid interpre-
tation of such data in terms of their limitations
as well as their proven values. They serve as
warnings that estimates are no more than esti-
mates; that the. estimate must. be used by persons
versed in their applications, or lack of applica-
tions, to specific projects; that estimates are no
substitute for actual cost experiences by designers
and officials; and that estimates, at best, cannot,
and do not, reflect total project costs.
789—993 O—64--——2
I
-------
In providing rule-of-thumb data on construc-
tion costs, there is always the hazard that the un-
initiated may judge the merits of comparative
treatment processes on the basis of comparative
costs, alone. This report would be remiss if it did
not point out that relative costs, alone, cannot dic-
tate the choice of treatment process in a great
majority of sewage works projects. Such deci-
sions must be based on the degree of treatment
needed and the method which can best supply
it, on the availability of sites suitable for the type
of treatment chosen, and on the applicability of the
general environment and area development, both
present and long-range, to certain treatment proc-
esses.
Throughout this report one factor is stressed
repeatedly: That the cost estimate data do not
cover certain important items in the overall cost
of the project as built in, and on, the ground.
These noncovered items include administrative,
engineering, financing and other services, and, of
great importance, the land costs. Certain types
of treatment require more land area than others
and this must be given serious consideration be-
fore design decisions are made.
The annual investment in municipal waste treat-
ment construction has more than tripled in the
past 7 years. This, plus the many changes which
have evolved in treatment methods, such as modi-
fications in the activated sludge treatment process,
increased use of stabilization ponds and package-
type treatment plants, has increased the need for
shortcut preliminary cost estimating aids.
Though they be no more than cost “indicators,”
such tools can be of value to consulting engineers,
water pollution control agencies, municipal offi-
cials, and others.
Recognizing this need, the Public Health Serv-
ice in 1958, initiated a study in the construction
costs of sewage treatment facilities. From this
basic beginning the investigations have been up-
dated and made to reflect more specifically the
costs of construction in the municipal sewage
treatment field. The comprehensive estimating
tables and curves presented in this report are the
results of these studies.
Earlier Construction Cost Studies
Since the studies here documented are an exten-
sion and a sharpening of findings from previous
cost studies, it is appropriate to record briefly the
data presented in reports of these earlier investi-
Cost Study of 20 Plants: 1958
The first of these investigations was published
in 1958 (1), 2 years after the Federal incentive
grants program was authorized by PL—660. This
study concerned itself with engineering design
practices, costs of construction, and estimated op-
eration and maintenance costs for projects assisted
under the program. The analysis was based on 20
small secondary sewage treatment plants in the
upper Midwest. Four projects were selected at
random from each of five States: Illinois, Indi-
ana, Michigan, Ohio, and Wisconsin. Included
were six activated sludge plants, eight standard-
rate trickling filter plants, and six high-rate trick-
ling filter plants.
That portion of the investigation dealing with
construction costs drew its data from actual con-
tract prices. It did not include the cost of land,
nor engineering, administrative and legal services.
The cost data from these 20 projects were eval-
uated against the parameters of design population,
population equivalent, and design flow to estab-
lish unit cost curves of the second degree for each
parameter.
Cost Study of 380 Plants: 1958
Availability of data on the increasing num-
ber of treatment plants receiving Federal construc-
tion grants continued to offer further opportunity
for development of cost estimating techniques.
Another invest.igation was reported in 1958, based
on 380 projects in various parts of the United
States. This report presented construction cost
estimating curves for the general categories of
primary treatment, secondary treatment, and sta-
bilization ponds (s).
These data reflected probable construction costs
and did not include land, engineering, administra-
2
-------
tive, and legal costs. Interceptor and outfall sew-
ers, pumping stations and similar ancillary works
were excluded; projects involving plant additions
and enlargements were not considered. The En-
gineering News-Record Construction Cost Index
(ENR—C) was employed in the study to convert
all cost data, nationally, to a “common denomina-
tor” of cost. The estimating data were statisti-
cally developed by the application of least square
methods.
The Public Health Service then continued its
probe of sewage treatment plant costs on an even
more specialized basis. The third analysis of en-
gineering and financial criteria was based on data
from 31 new sewage stabilization ponds located
in seven Midwestern States (3). The projects
involved treatment of domestic sewage with little
or no industrial wastes.
Data for this analysis included the contract. cost
of excavation and earth placement, cost of piping
within the confines of the pond, inlet and outlet
structures, and the cost of fencing and seeding.
These costs were analyzed and related to surface
area on a cost per acre basis to equalize surface
loading variations. It was reported that the con-
struct.ion costs, as defined, accounted for about
80 percent of the total cost of a stabilization pond.
The remaining 20 percent included engineering,
legal, administrative, and contingency costs. It is
pointed out that the cost of land, often involving
much larger areas than the so-called standard
treatment plants, was not included in the cost
evaluation. Cost of pumping stations, intercep-
tors, and outfall sewers were also excluded.
Study of Specific Processes: 1960
The following year, the Public Health Service
reported a similar study which expanded and re-
fined the preceding investigations (4). This in-
vestigation added specifically to previous cost
studies. Instead of covering the generalized cat-
egories of primary, secondary and stabilization
pond treatment, it evaluated costs for six specific
types of treatment: Imlioff tank; conventional
primary treatment with separate sludge digestion;
activated sludge; trickling filters with separate
sludge digestion; trickling filters with Imhoff-
type treatment; and stabilization ponds. The
analyses produced data based on cost per capita
and design population, presented in the form of
estimating curves. The data from a wide geo-
graphical area were analyzed by the statistical
method of least squares. Costs were evolved by
the same criteria used in previous studies, includ-
ing the ENR-C Index.
In February 1961, the investigations into cost
estimating aids were extended to cover the opera-
tion and maintenance (0 & M) of sewage treatment
plants (5). The report presented data on actual
annual costs directly associated with plant opera-
tion and maintenance, excluding costs for central
administration, billing and collection of sewer
charges, and expenditures for capital mainte-
nance.
The data, collected from 320 sewage treatment
plants that had been in operation for a minimum
of 5 years, were analyzed by the statistical meth-
ods of least squares. These costs were evaluated
on the basis of design flow, population served,
and population equivalent, and resulted in the
development of unit cost estimating curves of an
inverse function form. Only unit cost data for
design flow and population served were presented
in the studies because insufficient data were avail-
able for cost estimations on the basis of popula-
tion equivalent. The resulting cost data were pre-
sented for four types of treatment: Primary;
standard-rate trickling filter; high-rate trickling
filter; and activated sludge.
These studies emphasized the need for a more
specific evaluation of sewage treatment plant con-
struction costs over a period of time than could be
obtained through the use of the ENR—C Index—
an index covering the broad complex of all con-
struction work. To meet this need, the Public
Health Service undertook the development of, and
reported on, a fixed base, weighted cost index,
which reflects the changing purchasing power of
the funds invested in sewage treatment plant
construction.
Cost Study of Stabilization Ponds: 1959 Operation and Maintenance Cost
Studies: 1961
3
-------
Cost Study of $1 Million High-Rate Filter
Plant: 1963
The index was established on the basis of a $1
million hypothetical high-rate trickling filter
plant using prices prevailing in the Kansas City,
Mo., area. The quantities of labor, material and
construction equipment, and overhead and profit
were established on the basis of the model plant
and remained constant. Unit prices varied from
month to month in accordance with published
prices in the Engineering News-Record and U.S.
Bureau of Labor Statistics Wholesale Prices. To
make this cost index more representative of the
whole country, indexes were developed for 20 trade
areas covering the United States. These trade
area indexes were averaged on a monthly basis to
obtain the national index.
Current Cost Study
The authenticity of cost estimation data is de-
pendent on the number of projects involved in es-
tablishing criteria, the inclusiveness of the types
of projects, and the local conditions under which
the projects have been constructed. The growing
number of sewage treatment projects of various
types receiving Federal construction grants has
made it possible for the Public Health Service to
improve its cost analysis procedures and to evolve
construction cost estimating data of greater value
to municipal officials, consulting engineers and
others.
Availability of these expanded data has made it
possible for the Service to go beyond a mere up-
dating of past cost estimating information. This
report., therefore, is based on data refinements of
an important nature. They represent:
1. The use of a more responsive index to
eliminate variations in treatment plant costs
due to geographic location and time differ-
ences
2. Presentation of unit costs on the basis of
design population, design flow, and design
population equivalent
3. Evaluation o construction costs for a
greater number of specific types of treatment.
