REVISED ECONOMIC IMPACT ANALYSIS
OF PROPOSED REGULATIONS
ON ORGANIC CONTAMINANTS IN DRINKING WATER
SUBMITTED TQ:
OFFICE OF DRINKING WATER
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C.
BY:
TEMPLE, BARKER & SLOANE, INC.
15 WALNUT STREET
WELLESLEY HILLS, MASSACHUSETTS 02181
JULY 5, 1978
-------
CONTENTS
PREFACE
I. SUMMARY
II. GAC COSTS FOR THE INDIVIDUAL WATER SYSTEM
III. NATIONAL ECONOMIC IMPACT OF THE REGULATION
Page
i
1-1
II-l
III-l
IV. FEASIBILITY OF FINANCING GAC TREATMENT
IV-1
Appendices
APPENDIX A:
APPENDIX B:
APPENDIX C:
APPENDIX D:
ANALYSIS OF GAC COSTS FOR SELECTED CITIES
ESTIMATED COST FOR GRANULAR ACTIVATED
CARBON (GAC) FACILITIES-NEW ORLEANS
WATER TREATMENT PLANTS
CAPITAL MARKETS RATIO ANALYSIS FOR
SELECTED WATER SYSTEMS
ORGANIZATIONS CONTACTED DURING THIS REVIEW
A-l
B-l
C-l
D-l
-------
I. SUMMARY
Since tbe U.S. Environmental Protection Agency (EPA) pro-
posed a regulation for the control of organic chemical contam-
inants in drinking water in February 1978, the Agency has re-
ceived a number of comments regarding the costs and financial
feasibility of compliance with the regulations.1 This report
presents the results of an eight-week reassessment of the eco-
nomic analyses prepared in 1977 as the regulation was being
developed.*
The focus of this reassessment has been on the costs and
financing of granular activated carbon (GAG) treatment systems.
Although not the only treatment technique available for reduc-
tion of trihalomethanes, GAC is the most expensive treatment
technique contemplated under these regulations and the one on
which the greatest number of public comments has been received
thus far in the public comment period.
Three specific subjects have been addressed in this review:
• The capital and operating costs for granular
activated carbon treatment installed at in-
dividual water systems
• The implications of changes in those unit
costs for the economic impact of the pro-
posed regulation at the national level and
to residential customers of affected water
systems
• The ability of water systems to raise the
capital which would be needed to install
GAC treatment.
U.S. Environmental Protection Agency, "Interim Primary Drinking
Water Regulations; Control of Organic Chemical Contaminants in
Drinking Water," Federal Register, February 9, 1978.
2
Economic Impact Analysis of a Trihalomethane Regulation for
Drinking Water, prepared by Temple, Barker & Sloane, Inc. for
EPA, Office of Water Supply, August.1977; and Economic Analysis
of Proposed Regulations on Organic Contaminants in Drinking
Water, prepared by Temple, Barker & Sloane, Inc. for EPA,
Office of Water Supply, December 13., 1977.
T
B
S
-------
1-2
The examination included five major activities. First,
selected equipment manufacturers and carbon suppliers were
again contacted to verify or supplement previous data. Second,
GAG cost estimates submitted to EPA by some water utilities were
analyzed. An effort was made to understand fully the basis
of estimates developed by three cities which had prepared the
most thorough projections: New Orleans, Indianapolis, and
Louisville. In fact, two of them were visited in person and
the third was contacted by telephone in this process. This
effort was meant to supplement four case studies conducted
in 1976 as a basis for the costing methodology.
A recognized consulting engineering firm, Gannett Fleming
Corddry and Carpenter, Inc., was engaged to visit New Orleans
and to prepare an independent preliminary estimate of.capital
costs for GAG treatment at New Orleans. The purpose of this
element of the review was to have a professional engineering
firm examine the specific costs encountered at one water sys-
tem. It was felt that this would help reconcile differences
between EPA and the industry in this instance and could pro-
vide some feedback on more general assumptions.
In the evaluation of national and customer level impacts
of the revised cost estimates, the Temple, Barker & Sloane, Inc.
computerized Policy Testing Model (PTm)3 of water utilities
was again utilized. The model traces the effects of construc-
tion and operating cost impacts through the industry's financial
structure to identify those economic effects.
Finally, to address the capital markets.issues regarding
the financing of GAG installations, representatives of Moody' s
Investors Service and two other financial institutions were
contacted. These discussions identified the key financial
ratios by which the financial investment community evaluates
water utility bond issues. The project team then analyzed
those ratios and the general financial condition of a sample
of 27 water systems to determine the relative ease or diffi-
culty such systems would have financing GAG installations at
a range of costs.
The actual costs faced by a water system installing GAG
would vary widely depending on a variety of factors. Perhaps
most important is the quality of its raw water, which will
be reflected in the contact time and regeneration frequency
resulting from the pilot studies. Another is the layout of
3
For a description of PTm see Appendix A, Economic Impact
Analysis of a Trihalomethane Regulation for Drinking Water,
op. cit.
-------
1-3
the existing treatment plant and the ease or difficulty of
modifying it to accommodate the GAG equipment. Finally, each
water system will have to make certain policy decisions con-
cerning such matters as the amount of growth in demand to pro-
vide for and the amount of redundancy to allow in contactors
and furnaces. The cost estimates presented here reflect more
conservative assumptions in all these areas.
The major findings of this review process are:
• The unit costs for GAG installations are now
being revised upward from the figures pub-
lished in 1977. Capital costs are being
increased generally by 50 to 80 percent,
primarily to adjust for inflation, to allow
for contingencies and higher design, legal
and financing fees, and to incorporate some-
what more conservative design parameters.
Operating and maintenance expense estimates
remain at levels very close to the former
estimates.