These refinements have vastly extended the
value of data resulting from earlier investigations.
The report., therefore, tends to be more responsive
to current developments in the sewage treatment
field than previous studies.
Procedures of Analysis
Source of Data
The study is based on a tabulation of design
and cost information for 1,504 sewage treatment
projects constructed under the PL—660 program.
This source provided the most comprehensive data
yet available for the sewage treatment field, in
terms of numbers of projects and geophysically
widespread construction conditions. All sections
of the Nation, except Alaska, the District of Co-
lumbia, Guam, and the Virgin Islands are rep-
resented. A detailed listing by States for each
type of treatment involved in the study appears
in table I.
Criteria for Analysis
The cost data employed in the study represent
actual cost information of projects constructed un-
der the incentive grants program. These costs,
therefore, can be compared with earlier studies,
(1), (s), (3), (4), and (5), since they also repre-
sent only contract costs and not total cost of the
various types of treatment facilities. Only those
projects for which costs could be identified were
included in the study. Costs were not included
for such facilities as interceptor and outfall sew-
ers; pumping stations not. contiguous to the plant;
and administrative., engineering, and legal services.
A detailed review of the costs making up the
study, and of the costs excluded from the analysis,
showed that construction costs of a project repre-
sent approximately 80 percent of the total. The
additional costs which were excluded for this in-
vestigation can be incorporated in probable total
project costs by increasing the construction esti-
mates by a factor of 20 percent. The study does
4
-------
TABLE I
Distribution of Projects by State and Type of Treatment
State
Type of treatment
Imhoff tank
ImliolI-type
Primary treat-
ment —separate
sludge digestion
Stabilization
ponds
Trickling dltcr—
separate sludge
digestion
Trickling filter—
Imboff-type
Activated
sludge
PCC
MGD
PE
FCC
MGD
FE
P00
MCD
PE
FCC
MGD
PE
P00
MUD
PE
P00
MUD
PE
FCC
MUD FE
Alabama
Alaska
Arizona
2
1
5
3
5
2
5
2
_
18
1
9
9
2
1
1
8
5 5
Arkansas
Cailforma
Colorado
Connecticut
Delaware
1
1
1
1
16
5
1
7
3
1
21
3
11
19
3
7
- - -
5
10
9
5
2
4
3
2
4
4
1
1
8
2
1 1
3 5
2 2
District of Columbia
Florida
7
5
15
7
7
5
2 3
Georgia
Guam
1
4
1
1
1
13
3
3
2
2 2
Hawaii
Idaho
1
10
1
4
--__
11
9
1
1
1
1
1 1
Il linois
I
2
9
‘
18
16
11
6
6
9
4
5
5
4 4
Indiana
3
2
5
2
6
5
6
2
2
5
2
2
6
4 5
Iowa
5
1
7
5
12
5
5
19
6
6
Kansas
1
6
4
33
24
99
4
4
8
2
2
Kentucky
Louisiana
1
1
1
1
1
5
4
3
2
7
7
12
9
7
5
7
5
2
8
1
2
1
2
5
5
2 2
1 1
Maine
1
3
3
1
1
2
2 2
Maryland
Massachusetts
1
1
1
1
2
1
2
1
1
1
1
1
2
2
2
2
5
1 1
3 3
Michigan
Minnesota
13
4
6
1
1
11
1
10
6
22
1
8
2
9
2
3
2
2
2
3
4
1
I I
Mississippi
1
37
23
2
Missouri
1
1
71
50
6
2
2
3
2
2
1
Montana
1
1
2
2
32
19
Nebraska
2
1
7
5
42
25
7
1
9
6 7
Nevada
1
New Hampshire
2
3
1
New Jersey
NewMexico
1
3
2
1
----
2
8
1
3
1
3
2
1
NewYork
6
2
1
11
5
2
2
1
3
2 2
North carolina
1
3
3
3
1
7
6
24
10
8
9
6
6
5
3 3
North Dakota
74
18
Ohio
3
4
1
10
2
1
8
4
4
8
1
1
18
11 11
Oklahoma
1
29
5
3
3
3
1
Oregon
Pennsylvania
FuertoRico
1
I
I
6
12
1
6
12
7
9
14
1
1
4
1
1
4
1
2
2
1
1
1
2
5
1 1
3 3
Rhode Island
1
I
SouthOarolina
3
1
6
6
10
3
3
1
2
1
South Dakota
48
24
1
Tennessee
2
9
3
2
2
17
10
10
5
4
4
2
1 1
Tens
Utah
1
1
21
11
8
7
7
7
14
8
7
6
1
4 4
1 1
Vermont
Virginia
2
8
1
7
1
1
6
12
6
4
6
1
1
4
1
1
3
1 1
Virgin Islands
Washington
2
2
2
9
2
17
8
3
2
2
1
WestVirginla
Wisconsin
Wyoming
Total
1
5
43
8
2
45
6
1
24
i i
10
2
227
4
3
99
.
- - - -
1
13
23
564)
1
12
12
350
2
26
1
370
1
7
1
141
1
10
1
143
1
2
121
1
2
51
1
2
51
16
1
135
9 12
1 1
77
86
not include land costs because of their wide and
unpredictable Variations.
If actual construction costs of completed sewage
treatment plants are to be of value in estimating
the probable cost of proposed projects of similar
types, the “known” data must be translated into
costs per some established unit or units Which will
apply as well to the “unknown” installation.
Three such units, or parameters, of cost are avail-
able as “common denominators”:
1. The flow for which the plant is designed.
2. The population which the plant is de-
signed to serve.
3. The population equivalent to which the
plant is designed to serve, including the or-
ganic loadings from industrial-commercial
operations tributary to the community sewer
system.
Review of the data for the 1,504 plants selected
for the study indicated that any of these 3 pa-
rameters could be used since sufficient information
was available in all categories. Upon the basis of
this finding, the decision was made to employ all
three parameters in developing unit construction
costs. This was done to make available to govern-
5
-------
mental and private organizations using estimating
aids, a choice as to the most reliable cost relation-
ships for any particular project.
Treatment Processes Covered in This
Cost Study
To facilitate the cost studies it was necessary to
select the treatment categories into which the
available cost data could be grouped. The follow-
ing seven subdivisions were chosen for analysis on
the basis of available data:
1. hnhoff tank plants
2. Imhoff-type plants
3. Primary treatment—separate sludge di-
gestion plants
4. Stabilization ponds
5. Activated sludge plants
6. Trickling filters—separate sludge diges-
tion plants
7. Trickling filters—Imhoff-type plants.
To obtain a more specific definition of the fac-
tors in.flucncing costs of construction, it would
have been desirable to subdivide each treatment
category further as to type—such as modifications
of the activated sludge process; trickling filter
design; chlorination facilities; and other features.
However, the available data did not include a
sufficient number of projects to make this type of
differentiation valid. Therefore, all plants were
classified on the basis of the principal treatment
unit or process employed. For clarification pur-
poses, a brief description of the projects in each of
these subdivisions follows. (For additional de-
scriptive information, consult the “Definition of
terms,” p. vi.)
lmhoff tank plants (IT).—Projects included
in this treatment category are those primary
plants having self-contained digestion units, but
which do not have mechanically equipped settling
compartments. Plants in this group are not fol-
lowed by any additional form of treatment; they
provide primary treatment only.
Imhoff-tgpe plants (ITT).—All primary
plants which are of the self-contained digestion
classification and do not meet the description of
the preceding category are grouped under this
heading. These plants are equipped with mechan-
ical skimmers and/or sludge removal equipment.
As in the preceding category these systems are the
only means of treatment provided.
Primary treatment—separate sludge diges-
tion plants (P-SD).—This group represents the
remainder of the primary-type plants. All sys-
tems in this category employ gravity settling as
did the preceding two groups. However, these
plants employ separate structures for sludge diges-
tion and/or storage. This classification, therefore,
excludes Imhoff tank plants or Imhoff -type plants.
Stabilization ponds (SP).—The projects in-
cluded in this category are all stabilization ponds
(lagoons) which are designed as the principal
form of aerobic treatment for raw sewage. No dis-
tinction is made as to number of cells provided, or
flow patterns involved. Both single and multicell
ponds, as well as series and/or parallel-operated
systems, are included.
Activated sludge plants (AS).—This category
is composed of projects which employ primary
settling, aeration by either diffused air or mechani-
cal means, and final settling.
This group includes both prefabricated package-
type plants and site-constructed facilities. Had
sufficient information been available to make a dis-
tinction between package-type and site-constructed
plants, a more realistic picture of the activated
sludge category would have been possible.
Trickling Iilter—separate sludge digestion
plants (TF-P).—In this category are those proj-
ects which employ the trickling filter method of
treatment, preceded by primary treatment utilizing
separate sludge digestion., and followed by final
clarification. Systems which employ multiple-
stage units were not separated from the single-
stage type. In addition, no distinction was made
in this study between standard-rate and high-rate
filter plants. Trickling filter plants which include
other secondary treatment processes, such as acti-
vated sludge or stabilization ponds, are not iii-
cluded in this category.
Trickling filters—Imhoff-type plants (TF—
JTT).—The projects in this category include
trickling filters which are placed between a pri-
mary treatment unit and some form of final clari-
fier. The difference between this and the preced-
ing category (TF—P) is that the primary treatment
stage in this category is of the Imhoff type.
Again, no distinction has been made between
standard-rate and high-rate systems and no differ-
entiation was made between the flow patterns of
single and multiple-stage plants.
6
-------
Selection of an Index
Sewage treatment plant construction costs vary
from month to month, as well as from area to area.
To adjust for this variable condition and to estab-
lish cost limits which can be applied with reason-
able accuracy, it was necessary to use a cost index
to reflect the relationship of the price of a com-
modity to the price of a similar commodity for a
specific base period, and/or specific location (6).
In previous unit cost analyses ( ) and (4), the
Engineering News-Record Construction Cost In-
dex (ENR—C) was employed for converting cost
variations to a standard base. This index, how-
ever, as its title indicates, was designed for use in
general construction and quite naturally, did not
provide for the specific needs of the sewage plant
construction field. Obviously the use of a cost
index specifically developed for sewage treatment
construction would yield more representative and
workable estimating data. Therefore, the Public
Health Service Sewage Treatment Plant Construc-
tion Cost Index (PHS—STP) was developed and
has been used in this report.
The inclusion of a cost index has served to
“stabilize” the cost variations which are inherent
in any cost comparison. The use of an index which
is related to a fixed base, as is the PHS—STP in-
dex, adjusts the costs from any particular time
period to the base period of the index (1957—
59 = 100). Thus, all costs used in the study are in
terms of 1957—59 dollars.
To account for area cost variations, the 20 trade
areas of the PHS-STP index were used. These
areas are shown in figure 1, with each area’s in-
fluence city designated. It was assumed that the
costs of all projects located in each area would vary
directly as the index for the influence city, and they
were so adjusted. The projects from Hawaii and
Puerto Rico we.re adjusted on the basis of the Na-
tional Index value.
Mechanics of Analysis
With the parameters of cost comparison defined,
it was necessary to select the appropriate statis-
tical methods by which to analyze the data and
from which resulting conclusions could be made.
The use of more than two parameters opened up
new possibilities of analysis. Not only was it
possible to analyze the data in three independent
correlations, but it appeared reasonable to investi-
gate the development of a multiple comparison of
cost with more than one cost parameter.
The first step in the procedure of comparing
unit costs to each of the other parameters, was the
visual representation of the sample. A Cartesian
presentation showed a skewed distribution of the
sample to the right. In an attempt to reduce the
sample spread, a log-log plot was made. With
the distribution spread considerably reduced
through the use of logarithms of both variables,
the method of least squares was employed to fit an
appropriate curve to the data. By testing the
standard forms of curves by this statistical tool, it
was found that a straight line within a log-log
environment best fitted the distribution (7) (8).
The specific description of the curve which best
fits a particular sample distribution in this study
can be determined by a simultaneous solution of
the two normal equations:
where,
Na+b X==r Y (1)
a2X+b X 2 =Y.XY (2)
N=Number of observations.
X== Sum of iogs of the design population
(population equivalent, flow);
= Sum of squares of the log of the design pop-
ulation (population equivalent, flow);
Y==Sum of logs of 10 times the unit cost
(except in cost per unit flow correlation
when it is the sum of log of the unit
cost);
Y 2 =Sum of squares of the logs of 10 times the
unit cost (except in case of flow correla-
tion in which factor of 10 is not included);
XY=Sum of cross products of logs of X and Y;
a= V intercept (constant), and
b==Slope of the line.
By t,he simultaneous solution of equations (1)
and (2), the constants a and b can be found, deter-
mining the equation which best describes a sample
distribution. The equation is—
for the cost per capita or population equivalant
studies:
log lOYa+b log X (3);
and for the cost per unit flow study:
log F=a+b log X (3d)
7
-------
FIGURE 1.—Map of 20 Index Cities and Their Assigned Areas of Cost Influence
Equations (3) and (3e) represent the curves of
best. estimate for the sample distribution, so as to
minimize the variance between the known values
and the estimated values. Since the actual cost
estimating curve could not be expected to intersect
all of the unit. project costs, an indication of their
location with respect to the computed curve is
needed. This distribution is described by the
standard error of estimate which relates the sum
of the squares of the difference between the actual
values and the estimated values.
Combining the calculated estimating curve (3)
or curve ( 3 ) with the standard error of estimate,
the sample’s mean value as well as its scatter are
fully described. This completely characterizes
the distribution. Figure 2 graphically illustrates
his relaiionship.
Figure illustrates the typical estimating curve
used in this report. In every case the sample for
each category of treatment. is shown in a log-
arithinic coordinate system. Upon each distribu-
tion is superimposed the appropriate statistically
developed estimatiilLr curve. The solid line is the
expected cost curve as determined from the cal-
culated equation. The dashed lines above and be-
low the solid line define the cost interval. The
interval is calculated to contain one standard error
of estimate (SB) on either side of the calculated
curve.
The correlation coefficient (r) serves as a meas-
ure of the relationship of one variable to the other.
Therefore, to measure how well the cost is de-
scribed by the unit factor or parameter used, this
coefficient r’ is calculated. The value of the cor-
relation coefficient varies between 0 and ±1.
All data used in this analysis were handled with
electronic dat.a processing equipment using the
punchcard technique. This procedure will allow
new data to be introduced at. any future time with
little difficulty, and it eliminates errors common
to manual calculating methods.
Example of Cost Analysis
To illustrate the procedure of analysis used in
this investigation, the activated sludge treatment
category is presented as an example. This ex-
ample concerns itself only with the calculations
involved in the study. Later sections of this re-
port discuss geographical representativeness of
the particular sample.
i.us .aov. ffio , — c 1%,
£a ...rt. N1-R.c.id Cs. ,vcftsu Cost dsz
is
8
-------
Si
FIGURE 2.—Typical Unit Cost Estimating Curve
This analysis relates design population to the
cost. per capita. For this purpose, data from 138
projects were available. The required values for
the solution of the normal least squares equations
(1) and ( ) are:
N=138
X=49 1.69060
X 2 = 1,810.184795
Y= 359.226848
r= 961.938606
XY= 1,292.629537
Substituting these values, then, in equations (1)
and (2) and solving them simultaneously results
in the determination of the constants a and b:
a — 3.6533024
b= —0.2782395
These constants, when substituted in the general
equation (3) with appropriate sign, result in the
equation which best describes the 138 projects on
a cost per capita basis:
log 10 Y=3.65330240.2782395 log X
Since this equation, by definition, represents the
best description of the sample and is only the in-
S
2
a.
3
S
a.
0
S S
Si
100 1,000 . 10.000
Population Served (Design Capacity)
100,000
739—993 O—44-—-—3
9
-------
TABLE II
Construction cost per capita (design capacity)
Activated Sludge Treatment
(1957—59 dollars)
Design popu-
lation
Construction costs
Valid size
range
Lower limit
Expected
value
Upper limit
100
1, 000
10, 000
100, 000
$84. 39
44. 47
23. 43
12. 35
$124. 98
65. 85
34. 70
1& 29
$185. 07
97. 52
51. 39
27. 08
250
to
100, 000
Ratio of the upper limit = 1.4809
Ratio of the lower limit = 0.6753
Correlation coefficient (r) = —0.73
Population S.rv.d (Deiign Capacily)
FIGURE 3.—Construction Cost per Capita—Activated Sludge Plants
$L000
$100
2
‘3
U
$10
10,000
100,000
10
-------
dicated estimating line, it is necessary to define a
reasonable cost range on either side of this curve
to allow for a spread of costs. To establish this
cost range, one standard error is calculated for the
sample at its mean, and applied above and below
the estimated cost level to produce an upper limit
cost and a lower limit cost. These upper and lower
limit costs, when related to the expected cost, yield
a ratio of the upper limit and a ratio of the lower
limit, respectively. These two extremes define the
interval in which the construction cost of activated
sludge systems can be expected to fall approxi-
mately two-thirds of the time.