• The range of technical assumptions has been
narrowed somewhat in this review with a re-
sulting impact on the change in the national
cost estimates. The lower end of the pre-
vious range of cost estimates has been raised
because a six-month regeneration frequency is
no longer included in this analysis, although
it may be adequate in some cases. The high
end of the range has been reduced somewhat to
reflect the estimate that 11 systems affected
only by the THM regulation and 61 systems in
total would install GAG treatment. The na-
tional capital cost based on these assumptions
and the December 1977 unit costs would be $352
to $585 million. The new unit costs presented
in this report raise this range to $616 to $831
million.
• The estimates of increases in local residential
water bills for customers of systems installing
GAG treatment have been increased by approxi-
mately 30 to 50 percent and could be higher for
systems with significant site specific problems
in implementation. The effect on residential
bills for the average customer will range from
$7 to $26 per year depending upon system size,
design parameters, and loe-al conditions.
-------
1-4
• Financing in the capital markets should be
possible through normal financing channels
for almost all of the systems under either a
low or high GAG cost scenario presuming that
rates are increased to cover the financing
costs and the 0/M expenses associated with
GAC treatment addition. Under the low cost
scenario almost all could finance GAC with
little or no difficulty, but under the high
cost scenario about half of the systems would
have some difficulty and would have to phase
in their financing, experience some decline
in financial strength, and/or perhaps increase
revenues more than just the amount required
to cover GAC capital and operating costs. A
small number of systems would have major dif-
ficulty in financing and would probably have
to arrange special financing or apply for
relief under the Safe Drinking Water Act.
The analysis performed during this economic review will
continue as more comments are received by EPA during the re-
mainder of the public comment period.
-------
II. GAG COSTS FOR THE INDIVIDUAL WATER SYSTEM
INTRODUCTION
Following EPA's proposal in February 1978 of an organics
regulation for drinking water, many of the public comments
dealt with cost estimates for the installation and use of GAG
treatment. TBS and its subcontractor, Energy Resources Company,
Inc. (ERCO), have reviewed the estimates presented at the public
hearings and identified the major areas of difference in design
and costs. The cost estimates used in EPA's earlier analyses
have been reviewed and, in some areas, revised based upon the
comments of the industry. Continuing analyses are expected
to yield more information on alternative designs, local cost
variations, and minor engineering requirements.
This chapter discusses the areas of difference between the
TBS/ERCO estimates used by EPA and the estimates and comments
provided by the water utility industry. It points out the bases
for both sets of estimates and identifies the degree to which
the earlier EPA estimates are being modified on the basis of
this review.
The sections which follow discuss the major design and
cost issues. First, the system design issues such as the
sizing of hydraulic components and contactors are discussed.
The design size of the system affects most of the costs dis-
cussed subsequently. Second, each of the major capital cost
components is reviewed, with sources of data and the basis
for costs in both TBS/ERCO and industry estimates noted where
possible. Finally, operating and maintenance (0/M) cost is-
sues and issues related to accounting and financing practices
are covered.
After presenting the issues individually, the chapter
presents a compilation of the revised cost estimates for water
systems of three sizes. A comparison with some of the indus-
try's estimates is presented in Appendix A. Appendix B con-
tains an independent preliminary engineering cost estimate
of GAC costs for New Orleans prepared by Gannett Fleming
Corddry and Carpenter, Inc.
-------
11-2
SYSTEM DESIGN ISSUES
A few key design concepts are at the heart of the major
differences in engineering cost estimates prepared by TBS/ERCO
and some utilities in the industry. These include:
• Selecting the flow rate at which to size the
initial carbon fill. If the pilot test re-
sults indicate 18 minute contact time is de-
sirable, for example, should carbon be added
to provide 18 minute contact time at design
flow, at peak month average daily flow, at
annual -average daily flow, or at some other
level? The TBS/ ERGO estimates are based up-
on lower flows than are some of the industry's
projections, yielding correspondingly lower
costs.
• Deciding upon basic contactor design from the
major alternatives of in-ground reinforced con-
crete filter-like units, above-ground steel or
concrete gravity flow contactor vessels, and
above-ground pressure contactors. The above-
ground units assumed by TBS/ERCO are specific-
ally designed for GAG application, but some
utilities would prefer units similar to the
rapid sand filter units with which they are
familiar.
• • Designing for specific local conditions, es-
pecially for multiple treatment plants in some
water systems, versus designing for idealized,
"typical" systems. For its national analysis
TBS/ERCO opted for the latter design basis,
and as a result, the current review is adding
- " ~~ to the estimates allowances for contingencies
and for a range of predictable local conditions.
Each of these design issues is discussed conceptually below.
The actual cost implications of each are addressed in the following
section on capital costs.
One important set of design assumptions should be identified
at the outset because it underlines all of the economic estimates
and is uncertain. These assumptions concern the eventual outcome
of the pilot tests with respect to GAG performance. The EPA anal-
ysis is based on two key assumptions: (1) that 9 minute empty bed
contact time for T.HM removal and 9 to 18 minute empty bed contact
time for synthetic organics removal-will be sufficient to yield
-------
11-3
60 day carbon bed life; and (2) that the adsorption capacity
of each pound of carbon will be relatively constant over the
range of contact times experienced by a plant designed to
achieve 9 to 18 minutes. These assumptions are intended to
provide a range which, in EPA's judgment, should encompass
most systems which will be affected by the regulation. How-
ever, if the ultimate requirements are substantially more
or less severe, then the economics would be significantly
changed in the same direction.
Contactor and Carbon Fill Sizing
The GAC system design used as a basis of EPA's economic
analysis of the organics regulation is expected to be able to
accommodate all possible flow rates through a plant, up to and
including a plant's design or hydraulic capacity. All system
elements which affect flow rates, such as piping and pumps,
would be sized at desi'gn capacity. Other hydraulic elements,
such as contactors, would be sized to accommodate variable flow
rates, with corresponding variable treatment rates (i.e., con-
tact times).