Since the data used for the independent variable
were limited to only certain portions of the log-
arithmic scale and did not extend to its extremes,
use of the developed cost curves beyond the limits
of the data would be of doubtful validity. In an
attempt to provide the estimators with some guide
to these limitations, a valid size range has been
determined for each estimating equation.
Before this information can be totally func-
tional in making cost estimates it should be repre-
sented in a graphical form for rapid use. To
facilitate the development of a graphic presenta-
tion of this curve, the per capita costs for four
even populations were chosen for ease of calcula-
tion. They were solved from the activated
sludge equation on page 9. A tabulation of these
results is given in table II.
From the data in table II, a logarithmic presen-
tation of design population versus cost can be pre-
pared (fig. 3).
Illustration of Use of Curves
To illustrate the use of the curves as an estimat-
ing tool, a hypothetical problem is presented: Es-
timating the cost of an activated sludge plant to
be constructed in the Denver, Cob., area of in-
fluence for a community of 10,000 population in
March 1964.
Solving the calculated equation (p. 9) for the
expected cost per capita for a community of 10,000
people, a value of $34.70 per person is obtained.
Since this curve was developed on the basis of
1957—59 dollars, it is necessary to update
the estimate using the PHS—STP index. The
March 1964 index value for the Denver area of
influence was 1.0343.
Combining the cost figure, and the PHS—STP
index value, the expected construction cost
is: 1.0343 ($34.70) = $35.89/capita. To convert
this cost to the estimated contract. cost, it is neces-
sary to multiply the cost per capita by the design
popultaion: $35.89 (10,000) $358,900. The cost
of $358,900 is the most probable construction cost.
To obtain the total estimated cost of the project,
the construction cost would be increased by ap-
proximately 20 percent.
To establish the possible cost range for this proj-
ect, in addition to the expected cost estimate, it
is necessary to multiply the expected construction
cost., as tabulated above, by the ratios of the upper
and the lower limits. The result would be the
most probable cost range for the project.
Treatment Plant Costs per Capita
The following stati tkal evaluations of sewage treatment plant construc-
tion costs are based on: (1) cost per capita; (2) cost per population equiv-
alen.t; and (8) cost per unit flow. All data are based on design capacity.
This study covered the relationship between
sewage treatment plant construction cost per cap-
ita and design population. Data from 1,504 proj-
ects have been used. Table I (p. 5) shows the
types of treatment and specific geographic loca-
tions of the plants included in the study.
It should be noted that. the statistical reliability
of the per capita cost data is greatly enhanced
by the large number of projects included in each
sample. The following discussion covers the in-
dividual treatment types on a cost per capita
basis:
I’
-------
FIGURE 4.—Construction Cost per Capita—linhoff Tank Plants
Imhoff Tank Plants
The geographic distribution of Imhoff tank
plants was such that over 86 percent of the proj-
ect s were located east of the Mississippi River.
Six of the 43 plants were located in Kansas Ar-
kansas, Nebraska, and Washington—the only
Western States for which data were available for
this phase of the cost study.
The cost curve best fitting this group of projects
appears in figure 4 and is represented in equation
form by:
log 10 Y =4.1833312—0.50830469 log X
Solution of this equation for specific population
values appears in table III.
TABLE III
Construction cost per capita (design capacity)
Imhoff Tank Plants
(1957—59 dollars)
Design popu-
lation
Construction costs
Valid size
range
Lower limit
Expected
value
lJpper limit
100
1, 000
10, 000
100, 000
$80. 23
24. 89
7. 72
2. 40
$146. 80
45. 54
14. 13
4. 38
$268. 59
83. 33
25. 85
8.
300
10, 000
Ratio of upper limit = 1.8297
Ratio of lower limit = 0.5465
r = —0.60
$100
2
a
0
LI
0
LI
$10
SI
Popu’ation Ssn sd cDuign Capacity)
100,000
12
-------
t 4 1U .
“ui,i:iiti
I1 , , ,
• S
I
I
I
I
Si I 1 I I I i _ - -— — — -- —- I I I I I I I I
100 100,000
Population Svv.d (Design Capodly)
FIGURE 5.—Consfruction Cost per Capita—Imhoff-Type Plants
linhoff-Type Plants
The planLs included in this category were con-
centrated east of the Mississippi River, with only
10 percent dispersed through the Western States.
In addition to this heavy weighting to the east, ________________________________
the sample is strongest in the States of Virginia
and West Virginia where 30 percent of the proj-
ects were centered.
For the 47 projects providing cost data for this ________ _______ _______ _______
type of treatment, the curve best satisfying the
distribution is represented in equation form by:
log 10 Y=3.2640685—0.20820798 log X
Graphically, this curve is represented in figure 5
and can be reproduced by a log-log plot of the data
appearing in table IV.
2
a.
C
A.
LI
:hhuu.1t.. ,
III,I,ttI,Il,
510
TABLE W
Construction cost per capita (design capacity)
Imhoff-Type Plants
(195T—59 dollars)
Design popii-
lation
Construction costs
Valid size
range
Lower limit
Expected
value
Upper limit
100
1, 000
10, 000
100, 000
$50. 93
31. 53
19. 53
12. 09
$70. 41
43. 59
26. 99
16. 71
$97. 33
60. 26
37. 31
23. 10
500
to
10, 000
Ratio of upper limits = 1.3823
Ratio of lower limits = 0.7234
r = —0.40
13
-------
)1,I IU
TABLE V
Construction cost per capita (design capacity)
Primary Treatment—Separate Sludge Digestion
Plants
(1957—59 dollars)
Design popu-
lation
Construction costs
Velid size
range
Lower limit
Expected
value
Upper limit
100
1, 000
10, 000
100, 000
$77. 12
37. 52
18. 25
8. 88
$114. 96
55. 93
27. 21
13. 24
$171. 36
83. 37
40. 56
19. 73
600
to
150, 000
Ratio of upper limit = 1.4906
Ratio of lower limit = 0.6709
r =—0.66
100,000
FIGURE 6.—Construction Cost per Capita—Primary Treatment—Separate Sludge Digestion Plants
Primary Treatment—Separate Sludge
Digestion Plants
More than TO percent of the treatment plants
in this sample were located east of the Mississippi.
Very little cost data were available from the
southwest quadrant of the Nation.
The expected cost curve derived from the 233 ________ ________ ________
plants included in this sample appears in figure
6 and has the equation form of: ________________________________________
log 10 J =3.6863572—0.31290299 log X
Should cost figures be desired for other popula-
tions, within the cost range of 600 to 150,000 design ________________
population, other than those tabulated in table V,
they can be extrapolated from figure 6 or obtained
by use of the equation.
2
a.
‘3
I
I
$10
SI
100
S
1,000 10000
Population Se,ved (Design Capacity)
I I iLl ill
14
-------
S i
2
‘3
U
100,000
FIGURE 7.—Construction Cost per Capita—Stabilization Ponds
Stabilization Ponds TABLE VI
The stabilization pond projects were distrib-
uted over most of the country; however, the sam-
pie was very light in the Northern New England
and Atlantic seaboard areas. The geographic
focal point was the North Central United States,
where approximately 60 percent of the sample _______ _______ _______
originated.
Data from 560 stabilization ponds served as the _________ _______ _______ _______
basis for the construction cost curve which ap-
pears in figure 7. The equation which best de-
scribes this cost distribution is:
log 10 Y= 3.2663415 — 0.35310975 log X
Table VI presents a summary of the per capita
costs calculated for four specific populations.
Construction cost per capita (design capacity)
Stabilizalion Ponds
(1957—59 dollars)
Design popu-
latlon
Construction costs
range
Lower limit
Expected
value
Upper limit
100
1, 000
10, 000
100, 000
$19. 96
8. 85
3. 92
1. 74
$36. 31
16. 10
7. 14
3. 17
$66. 10
29. 31
13. 00
5. 77
100
to
60, 000
Ratio of upper limit = 1.8199
Ratio of lower limit = 0.5495
r ==—0.S4
$100
$10
$1
L
Pop.la ioii Ss,vsd (Dssàgn Copociiy)
15
-------
2
0.
33
0.
V
FIGURE 8.—Construction Cost per Capita—Trickling Filters—Separate Sludge Digestion Plants
Trickling Filter—Separate Sludge
Digestion Plants
The 374 projects making up this category were
dispersed throughout the country. Five States—
Kansas, Minnesota, North Carolina, Texas, and
Wisconsin—contained 30 percent of the sample.
The geographic location of the trickling filter—
separate sludge digestion plants is shown in table
I(p. 5 ).
A summary of costs at different population
levels appears in table VII. This tabulation of
cost results from the solution of the equation
log 10 Y=3.8827576—0 .33673892 log X
which is graphically presented in figure 8.
TABLE VU
Construction cost per capita (design capacity)
Trickling Filter—Separate Sludge Digestion
Plants
(1957—59 dollars)
Design popu-
lation
Construction costs
Valid size
range
Lower limit
Expected
value
Upper limit
100
1, 000
10, 000
100, 000
$109. 83
50. 58
23. 29
10. 73
$161. 91
74. 57
34. 34
15. 81
$23& 70
109. 93
50. 63
23. 32
350
to
100, 000
Ratio of upper limit = 1.4743
Ratio of lower limit = 0.6783
r =—0.66
.
$10
Si
1,CCl 10,000
Popuktiom Serv.d (Design Capacity)
I I I I liii I I L I I
1O(
16
-------
FIGURE 9.—Construction Cost per Capita—Trickling Filter—Imhoff-Type Plants
Trickling Filter—Imhoff-Type Plants
The 1 ll treatment plants of this type are located
throughout the Central United States, and in sev-
eral of the Western States such as Arizona, Cali-
fornia, New Mexico, and Oregon. Almost 30 per-
cent of the projects are located in the two States of
Iowa and Texas. Data were available from very
few projects in the Northern New England States
and the southeastern section of the country.
Table VIII presents unit costs for the four se-
lected population levels. These results are graphi-
cally portrayed in figure 9. The most probable
estimating equation based on the 121 projects in-
volved in the sample is:
log 10 Y=4.4201461—0.52710008 log X
TABLE VIII
Construction cost per capita (design capacity)
Trickling Filter—Imhoff-Type Plants
(1957—59 doflars)
Design popu-
lation
Construction costs
Valid size
raflge
Lower limit
Expected
value
Upper limit
100
1,000
10,000
100,000
$169. 71
50. 42
14. 98
4. 45
$232. 24
69. 00
20. 50
6. 09
$317. 82
28. 05
8. 33
500
to
25, 000
Ratio of upper limit = 1.3685
Ratio of lower limit = 0.7307
r =—0.79
2
‘3
0
S.
1,000 -
Population Served (Design Capacity)
10,000 100,000
17
-------
. luIpsIw
1I
Primary
—. — Secondary Treatment
_-_-___ StabihzaiLn Ponds
1uIs1tu 1 1 I
Popsiation Served (Desl n Capacity)
FIGURE lO.—Construction Cost per Capita—Composite Comparison by Types of Treatment
Activated Sludge Plants
This category of treatment plants included 138
projects representative of all parts of the Nation
except the southwestern area. About 6 percent
of the projects were located east of the Mississippi
River; the influence of the Central United States
and west coast was exerted on the cost data in
this group in a more limited way.
The curve that represents the most probable
construction cost for this type treatment is shown
graphically in figure 3 (p. 10), and is expressed
mathematically by the equation:
log 10 Y=3.6533024—O.27& 3950 log X
Table II (p. 10) presents several solutions to
this equation for different population levels.
Comparison of Per Capita Cost by Types
of Treatment
The individual cost data for the seven different
methods of treatment make it possible to compare
the types, one to the other. Before presenting such
a comparison, it is of interest first to compare
graphically the three general categories of treat-
ment: primary treatment; secondary treatment;
and stabilization ponds. For this analogy the pri-
mary and secondary curves in figure 10 have been
developed from the composite cost data used in the
previously presented curves. This composite cost
information was developed from the following
sources:
$1,000
$100
2
0.
‘3
$10
$1 I I I I liii
100 1,000
I I I I I III
I I I I I t_I 1
10,
100,000
18
-------
,_.. Ii.
FIGURE 1 1.—Construction Cost per Capita—Primary Types of Treatment
(1) Primary curve (IT data; ITT data;
and P-SD data).
(2) Secondary curve (AS data TF—P
data; and TF—ITT data).
The relationship of the secondary cost curve to
the primary curve demonstrates the obvious fact
that the construction of secondary treatment plants
is more expensive than construction of primary
facilities. The relative difference in costs between
stabilization ponds and the two other general cate-
gories is not surprising. These curves reflect only
construction costs and do not include land costs.
Land demands for stabilization ponds may play a
significant role when comparing total costs of this
form of treatment to that of primary and second-
ary processes.
It is desirable to reiterate here that construction
cost—and, in fact, total cost—is not the only factor
involved in choice of degree of treatment and
treatment processes. The choice is often dictated
by other considerations.
Comparative costs—primary treatment.—A
comparison of the three construction cost curves
relating to primary treatment. is shown in figure 11.
The IT cost curve would tend to intersect with
the P—SD curve at a population level of approxi-
mately 300. With increasing population the
Inthoff curve and the conventional primary cost
curve diverge at a very rapid rate. This diver-
gence is due, in part, to the increase in equipment
demands for P—SD facilities. In addition to the
equipment factors, the lower cost of constructing
iiiiinui..i ITT
p——- — P-SD
2
.
‘5
3
$1
t
Popvk$on Ss v.d (Dsse n C ipocity)
1,000 10 000
Ill J
100,000
19
-------
FIGURE 12.—Construction Cost per Capita—Secondary Types of Treatment
the Imhoff self-contained digestion units as com-
pared to separate sludge digestion units of the
conventional primary-type further explains this
cost divergence.
Imhoff-type facilities at the lower population
levels appear to be the least expensive of the three
forms of primary treatment shown in figure 11.
Even with the addition of mechanical equipment,
this type of plant tends to cost less per capita to
build than do Imhoff plants. This cost differen-
tial is explained by the almost exclusive use of
prefabricated package-type plants in these lower
population ranges, which are assumed to be less
expensive to build than are on-site-constructed
Imhoff tanks.
With increasing population the ITT curve inter-
sects the IT cost curve and apparently would tend
to intersect the P—SD curve. This might be attrib-
uteci to the substantial increases in construction
costs of ITT plants at these higher population
levels. These increases could be caused by the
more frequent use of field-constructed facilities
over factory-assembled plants, combined with the
rapid increase in mechanical equipment costs for
increasing populations.
Comparative costs—secondary treatment.—
The final phase of this cost comparison deals with
the graphic collation of the three curves appearing
in figure 12 which cover the secondary treatment
category.
The trickling filter—Imhoff-type plant curve
for the most part reflects a per capita cost which
2
3
a.
0
U
Population Sstvsd (D.sign Capacity)
100,000
20
-------
is lower than either the TF—P curve or the AS
curve. An initial comparison of TF—ITT plants
with TF—P facilities discloses that the only appar-
ent difference between the two is in the method of
primary treatment employed in each. Similar
reasoning would apply to these curves (TF—P and
TF—ITT) as that which applies to the behavior of
the P—SD curve and the IT curve.
The third curve in figure 12 reflects costs per
capita for activated sludge treatment facilities.
It should be noted that this curve is developed
from projects which include prefabricated pack-
age-type plants as well as on-side-constructed
facilities. If sufficient information were avail-
able, a more realistic evaluation of construction
costs for activated sludge systems could have been
obtained by separating the sample into various
modified processes as well as various construction
methods.
At the lower population levels, the AS curve
reflects a lower cost per person than either of the
two trickling filter curves. This condition may
be attributable to the increased economy of
factory-fabricated package-type plants used to
serve these lower population ranges.
With increasing population, however, the AS
curve intersects both filter curves. This intersec-
tion of the two filter curves by the AS curve was
unexpected, since earlier cost. investigations did
not show the activated sludge curve intersecting
the TF—P curve. Based on a considerably larger
sample and improved statistical correlation, it is
believed that the change in slope reflects existing
conditions in the sewage treatment field. It is
further hypothesized that the AS curve could have
been influenced by the use of cost data for modified
AS systems, the construction of which varies in
unit. cost from so-called conventional activated
sludge plants.