Contactors in the original TBS/ERCO analysis were designed
to provide the desired contact time at a system's average daily
flow, and could handle flows up to design capacity with propor-
tional reductions in contact time. That means, for example, in
the case of New Orleans if 18 minute contact time were desired,
that on average over a year, water would receive 18 minutes con-
tact with GAC. At peak periods the contact time would be less,
down to 14 minutes for the average day in the maximum month over
the last five years. During periods when the demand is less than
the annual average flow, the contact time would of course exceed
18 minutes. It was assumed that the adsorption efficiency of
each pound of carbon would not change appreciably within this
range of contact times.
The major reason for selecting average day flow as the basis
for sizing in the December 1977 analysis was economic. Other
bases, such as operating practices, could be selected as design
objectives by individual systems. For instance a system could
design additional contactor and carbon capacity to meet an oper-
ating criterion that regeneration cycles never be less than some
target number of .days.
Some of the water systems which prepared their own cost
estimates based the contactor and carbon fill sizing on the
design flow of the systems. Thus, if 18 minute contact time
-------
II-4
were desired as a result of the pilot test work, enough carbon
would be used to provide for that at the system's design flow.
Since many system's average daily flow is 50 to 60 percent of
'capacity, such sizing would actually yield a contact time of 30
to 36 minutes on average throughout the year.
The design with such higher actual contact time can be
rationalized, but is unlikely to actually be financed and built
when detailed engineering studies are performed. One rationale
offered for the higher contact time design is that it provides
a safety margin over the target contact time. However, a margin
to allow for variations in flows, water quality, temperature,
etc. should already be incorporated into the target contact
time which results from the pilot testing.
Another justification for larger contactor and carbon siz-
ing is to allow for future growth in system demand. However,
this should also be incorporated explicitly into the design,
in terms of projected flow rates. In fact, system growth has
been incorporated into the TBS/ERCO analysis in three ways.
First, the system production requirements used in the analysis
are projected 1981 figures, to account for growth between now
and then. Second, the contactor systems are expected to be of
modular design so that additional contactors could be added
over time as demands increase. Finally, many large systems
are experiencing little or no real growth of demand, and have
relatively stable production needs.
A final rationale for significantly increased carbon fill
sizing is that increases in contact time will simply lead to
longer carbon bed life, longer regeneration cycles, and there-
fore intuitively should result in reduced operating and main-
tenance costs for carbon regeneration. Unfortunately, while
longer regeneration cycles would result, the 0/M costs for
regeneration would remain relatively constant.
. As Figure II-l shows, the volume of carbon required in a
water system is, for practical purposes, linearly related
to the contact time desired for a given flow. For each one
MGD flow to be treated with a 10 minute contact time, for
example, 24,000 pounds of GAG are required. Thus, for 20
minute contact time with the same flow, 48,000 pounds are
required. If the carbon bed life has doubled when going from
10 minutes to 20 minutes, the daily volume of carbon to be
regenerated will remain unchanged. No 0/M cost savings
result and higher capital costs are incurred.
-------
II-5
Figure 11-1
VOLUME OF CARBON REQUIRED
FOR SELECTED CONTACT TIMES
72
48
'3"
10 20 30
Empty 8«d Contact Tim« (minutari
•Asniouna, 26 pounds of carton par cubic foot.
III
I I III
In its revised cost estimates, TBS/ERCO sized contactor
and carbon volume on the average day in the maximum month of
the year. On average throughout the year this will give longer
contact times than the target amount. If 18 minute contact time
were desired, for example, this would yield about 22.5 minutes
of contact time on average over a year at most water systems
since the flow on the average day in the maximum month is ap-
proximately 25 percent higher than the average daily flow over
the year.
This means that longer regeneration cycles would also re-
sult. For example, if 60 day cycles were expected with 18 minute
contact time, then the average over the year in the example above
would actually be 75 days (25 percent longer corresponding to 25
percent increased contact time). These same relationships could
lead to regenera-t-ion cy-cles of 90 days and--longer if carbon fill
were sized to a system's design capacity.
An important factor in carbon fill sizing is the relation-
ship of contact time to the adsorptive capacity of carbon, about
which there is uncertainty. If the adsorptive capacity is con-
stant over the range expected, that is, each pound of carbon
ultimately adsorbs the same quantity of organics regardless of
flow rate, then shorter contact times will be dictated since
capital costs can be reduced. If the adsorptive capacity of
each pound of carbon decreases with decreased contact time,
then a trade-off will have to made between increased operating
and maintenance costs for carbon regeneration with short con-
tact times versus increased capital .costs for contactors and
carbon volume with long contact times.
-------
II-6
Pilot testing in Philadelphia over the past eight months
shows that for TOG reducing the contact time does not reduce the
adsorptive capacity of each pound of carbon for contact times
ranging from 7.5 to 30 minutes. This information would support
reducing capital expenditures for carbon and contactor volume.
Obviously the data generated by Philadelphia is preliminary and
more pilot testing is necessary. However, it does suggest, at
least for TOG in this case, that adsorptive capacity is inde-
pendent of contact time and that capital costs can be reduced
by designing for less contact time within some range.
Contactor Design
Once the total carbon and contactor volume is determined,
then decisions have to be made as to the size of individual
contactor units. Typically, the larger the individual con-
tactor units, the lower the overall contactor system capital
cost. For an above-ground contactor, the cost of items such
as manufactured equipment, valves and instrumentation are
relatively fixed and' increasing contactor diameter or depth
does not appreciably increase cost. Other items such as con-
crete, steel, labor and pipe do increase with larger contactor
diameter or depth, but not proportionately. For example, in
the Process Design Manual for Carbon Adsorption^ doubling
the effective volume per contactor from 10 to 20,000 cubic
feet only increases costs per contactor by 28 percent. Table
II-l demonstrates the cost savings, for treating 200 MGD by
using fewer contactor units of a larger diameter. The costs
are based on upflow counter current units with a combined
surface area of 23,205 square feet. As the figure shows, by
doubling the contactor diameter from 15 to 30 feet the con-
tactor system cost would drop from $20 to $8 million. Culp-
Wesner-Culp has reported that increasing bed depth of in-
ground contactors (which are similar to rapid sand filters)
by 67 percent from 5.0 to 8.3 feet results in an increase of
costs of less than 5 percent.2
Process Design Manual for Carbon Adsorption, United
States Environmental Protection Agency, Technology Transfer,
October 1973, Figure 5-1, page 5-4.
2
Estimating Costs as a Function of Size and Treatment
Efficiency (Draft Report),United States Environmental
Protection Agency, May 1978, Tables 21 and 22, for 28,000
square feet surface area.
-------
II-7
Table II-l
CONTACTOR COSTS
FOR A,200 MGD SYSTEM^
Diameter
(feet)
15
20
30
Carbon
Depth
10
10
10
Contactor
System
Cost
$20,120,119
$13,672,074
$ 8,129,341
Process Design Manual for Carbon Ad-
sorption, U.S. Environmental Protection
Agency, October 1973, Figure 5-1, p. 5-4.
Note: Costs based upon upflow counter-
current packed bed above-ground
contactors.
In the economic analysis supporting the proposed organics
regulations, above-ground contactors from 15 to 30 feet in
diameter were chosen. Contactors greater than 12 feet in
diameter would need to be constructed on site because of
highway restrictions in the transport of large items. For
large contactors a choice of steel versus concrete would have
to be made since on-site fabrication would be necessary.
There are a number of,types of above-ground carbon con-
tactors including:
• Upflow packed beds
• Upflow expanded beds
• Downflow packed beds in
gravity and pressure units.^
Flow rates in these types of above-ground contactors can
range from 2 to 10 gallons per minute per square foot of surface
area while depths range from 10 to 30 feet. Diameters can range
up to 30 feet.
Process Design Manual for Carbon Adsorption, United States
Environmental Protection Agency, Technology Transfer, October
1973.
-------
II-8
Upflow beds have an advantage over downflow beds in the
efficiency of carbon use because they can more closely approach
continuous counter-current contact operations which means the
most recently regenerated carbon is first in contact with the
water stream to be treated. Upflow beds may be designed to
allow addition of fresh carbon and the withdrawal of spent
carbon while the column remains in operation. Upflow packed
beds require a high clarity influent (usually a turbidity less
than 2.5 JTU). However, carbon fines in the effluent of up-
flow units can be a problem.
Downflow contactors can be used for both adsorption of
organics and for filtration. Provision must be made periodi-
cally to thoroughly wash downflow beds to relieve the pressure
drop caused by the accumulation of any suspended solids. Downflow
beds require a false bottom and support system, backwash facili-
ties, and controls similar to those used in rapid-sand filters.
Upflow units may be flushed through a simple well screen inlet-
outlet system. Often backwashing of upflow packed carbon con-
tactors, which are preceded by filtration, merely consists of
increasing upflow flow rates.
The alternative to the use of above-ground circular con-
tactors is the use of in-ground units similar to rapid sand
filters. These units would have to be designed to facilitate
the removal of carbon from the bed and the return of carbon to
the bed. This might present some problems initially since the
design of rapid sand filters does not usually include automatic
removal of the filter media. There is conflicting evidence on
the relative costs of such filter-type beds and above-ground
contactors. Some preliminary figures from a research study
by Culp-Wesner-Culp suggests that the filter-type units are
less expensive. However, estimates prepared by New Orleans
suggest that they would be much more expensive. No figures
are available for comparable design assumptions for the two
contactor types.
System Versus Plant Basis for Design
Many large cities which responded to the proposed regulation
are served by more than one water treatment plant. GAG treatment
installed at two smaller treatment plants is likely to be more
expensive than at one larger treatment plant for a number of
reasons.
• Capital items such as contactors and furnaces
cost less on a unit basis as they increase
in size.
-------
II-9
• Modifying two smaller treatment plants
to install adsorption following filtration
would be more expensive than modifying one
large plant. For instance one large pumping
station is likely to be less expensive than
two smaller ones of the same total capacity.
• Labor costs would increase since two smaller
furnaces at separate locations would require
more personnel than at one centralized furnace.
• Fuel costs might increase since smaller fur-
naces might require more fuel per pound of
carbon regenerated than larger furnaces.
Capital items such as initial carbon fill and operating
and maintenance expenses such as carbon loss upon regeneration
likely will not change appreciably within broad size categories.
There might be some discounting of carbon as the quantities pur-
chased increase, but a utility could purchase carbon for both
plants at the same time and take advantage of whatever volume
discount might be available. Carbon loss upon regeneration also
would not change significantly because furnaces over broad size
ranges can achieve the same performance.
Although the costs for each size category assumed that the
size category was served by one plant, this underestimation ,of
costs is counterbalanced to the extent that some systems will
not have to treat all their water. A number of systems are
served by dual sources such as surface and groundwater, or by
different surface sources one of which might not need treatment.
The revised TBS/ERCO cost estimates presented later in this
chapter now include two provisions which relate .to these cost
issues and which were not present in the earlier estimates. One
is an allowance for contingencies on all engineered items. The
other is an allowance of from zero to 25 percent add-on for site
specific costs in excess of the standard estimates.
CAPITAL COST ISSUES
A number of capital cost issues relating to GAC treat-
ment are discussed individually in the sections below.