Treatment Plant Costs per Population Equivalent
In the design of sewage treatment facilities for
communities receiving both industrial and domes-
tic sewage flows, an accepted criterion of design is
that of population equivalent served. This pa-
rameter converts the industrial component of
sewage flow to a strength equivalent to that of
typical domestic sewage. The use of population
equivalent as a design factor, is generally limited
to secondary treatment facilities because they are
designed to remove and reduce organic loadings.
The steady increase in the construction of sys-
tems receiving both industrial wastes and domestic
sewage led to the decision to collate construction
costs and population equivalents served by proj-
ects of the secondary treatment type. A sample
of 280 projects serves as the statistical basis of
this phase of the study.
Activated Sludge Plants
The statistical basis of the cost data in. this
category is a sample of 86 projects, with 75 percent
located east of the Mississippi River. Nine States
west of the Mississippi account for about one-
quarter of the cost data. Over one-quarter of the
plants were located in the two States of Ohio and
Wisconsin.
The construction cost estimating curve best de-
scribing this sample is shown in figure 13. The
equation, presented in its log form is:
log 10 Y==4.8716858—0.35073541 log I
A tabulation of data that can be used to repro-
duce this cost curve appears in table IX. The
population equivalents listed were chosen for con-
venience of calculation.
TABLE IX
Construction cost per population equivalent
(design capacity)
Activated Sludge Plants
(1957—59 dollarR)
Construction costs
Design —_________
population
equivalent Lower limit Expected Upper limit
value
Valid s1z
range
100
1, 000
10, 000
100,000
$98. 72
44. 02
19. 63
8. 75
$147. 98
65. 99
29. 42
13. 12
221. 83
98. 92
44.
19. 67
250
to
60, 000
Ratio of upper limit = 1.4990
Ratio of lower limit = 0.6671
7. =—0.74
21
-------
s1,
Pop.Id om Eqelvulent (DssI Capadty)
FIGURE 13.—Construction Cost per Population Equivalent—Activated Sludge Plants
Trickling Filter—Separate Sludge
Digestion Plants
Cost data for 143 projects were employed in t.his
facet of the study. Projects were from all parts
of the United States with the exception of 11
States for which no data were available. The
Northern New England area did not influence the
curve. The seven States of North Carolina, Ten-
nessee, Wisconsin, Alabama, Minnesota, Texas,
and Utah, on the other hand, had the greatest in-
fluence on the estimating cost data since approxi-
mately 40 percent of the sample came from these
States. Table I (p. 5) indicates the States which
influenced this cost study.
The expected cost equation is:
log 10 Y=5.1805624—0.42 66682 log X
Costs for selected populations are tabulated in
table X. These data served as an aid in develop-
ing the curve shown in figure 14.
Trickling Filter—Imhoff-Type Plants
This type of treatment included no population
equivalent cost data from States lying west of the
Rocky Mountains. With the exception of cost in-
s100
e
5
0 ’
‘U
C
0
ft
3
1POO
100,000
22
-------
TABLE X
Construction cost per population equivalent
(design capacity)
Trickling Filter—Separate Sludge Digestion Plants
(1957—59 dollars)
Design
population
equivalent
Construction costs
Valid size
range
Lower limit
Expected
value
Upper limit
100
1 000
10 000
100, 000
$149. 25
56. 01
21. 02
7. 89
$213. 41
80. 09
30. 05
11. 28
$305. 16
114. 51
42. 97
16. 12
350
to
75 000
Ratio of upper limit 1.4299
Ratio of lower limit = 0.6994
r =—0.74
100,000
FIGURE 14.—Construction Cost per Population Equivalent—Trickling Filter—Separate
Sludge Digestion Plants
C
0
U.
C
0
3
100 1,000 10,000
Population Equivalont (Design Capacity)
23
-------
Population Equivalent (Dssi n Capacity)
100,000
FIGURE 1 5.—Construction Cost per Population Equivalent—Trickling Filter—Imhoff-Type Plants
formation from the eight States of Arkansas,
Iowa, Kansas, Louisiana, Minnesota, Missouri,
Oklahoma, and Texas, all the data were from
States lying east of the Mi si ippi River. It
should be noted that representation was only
nominal from the southeastern portion of the Na-
tion. Over one-quarter of the cost data from the
51 projects making up the sample were provided
by two States: North Carolina and Texas.
The estimating cost curve for this type of treat-
ment is:
log 10 Y==5.6743457—0.59384713 log X
Unit cost results appear in table XI and are shown
graphicafly in figure 15.
TABLE XI
Construction cost per population equivalent
(design capacity)
Trickling Filter—Imhoff-Type Plants
(1957—59 dollars)
Construction costs
Design
population
equivalent
—___________________
Valid size
range
Lower limit
Expected
value
Upper limit
100
1, 000
10, 000
100, 000
$195. 95
49. 92
12. 72
3. 24
$306. 66
78. 13
.19. 90
5. 07
$479. 92
122. 27
31. 15
7. 94
400
to
20, 000
Ratio of upper limit = 1.5650
Ratio of lower limit = 0.6390
r=—0.73
$100
C
C
0
0
Lu
C
0
a
0.
Si
100
10,000
24
-------
Sli
Population Equivalent (Design Capacity)
100 ,000
FIGURE 16.—Construction Cost per Population Equivalent—Cost Comparison by Types of Treatment
Comparison of Population Equivalent
Costs by Types of Treatment
As in the case of the cost per capita study, a
comparison of the data on costs per population
equivalent is desirable. Figure 16 reflects the
three cost per PE curves for secondary treatment
plants.
Since the PE, by definition, is always higher
than the design population when industrial wastes
are included in the sewage flow, the cost per PE
will naturally be less than the cost per capita for
the same project. In all cases, the cost per PE
curves have a greater slope than the cost per capita
curves. This slope differential evidently is caused
by the increasing percentage of industrial wastes
with increasing population.
The relationship of the three curves in figure
16 is generally the same as for the cost per capita
curves of the three secondary treatment processes
shown in figure 12 (p. 20). This was expected
since, in a large number of the projects, particu-
larly in the smaller communities, the design popu-
lation was equal to the PE. In the remainder,
the effect of PE would have been felt equally by
similar secondary treatment processes.
. — — — — TF-P
I — I — I — I
C
0
L I ’
C
0
.
&
3
510
Si
100
1,000 10,000
25
-------
$10,000,000
$1,000,000
C
0
I
a-
$100,000
“I I I I 11111
.01 .1 1 10
MIlhon Gallons Per Day Design Flow
FIGURE 17.—Construction Cost per Unit Flow—Imhoff-Type Plants
I I I I lIlt I I I I liii
26
-------
Treatment Plant Costs per Unit Flow
The third group of cost comparisons collates
construction cost against design flow in million
gallons per day (MOD). The sample involved
in establishing these cost data was composed of
plants receiving principally domestic flow; how-.
ever, in many cases both domestic and industrial
wastes flows were involved. Cost data from over
750 projects formed the statistical basis for the
estimating curves on cost per unit design flow.
Since hydraulic loading is one of the basic fac-
tors in the design for both primary and secondary
systems, curves were developed for both categories
of treatment. For lack of sufficient data, Imhoff
tank plants have been. eliminated from this por-
tion of the report.
The only other difference between this portion
of the report and the preceding two sections is that
the construction cost is not multiplied by a factor
of 10. Hence, the general equation for the curves
will take the form:
log Y=i z+b logX (3 )
Tmhoff -Type Plants
The geographic distribut.ion of the 24 projects
from which the cost curve for this type of treat-
ment was developed is not balanced. The sample
covers almost every State east of the Mississippi
but only t.he three States of Washington, Mon-
tana, and Louisiana west of the Mississippi. The
State of West Virginia exerted the greatest in-
fluence on the curve by supplying 25 percent of
the sample.
The most descriptive curve developed by the
cost information available in this sample is repre-
sented by this equation:
log Y= 5.2132021 — 0.27927265 log X
Figure 17 reflects this equation in graphic form.
A tabulation of the expected costs for four se-
lected flows appears in table XII.
TABLE X I I
Construction cost per unit flow (design capacitg)
Imhoff-Type Plants
(1957—59 doUars)
Design flow
(MOD)
Construction costs
Valid size
range
(MOD)
Lower limit
Expected value
Upper limit
0. 1
1. 0
10. 0
100. 0
8337, 128. 70
177, 224. 23
93, 164. 53
48, 975. 39
$451, 498. 84
237, 347. 14
124, 770. 38
65, 590. 18
$604, 668. 78
317, 866. 61
167, 098. 44
87, 841.
0. 095
to
1. 100
Ratio of upper limit 1.3392
Ratio of lower limit = 0.7467
r =—0.56
Primary Treatment Separate Sludge
Digestion Plants
The 99 projects of this type for which cost data
could be obtained were not uniformly distributed
over the country. Most were located east of the
Mississippi River with some of the data stemming
from States in the northwestern quarter of the
Nation. With the exception of seven projects in
California, the southwestern portion of the coun-
try exerted little influence on the cost data. The
six States of Vermont, New York, Michigan, Illi-
nois, Idaho, and California accounted for 38 per-
cent of the sample.
Cost ranges and expected costs for specific flows
have been tabulated in table XIII. These data
were the basis for the estimating cost curve ap-
pearing in figure 18. The equation which best de-
scribes the sample is:
log Y=5.70855960.44604400 log X
27
-------
TABLE XI II
Construction cost per unit flow (design capacity)
Primary Treatment—Separate Sludge Digestion
Plants
(1957—59 dollars)
Design
flow
(MOD)
Construction costs
Valid
size
range
(MOD)
Lower limit
Expected value
Upper limit
0. 1
1. 0
10. 0
100. 0
$429, 165. 71
153, 667. 03
55, 022. 07
19, 701. 21
$655, 347. 62
234, 653. 70
84, 020. 18
30, 084. 28
81, 000, 733. 50
358, 322. 54
128, 301. 08
45, 939. 51
0. 40
to
ooo
Ratio of upper limit = 1.5270
Ratio of lower limit = 0.6549
r = — 0.76
10
FIGURE 18.—Construction Cost per Unit Flow—Primary Treatment—Separate Sludge
Digestion Plants
SI
0
0
8
U
Se
s1ol000
.01
.1
Million Gallons P., Day Design Flow
28
-------
FIGURE 19.—Consthiction Cost per Unit Flow—Stabilization Ponds
Stabilization Ponds
The projects which formed the basis of this cost
study were well distributed nationally except for
the New England area, where only two projects
were located—one each from Maine and New
Hampshire. The eight States of Arkansas, Kan-
sas, Mississippi, Missouri, Nebraska, North Da-
kota, Oklahoma, and South Dakota, provided 61
percent of the data. The State of Missouri exerted
the greatest single influence on the cost data with
50 of the sample’s 350 projects located in that State.
The equation which best characterizes the avail-
able cost data statistically is:
log 1= 5.107643 — 0.42570017 log X
This equation is presented in graphic form in figure
19. Expected costs for specific flows have been
tabulated in table XIV.
TABLE XIV
Construction cost per unit flow (design capacity)
Stabilization Ponds
(1957—59 dollars)
Design flow
(MOD)
Construction costs
Valid size
range
(MOD)
Lower limit
Expected value
Upper limit
0. 1
1. 0
10. 0
100. 0
$99, 599. 29
37, 372. 83
14, 023. 48
5, 262. 06
8180, 32& 36
67, 664. 96
25, 390. 05
9, 527. 16
$326, 491. 46
122, 510. 01
45, 969. 67
17, 249. 29
0. 01
4. 50
Ratio of upper limit 1.8105
Ratio of lower limit = 0.5523
r =—0.65
s1,000I000
a.
0
I
$
.
Slop®
Million Gallons P., Day Design Flow
10
29
-------
$10,000I000
$100,000
$10,000
.01
FIGURE 20.—Construction Cost per Unit Flow—Activated Sludge Plants
Activated Sludge Plants
The 77 projects which provided data for the ac-
tivated sludge category were centered principally,
in the eastern one-half of the country and portions
of the southwestern area. The Northwestern
States exerted very little influence on the cost curve,
with information available only from Oregon,
Idaho, Wyoming, Utah, and Nebraska. Projects
for this sample were heavily centered in the States
of Ohio and Wisconsin where 30 percent of the cost
data originated.
The expected estimating equation is:
log Y 5.1954670 — 0.22968263 log X
Figure 20 is the graphic presentation of this equa-
tion. Table XV reflects the costs per unit flow for
the four specific flow loadings.
TABLE XV
Construction cost per unit flow (design capacity)
Activated Sludge Plants
(1957—59 dollars)
Design flow
(MOD)
Construction costs
Valid size
range
(MOD)
Lower limit
I
Expected value Upper limit
$544, 630. 08 $810, 743. 42
320, 936. 477, 750. 02
189, 119. 58 281, 525. 87
111, 443. 26 165, 895. 89
0. 1
1. 0
10. 0
100. 0
$365, 864 12
215, 594. 21
127, 044. 15
74, 863. 82
0. 015
to
4. 00
Ratio of upper limit = 1.4886
Ratio of lower limit = 0.6718
r =—0.60
g
0
I
Million Gallons Per Day Design Flow
10
30
-------
10,o00,000
0
0
0
I
FIGURE 21.—Construction Cost per Unit flow—Trickling Filters—Separate Sludge Digestion Planta
$1,000,000
$100,000
$10,000
Trickling Filter—Separate Sludge
Digestion Plants
The 141 projects which made up the TF—P cate-
gory were distributed throughout the Nation.
The three States of North Carolina, Tennessee, and
Wisconsin accounted for almost one-fourth of the
sample, with 10 projects each.
The estimating equation is:
log Y=5.7630257—0.43474996 log X
Solution of this equation for particular flows will
yield expected costs per MGD. A tabulation of
these costs for four specific flows appears in table
XVI. A graphic presentation of these data is
given in figure 21.
TABLE XVI
Construction cost per unit flow (design capacitg)
Trickling Filter—Separate Sludge Digestion
Plants
(1957—59 dollars)
Design
flow
(MOD)
0. 1
1.0
10. 0
100. 0
Construction costs
Valid
size
range
(MOD)
Lower limit
Expected value
Upper limit
$529, 168. 77
194,466.21
71, 465. 12
26, 262. 99
$782, 575. 04
287,591.43
105, 688. 06
38, 839. 70
$1, 157, 331. 50
425,312.09
156, 299. 55
57, 439. 11
0. 10
to
5 00
Ratio of upper limit = 1.4789
Ratio of lower limit = 0.6762
r = —0.73
. $
.
S
.01
.1
Million Gallons Per Day Design Flow
10
31
-------
$10,000,000
Million Gallon. Per Day Design flow
1
FIGURE 22.—Construction Cost per Unit Flow—Imhoff-Type Plants
10
Trickling Filter—linhoff-Type Plants
The 51 projects in the sample were, for the most
part, located east of the Mississippi River. Mas-
sachusetts was the only New England State
providing data: seven Western States were repre-
sented. Table I (p. 5) lists the projects contrib-
uting to this arnp1e. The greatest cost data in-
fluence for this curve was provided by the three
states of Texas, Iowa, and North Carolina, in
which 40 percent of t.he projects were located.
The equation best fitting the distribution of cost
data is:
log Y 5.3845334 —0.32009689 log X
TABLE XVII
Construction cost per unit flow (design capacity)
Trickling Filter—Imhoff-Type Plants
(1957—59 dollars)
Design flow
(MOD)
Construction costs
I
Lower limit Expected value Upper limit
Valid size
range
(MOD)
0. 1
1. 0
10. 0
100. 0
8399, 469. 35
191, 155. 27
91, 472. 28
43, 71. 62
8555, 059. 78 $771, 251. 57
265, 608. 871 369, 061. 61
127, 100. 08 176, 604. 65
60, 820. 35 84, 509. 44
0. 10
to
1.00
Ratio of upper limit = 1.3895
Ratio of lower limit = 0.7197
r =—O.63
$1.
a
C
0
Si
$10,000
I I Ill I I I liii
32
-------
J
0
1
FIGURE 23.—Construction Cost per Unit Flow—Cost Comparison by Types of Treatment
s100,000
$10,000
.01
The cost estimate curve for this category of treat-
merit appears in figure 2. Table XVII tabulates
the unit cost information used in preparing the
graphic presentation of the estimating data.
Comparison of Unit Flow Costs
by Types of Treatment
A comparison of the composite primary and
secondary treatment cost curves with the individ-
ual curve, for stabilization ponds, in terms of vol-
ume of flow provided in the design, appears in
figure 3. A similar comparison wa.s undertaken
in the “Treatment Plant Costs Per Capita” section
of this report (p.18).