-------
11-10
Contactors
The two main issues raised by industry concerning con-
tactors revolve around the amount of contactor volume necessary
and the costs of individual contactor units. As described pre-
viously, the contactor volume calculated in the economic anal-
ysis supporting the regulation was based on maintaining a 9 or
18 minute contact time at average daily flow. As flow varies
over a year, the amount of contact time would vary correspond-
ingly, and the unit would be able to process any quantity of
water up to the hydraulic capacity of the plant.
The unit costs for contactors were developed from the
Process Design Manual for Carbon Adsorption^ and included the
contactor unit itself, piping, valves, storage tanks, building
costs, and instrumentation. The units were above ground and
were assumed to be site fabricated and made of either steel
or concrete. They would be designed to follow filtration.
Backwashing capability is provided, although in the case of
packed upflow units, contactor backwashing can consist simply
of increasing the flow rates from 5 to 6 gpm/ft^ to 10 to 12
gpm/ft^ for 10 to 15 minutes.
The largest discrepancy in contactor costs occurred when
one water system developed costs for a contactor system based
upon the costs of a rapid sand filtration system, and compared
this to above-ground contactor units. In that case, New Orleans
used a cost of $500 per square foot of filter area and assumed
a carbon depth of 5 feet, yielding a cost of $100 per cubic
foot. The assumption of the amount of surface area and depth
per filter is critical because filters of larger volumes can
take advantage of economies of scale as described earlier^ As
an example, Culp-Wesner-Culp in a draft of a recent study5
had costs of $40.83 per cubic foot of bed capacity, assuming
a depth of 5 feet. If the depth were increased to 8.3 feet,
the cost dropped to -$25.36. Since the contact time dictates
the necessary volume, it appears that to obtain the necessary
volume, larger, deeper filters are.preferable. At an 18 minute
contact time TBS/ERCO estimates for above-ground contactors of
a ten foot depth ranged from $29 to $40 per cubic foot depend-
ing on the diameter of the contactor.
4
Process Design Manual for Carbon Adsorption, United States
Environmental Protection Agency, Technology Transfer, October
1973.
Treatment Costs as a Function of Size and Treatment Ef-
ficiency, United States Environmental Protection Agency,
May 1978.
-------
11-11
Closer agreement between industry and TBS/ERCO estimates
was obtained when costs were developed lor similar contactor
systems. For instance, Indianapolis developed cost estimates
for contactors for three plants with t-he result that unit costs
for one plant were lower than those published by TBS/ERCO and
for two were higher. Accordingly, no changes were made to the
unit costs used in the TBS/ERCO analysis—the only changes in
this area reflect the shift in the contactor sizing assumptions.
More detailed information on the comparison of the TBS/ERCO
costs with New Orleans, Indianapolis and Louisville is included
in Appendix A.
Modification to Hydraulics
The modification to hydraulics cost in the economic anal-
ysis was intended to represent the costs of pumping water to
and from the contactors and to cover any other site specific
costs such as the purchase of additional land to install the
carbon adsorption and regeneration operations. The costs were
intended to represent an average of what a number of systems
would incur. In estimating these costs, it was recognized that
there was a significant amount of uncertainty involved. A few
plants might need little additional pumping, while others would
need not only additional pumping but would incur significant
expenses in trying to locate the carbon adsorption and regen-
eration operations.
A number of responses indicate that these site specific
costs can be substantial. Indianapolis, for instance, has
estimated site specific costs of $8.95 million, which includes
suction wells and a pumping station and chlorine contact basins.
Applying the methodology in the TBS/ERCO economic analysis would
result in a cost estimate in this area of $3.3 million. The dif-
ference is accounted for by higher pumping costs and the addition
of 'chlorine contact basins with a capacity of 8 million gallons.
The TBS/ERCO economic analysis assumed that there would be some
disinfection ahead of carbon adsorption and that there was exist-
ing storage presently following filtration which could be used
to add any final dosages of chlorine. It now appears that as-
sumption may be optimistic for many systems.
The resolution to this issue of site specific costs in
revising the TBS/ERCO costs is to leave the allowance for mod-
ifications to hydraulics at its previous level and to include
a separate line item as an add-on to the total project cost
estimate for site specific costs in excess of the allowance.
On the basis of a limited sample of systems it appears that
these site specific costs could increase the "modifications
-------
11-12
to hydraulics" cost to three or four times the standard esti-
mate in extreme cases. That amounts to an increase in total
project cost, including contingencies and fees, of up to 25
percent in such cases. For the standard estimates a range
of zero to 25 percent add-on to total costs is now being
included for these factors.
GAG Costs
The initial capital cost of the GAG itself has not been
an area of much difference between the TBS/ERGO and industry
estimates. There are three assumptions needed to determine
GAG costs. First, the volume (in cubic feet) of GAG needed to
provide the desired contact must be established, as discussed
above.
The second assumption relates to carbon density, the num-
ber of pounds per*cubic foot of GAG. GAG made by different
manufacturers from different raw materials (various types of
coal, lignite, etc.) have different densities. According to
TBS/ERGO research the range is from approximately 23 to 30 Ibs
per cu ft. A review of the carbons most likely to be used for
drinking water treatment led to the original assumption of 26
pounds per cubic foot. Recent industry estimates have used
up to 30 Ibs per cubic foot, the highest end of the range of
densities. Although some carbons do have this much density,
on average the figure would be lower and 26 Ibs per cubic foot
remains, in the view of TBS and ERGO, an appropriate average
value for use on a national basis. If particular utilities
intend to use carbon of a certain type from one manufacturer,
then the use of the exact density, adsorptive properties, and
other specific characteristics of that carbon is appropriate
in making individual utility cost estimates. Since many dif-
fer ent--_carbon.s will be used -nationally, however, an average
value is appropriate for national estimates.
The third assumption is the purchase price per pound of
GAG. The original TBS/ERGO estimate was an average 1976 cost
of 45 cents/Ib. Cost increases over two years have brought
this to an average of approximately 53 cents in 1978. This is
the basis of the revised TBS/ERGO costs stated in 1978 dollars.
Various types of GAG have a wide range of prices, from
about 32 cents to 65 cents/lb. The choice of a suitable car-
bon for water treatment will be based on many factors other
than minimizing initial purchase costs, hence 'the purchases
will not all be at the low end of the range but at various
prices within this range. Several industry estimates have
-------
11-13
used 55 cents/lb as their estimated price, an estimate nearly
identical to the revised 1978 price of 53 cents being used by
TBS/ERGO.
Regeneration Furnaces
TBS/ERCO and industry cost estimates for regeneration
furnaces have shown considerable differences. Both sets of
estimates have been based primarily on multi-hearth furnaces
since these utilize a long established and proven technology
and since they generally have the highest capital cost of the
four commercially available furnace types, thereby stating
costs conservatively at the high end of the range. Rotary
kilns, electric infrared furnaces and fluidized bed furnaces
have all been used effectively for regenerating carbon, are
now generally available to the U.S. market and are priced
below—in some cases well below—the costs of comparable
multi-hearth furnaces. Nonetheless, the following discus-
sion focuses on multi-hearth furnaces only.
In its original estimates, TBS/ERCO sized furnaces to
accommodate the amount of carbon that would need to be re-
generated daily if a plant operated at design capacity.
While this provided a margin of safety since plants would not
be operating at design capacity for any appreciable length of
time, industry comments generally favored redundant furnaces.
It was decided to size the furnaces for future analyses such
that the amount of carbon that would need to be regenerated
on the average day in the maximum month could be accommodated
with the largest furnace out of service. This condition, it
is felt, should provide adequate back-up furnace capability
to allow for scheduled maintenance and repair and forced
outages. Since the national analysis assumes uniform organic
loading in the water, the spare furnace also provides extra
capacity for times when there is higher organic loading.
Table II-2 shows the percent utilization of the furnaces
under the initial and revised assumptions. Under the revised
assumptions the annual utilization rates would range from 28
to 53 percent. Given the low utilization rates and the high
capital costs of furnaces, many systems might explore alter-
natives such as stocking additional buffer carbon or operating
a single furnace and relying on a regional regeneration facil-
ity when the single furnace is out of service.
The cost estimating methodology for furnaces has been
identical for EPA's contractors and for the industry. Both
have first estimated daily regeneration requirements in terms
of Ibs/day, then determined the number of furnaces of a partic-
ular size needed to handle this requirement and finally used
-------
11-14
Contact Time
9 minutes
18 minutes
Table 11-2
FURNACE UTILIZATION
Population Category
75,000-100,000 100,000-1,000,000 over 1,000,000
PERCENT UTILIZATION UNDER REVISED ASSUMPTION*
49.8
28.6
43.0
37.5
44.9
52.7
PERCENT UTILIZATION UNDER ORIGINAL ASSUMPTION2
9 minutes
18 minutes
65.0
57.2
60.1
57.7
53.5
69.7
POUNDS OF DAILY FURNACE CAPACITY UNDER REVISED ASSUMPTION
9 minutes
18 minutes
12,210
42,460
42,460
97,240
207,140
352,440
POUNDS OF DAILY FURNACE CAPACITY UNDER,ORIGINAL
ASSUMPTION
9 mi nutes
18 minutes
9,350
21,230
30,360
63,250
173,800
266,750
POUNDS OF CARBON REGENERATED ON AVERAGE DAY
9 minutes
18 minutes
6,075
12,150
18,250
36,500
92,950
185,925
1
Carbon regeneration furnaces sized to accommodate carbon regenera-
tion needs at the average day in maximum month with the largest fur-
nace out of service.
Carbon regeneration furnaces sized to accommodate cabon regenera-
! tion needs at design capacity.
-------
11-15
an estimated cost per furnace to calculate the total. Each
of these three steps will be discussed below.
Calculating daily regeneration requirements is the first
step. It requires data on the pounds of GAG in place and the
frequency of regeneration. Just dividing one of these figures
into the other, is too simplistic. As the water plant's pro-
duction varies in peaks and valleys, so will the daily amount
of GAG to be regenerated. Furnaces must be capable of handling
requirements in the peak month, so the GAG volume and regenera-
tion frequency figures used must relate to the peak month.
The second step—determining the number of furnaces—is
based upon the daily regeneration requirement, but requires an
additional input, namely, the furnace loading rate stated in
terms of pounds per day of GAG per sq ft of effective hearth
area. This figure has ranged broadly from 45 Ibs/sq.ft. for
some wastewater experience to over 150 Ibs. for some drinking
water experiments. It is generally agreed that the organics. -
found in drinking water are more easily removed from GAG pores
than organics found in industrial and municipal wastes. There-
fore the loading rate for drinking water applications should
be higher. The amount of this increase, however, has not yet
been proven definitively. Estimates have ranged from 45 to
nearly 200 Ibs per sq ft. Two major furnace manufacturers are
using 110 to 120 lbs/ft2 in sizing carbon regeneration fur-
naces for drinking water treatment, a figure close to the 110
lbs/ft2 used by TBS/ERGO in its cost estimates.
Many industry estimates have cited statistics from a re-
cent article in Chemical Engineering^. According to the
authors of this article, reactivation rates shown in it (Table
1) are based on 60 to 80 Ibs/sq ft with a 10 to 25 percent
downtime reducing the effective rate to 45 to 74 Ibs/sq ft/day.
The result is that to achieve a given reactivation ..rate (such
as 60,000 Ibs per day) this article suggests a furnace up to
2.4 times as large as that assumed by TBS/ERGO. The authors
of the article have also incorporated maintenance downtime in
their sizing. Some industry estimates based on this article
have also added downtime to furnaces, thus double counting
this maintenance time and sizing furnaces more than three
times as large as those sized by TBS/ERGO.