As expected, the primary treatment curve (com-
posed of IT data; ITT data; and P—SD data) re-
flects a lower unit cost than does the secondary
treatment cost (based on AS data; TF—P data;
and TF—ITT data). However, the relationship
of these two curves, with respect to each other,
varies from that noted in the previous comparison
of the same two category curves in the cost per
capita study. The differences in slope which ex-
ist between the curves as seen in figure 23 and
the cost per capita curves in figure 10 (p. 18) occur
between the primary category curves and the sta-
bilization pond curves, respectively.
Primary Treatment
—‘—I— Secondary Treatment
— — — —U — Stabilization Ponds
Million Gallons Per Day Design How
33
-------
FIGURE 24.—Construction Cost per Unit Flow—Primary Types of Treatment
By subdividing the primary treatment category
curve into its two component parts, a comparison
of the ITT curve to the P—SD cost curve is shown
graphically in figure 24.
The ITT curve depicts lower unit costs at the
smaller flow levels, no doubt due to the preponder-
ance of package-type plant systems in this size
range. Package-type plant systems are assumed
to be less expensive to construct than on-site-con-
structed facilities for the small population groups.
The two curves intersect at the 1—MGD point
and, thereafter the ITT curve reflects a higher
unit cost. This phenomenon was anticipated, due
to the combined effect of the greater cost of on-
site-constructed facilities for the larger flows and
the rapid increase in costs of the mechanical
equipment used in these larger ITT-type plants.
Figure 25 presents a graphic comparison between
the AS cost curve and the TF—P curve per unit of
flow. The curves reflect the same relationship
to one another that they did in the other two studies
of cost per capita and cost per population equiva-
lent. The only difference that was observed be-
tween these two curves was their wider divergence.
This greater angle of inclination between the AS
and TF—P curves could have been caused by the
effects of increased flows from industrial sources
in the larger communities.
The TF—ITT curve in figure 25 shows an unusual
and unexpected pattern with respect to the other
two curves, as compared to earlier studies. Based
upon its statistical characteristics, however, the
curve best describes the sample which it represents.
tHItIII iiij (IT
U—
$1,ooo,
8
C
0
U
Million Gallons P .r Day Flow
.01 .1 1 10
34
-------
II11I11I*114r1I f4tf,d,.w. •,,, 4 :
— — — — . TF—P
— u i IF— ITT
i.u..i.uuu.u AS
I I I Iii
‘Toi
Million Gallons P .r Day D.s. n Flow
FIGURE 25.—Construction Cost per Unit Flow—Secondary Types of Treatment
st,ooos
0
I
3
s100,o®
I I I I I I
I I I __ I I III
10
35
-------
Limitations o Cost Estimating Data
Before these curves can be used as effective, real-
istic estimating aids by municipal officials, plan-
ners, developers, designers, regulatory agencies,
and investors in sewage works bonds, the limita-
tions of cost estimating tools must be understood.
There are two general limitations that affect the
use of the estimating data presented in this report.
The first stems from the geographic distribution
of the projects which made up the samples for each
type of treatment. Certain areas of the United
States may have had little or no influence on the
raw cost data, while certain other areas may
have been proportionately overrepresented. This
emphasizes the importance of the PHS—STP cost
index system, which largely compensates for cost
variations associated with different geographic
locations.
The second limitation which may affect the use
of these estimating curves is the limited plant sizes
or capacities covered by the curves. Extension of
any of the curves would be unwarranted, since
there is no statistical basis for such extrapolations.
The curves should be used only within their valid
size ranges, as stated in the presentation of results
for each curve, since reliable data did not exist
beyond these limits.
Within these limitations the estimating curves
should be useful for determining the probable cost
of construction for a particu1ar type of treatment
plant. It must be noted that cost estimates ob-
tained from these curves will be unit cost estimates
for construction only expressed in 1957—59 dollars.
TABLE XV1II
Available unit cost estimating curves
Treatment type
Cost per capita
Cost per popula-
tion equivalent
Cost per unit
flow
Figure
Page No.
Figure
Page No.
Figure Page No.
IT 4 12 None
ITT 5 13 None
P-SD 6 14 None
SP 7 15 None
AS 3 10 13
TF-P 8 16 14
TF- ITT 9 17 15
22
23
24
None
17 26
18 28
19 29
20 30
21 31
22 32
Updating of any estimate may be accomplished
through use of the PHS-STP index in accordance
with the illustration presented on page 11.
To obtain the total cost estimate for a project,
the construction cost estimate must be increased
by 20 percent to properly consider engineering,
legal, administrative, and other costs not included
in the construction contracts which the Public
Health Service used in its studies. Land costs
must also be added.
The cost curves presented in this report are set
forth as aids in the preparation of cost estimates
for sewage treatment plant construction. How-
ever, these costs are statistically derived and there-
fore can only be used as a guide to the actual costs.
For t.his reason, instead of limiting the costs esti-
mating curves to a single curve, each one has been
encased in a “cost envelope,” or an interval between
high and low costs. The construction cost of a
particular system can be expected to fall within
this cost interval approximately two-thirds of the
time. Sixteen individual cost estimating curves
have been developed for seven different types of
treatment plants. A cumulative summary of these
curves available for estimating purposes is pro-
vided in table XVIII.
Table XVIII indicates that more than one esti-
mating curve has been developed for six of the
seven treatment plant types. This choice of curves
raises the question, “Which curve gives the best
cost estimate?” It is believed that the unit cost
estimating curves based on design flow are the most
responsive indicators of sewage treatment plant
construction costs. However, each estimator must
determine for himself which parameter best serves
his purpose for his specific project.
When comparing the individual curves devel-
oped for the same type of treatment process there
is a distinct probability that several different esti-
mates will be obtained. In all cases, however, esti-
mates derived from these curves should fall within
the “cost interval” for the same individual treat-
ment category in the various cost studies. Minor
variations can be rationalized and should cause no
significant concern.
36
-------
Recapitulation of the Studies
(1) Detailed analyses of construction costs for
sewage treatment plants of various types and sizes
in all seetions of the United States, built with Fed-
eral grants-in-aid, have made it possible to develop
dependable preliminary cost estimating data, re-
fleeting the overall conditions influencing the cost
of construction of similar facilities.
(2) Utilization of the PHS—STP Index as the
base of the study has assured more representative
estimating cost data since the index more specifi-
cally reflects the conditions involved in sewage
treatment plant construction.
(3) More extensive cost data in each treatment
category, obtained from the increased number of
construction projects aided by Federal grants, have
statistically strengthened the estimating data de-
veloped from this information.
(4) More definitive subdivisions of major treat-
ment categories have made it possible to develop
more specialized cost estimating data for these
processes.
(5) Periodic reexamination of available cost
information will be. necessary to insure reliability
of future cost data, and to reflect changing trends
in sewage treatment processes and plant construc-
tion methods.
References
1. Howells, D. H., and Dubois, D. P., “Design
Practices and Costs for Small Secondary
Sewage Treatment Plants in the Upper Mid-
west.” Sewage and Industrial Wastes, 30,
11, 1327 (November 1958).
2. Thoman, J. R., and Jenkins, K. H., “How to
Estimate Sewage Plant (Costs) Quickly.”
Engineering News-Record, 161, 64 (Decem-
ber 25, 1958).
3. Howells, D. H., and Dubois, D. P., “The De-
sign and Cost of Stabilization Ponds in the
Midwest,” Sewage and Industrial Wastes,
31, 7, 811 (July 1959).
4. Rowan, P. P., Jenkins, K. H., and Butler,
D. ‘W., “Sewage Treatment Construction
Costs,” Journal Water Pollution Control
Federation, 32, 6, 594 (June 1960).
5. Rowan, P. P., Jenkins, K. L., and Howells,
D. H., “Estimating Sewage. Treatment
Plant. Operation and Maintenance Costs,”
Journal Water Pollution Control Federa-
tion, 33, 2, 111 (February 1961).
6. Anon., “Sewage Treatment Plant Construc-
tion Cost Index.” U.S. Department of
Health, Education, and Welfare, PHS Pub-
lication No. 1069, Washington, D.C. (1963).
7. Fisher, R. A., “Statistical Method for Re-
search ‘Workers,” Sixth edition, Oliver and
Boyd Ltd., Edinburgh, London (1936).
8. Croxton, F. E., and Cowden, D. J., “Applied
General Statistics,” Second edition, Pren-
tice-Hall, Inc., Englewood Cliffs, New
Jersey (1959).
37
U.S. GOVERNMENT PRINTING OFflCE: 1964 Q—739—993
------- |