R. H. Zanitsch and R. T. Lynch, "Selecting a Thermal Regen-
eration System for Activated Carbon," Chemical Engineering,
January 2, 1978.
-------
11-16
Having determined the number and size of the furnaces
needed for reactivation, the third and final step is to esti-
mate the cost per furnace. TBS/ERCO interviewed major furnace
manufacturers and obtained copies of current retail furnace
prices which include installation, instrumentation, and as-
sociated carbon handling equipment such as a feed screw,
quench tank, afterburner and a scrubber.
Industry estimates have primarily been drawn from the
magazine article cited earlier. The use of cost figures per
furnace from this article has the effect of overstating costs
by approximately a factor of 3, as discussed above. Additional
contacts by industry representatives with furnace manufacturers
will likely lead to revised information on furnace prices and
a narrower range of differences between the estimates.
Contingencies
Many industry estimates have included contingencies as a
percentage of total construction costs. Such provisions are
common practice in making preliminary cost estimates for large
construction projects.
The previous TBS/ERCO estimates did not include contin-
gencies in national estimates. However, the revised estimates
do incorporate contingencies on certain items to reflect cost
uncertainties. In order to provide conservative cost estimates,
contingencies of 15 percent are now being added by TBS/ERCO to
the capital costs of contactors, modifications to hydraulics,
carbon transfer equipment, buffer carbons storage facilities,
and furnaces. The costs of items such as carbon are well
defined and do not warrant an allowance for contingencies.
Engineering, Legal and Financing Fees
.Fees for engineering, pilot testing, legal services, other
services associated with construction, and financing would be
incurred by utilities adding GAC treatment. The published
TBS/ERCO estimates included only engineering fees at 6 1/2 to
8 percent of construction costs depending upon the size of the
system and, therefore, of the construction project. Based on
added information from industry comments, these fees have now
been increased to 15 percent of construction costs in the re-
vised estimates.
-------
11-17
OPERATING AND MAINTENANCE COST ISSUES
The operating and maintenance costs developed as part
of the economic analysis of the proposed regulations received
much less comment than the capital costs. The basic areas of
disagreement which had the largest impacts on costs were fuel
usage and carbon loss upon regeneration.
A number of industry operating and maintenance cost esti-
mates for carbon reactivation were developed from an article in
Chemical Engineering.7 For a 60,000 Ib per day reactivation
rate the costs of fuel, power, steam, and make-up carbon were
reported to be $1,570,000 yearly or $.072 per pound of carbon
regenerated. The carbon regenerated in this article was used
to treat industrial wastes. This compares to a cost of about
$.058 per Ib in the analysis supporting EPA's proposed regula-
tion. The higher cost is primarily attributable to higher fuel
usage—the article assumed 8,000 Btu per pound are required
while TBS/ERCO's economic analysis used a range of 4300 to
3700 Btus with the larger fuel usage in the smallest water
system size category. The TBS/ERCO values were developed from
an article in Chemical Engineering Progress.8 At $4 per mil-
lion Btus this difference accounts for approximately $0.13 per
pound. Operating data from Lake Tahoe show an average fuel
usage of 2900 Btus per pound with a range of 1820 to 4510 Btus
per pound on a number of runs.9 Estimates by Neptune Nichols
for,a 60,000 Ib. per day furnace used to regenerate carbon for
drinking water treatment are approximately 4600 Btu of which
35 percent is for afterburning. With this information it was
decided to use a figure of 5000 Btus per pound of carbon re-
generated as a sufficiently conservative assumption of fuel
usage in the regeneration of carbon used in drinking water
treatment.
There was agreement on a carbon loss rate of 7 percent
upon regeneration in the economic analysis and the article in
Chemical Engineering article. This is significant because
make-up carbon accounts for approximately 40 percent of total
O&M costs associated with carbon regeneration in the Chemical
Engineering article and up to 50 percent in the revised TBS/
ERGO O&M costs.
7
R. H. Zanitsch and R. T. Lynch, Ibid.
Q
R. A. Hutchins, "Thermal Regeneration Costs," Chemical
Engineering Progress, May 1975.
9
Process Design Manual for Carbon Adsorption, U.S.E.P.A.,
Technology Transfer, October 1973.
-------
11-18
Some systems, e.g., Louisville and Indianapolis assumed
a carbon loss rate of 10 percent. This difference is sig-
nificant at the volumes of regeneration that would be necessary.
At a regeneration rate of 60,000 Ibs per day, for example, a
loss of 10 percent vs. 7 percent amounts to a difference of 1800
Ibs per day. At $.55 per pound this comes to $990 per day or
$360,000 yearly. Carbon losses at Lake Tahoe varied from 2.5
to 8.6 percent with an average of 5.8 percent. On this basis
it was decided that 7 percent carbon loss upon regeneration
represented a mid point in the range of losses that might be
expected. The loss rate includes any losses due to abrasion
as the carbon is transported to and from the contactors.
ACCOUNTING AND FINANCING PRACTICES
Though not a major topic of discussion in comments on the
costs of the organics regulations, the accounting and financing
practices of the water utility industry have been the source of
some differences in cost estimates. Two issues have been dis-
cussed most. The depreciation methods and the appropriate de-
preciation lifetimes of GAC systems are the first issue. The
second is the allowance for funds during construction (AFDC).
The original TBS/ERCO estimates assumed a straight-line
depreciation over a 40 year period, for all new capital items.
This basis is consistent with general utility practice and
appropriate for a newly-installed GAC system, though a few
explanations are appropriate. Most regeneration furnaces,
particularly multi-hearth furnaces, will need major capital
repairs such as new refractory linings and hearths during a
40 year life. These costs have been included in 0/M costs
at an annual cost of 5 percent of the initial capital cost
in the TBS/ERCO cost estimates. The contactor units and in-
-plant-piping should -be-" similar to present treatment plant
construction and warrant the same depreciation treatment.
Carrying costs of a new capital project during construc-
tion are handled differently by various utilities within the
water industry. Many publicly-owned municipal utilities base
their rates on the principle of recovering all current costs,
including those of financing capital programs for future use.
This approach is commonly referred to as including construc-
tion work in progress (CWIP). expenses in the rate base. It is
a 'common practice for municipal systems and is being allowed
for investor-owned•utilities in some states, at least in the
electric utility industry.
-------
11-19
Other municipal utilities, especially investor-owned sys-
tems, capitalize the cost of interest during construction, and
may add this amount to their initial borrowing. This practice
may result in an increase of as much as 20 percent in the cap-
ital funding of a construction project and has caused confusion
in comparing project capital costs.
The EPA cost estimates have been computed on the former
basis, assuming current funding of construction carrying costs
through rate increases. The assumption, however, has no effect
upon construction costs or operating costs, and little, if any,
effect upon long-term water rates. It is a financing assump-
tion and affects only the timing, amount and possibly the ac-
cessibiliy of capital financing.
REVISED TBS/ERGO GAC TREATMENT COST ESTIMATES
This section presents the results of the revisions to
the cost items discussed in detail above. The following table,
Table II-3, summarizes the physical assumptions for contactors
and furnaces which have been revised for three system sizes.
These figures reflect designs which are capable of accommodating
flows at the system design capacities and which yield their
desired performance (nine or eighteen minute contact time) at
a flow equal to the average day flow in the maximum month.
-------
11-20
Table I I -3
CONTACTOR AND FURNACE ASSUMPTION FOR AVERAGE SYSTEM
Population Category
Contactors
Number
Diamater (ft)
Depth of Carbon (ft)
Effective Volume
Per Contactor (ft3)
Furnaces
Number
Effective Area
Per Furnace (ft^)
Operating Labor
Requirements (persons]
Fuel Cost Per Pound
Regenerated
Cost of Make-Up Carbon
Per Pound Regenerated
Carbon
Initial Fill (Ibs)*
Daily Regeneration**
Volume (Ibs)
75,000-100,000
9 min 18 min
10 11
15 20
10 10
1,766 3,140
3 2
37 193
6 6
$.024 $.024
$.037 $0.37
434,000 864,000
6,075 12,150
100,000-1
9 min
11
25
10
4,906
2
193
6
$.024
$.037
1,303,000 2
18,250
Based on peak month average daily flow.
**
Based on annual average
daily flow.
,000,000
18 min
15
30
10
7,065
2
442
6
$.024
$.037
,606,000 6
35,500
I
Over
9 min
36
30
10
7,065
3
29845
10193
12
$.024
$.037
,516,000
92,950
one million
13 min
71
30
10
7,065
3
1,068
12
5.024
5.037
13,032,000
185,925
Capital Costs of GAG
The capital cost estimates for GAG treatment have been
revised significantly on the basis of this recent review and
discussion with the industry. The revised cost estimates are
shown in Tables II-4 and II-5 on the following page, for contact
times of 18 and 9 minutes, respectively.
Figure II-2 graphically depicts cost per thousand gallons
as a function of the average flow in the maximum month. The
curve can be used to more easily determine what the project
cost total is for plants in between the various categories
costed in Tables II-4 and II-5. It also demonstrates the
sharply increasing unit costs at smaller plants. The figures
shown in the graph are the project cost totals and do not in-
clude any 'allowances for unusual site specific costs.
-------
11-21
Table 11-4
REVISED CAPITAL COSTS
18 Minute Contact Time
€0 D«y Regeneration Cycle
Population Served
75,000-
100.000
Average-Production in 1981 (ngd)
Capacity 1n 1981 (ragd)
Assumed Average Day 1n Max. Month (mgd)
Population Served
100,0000-
1.000.000
50.4
75.6
60.0
263,200
Over
1 million*
256.3
359.3
300.0
1,193,000
Granular Activated Carbon Initial Fill
Contactors
Furnaces
Modification to Hydraulics
Buffer Stock
Subtotal
Contingencies*
Engineering, legal, and Financial Fees
Project Cost Total
Range of Site Specific Costs
Final Total (millions)
$ 460,000
2,032,000
2,333,000
354,000
32,000
$ 1,381,000
3,695,000
3,499,000
1,055,000
97,000
$ 6,907,000
17,490,000
8,398,000
3,535,000
483,000
$5,711,000 $ 9,727,000 $36,813,000
783,000
974,000
$7,468,000
02 to 251
1,237,000
1,645,000
$12,609,000
OX to 251
4,413,000
6,134,000
$47,410,000
OS to 251
$7.5 to 9.3 $12.6 to 15.3 $47.4 to 59.3
15 percent applied to contactors, furnaces, and modification to hydraulics.
Production levels for systems expected to be affected by regulation.
Table 11-5
REVISED CAPITAL COSTS
9 Minute Contact Time
60 Day Regeneration Cycle
Population Served
75,000-
100JJOO
Average Production In 1981 (mgd)
Capacity in 1981 (mgd)
Assumed Average Day in Max. Month (mgd)
Population Served
100,0000-
1.000.000
50.4
75.6
60.0
263,200
Over
1 mil 11 on"
256.3
359.3
300.0
1,193,000
Granular Activated Carbon Initial Fill
Contactors
Furnaces
Modification to Hydraulics
Suffer Stock
Subtotal
Contingencies*
Engineering, Legal, and Financial Fees
Project Cost Total
Range of Site Specific Costs
Final Total (millions)
$ 230,000
1,540,000
1,400,000
854,000
16,000
$ 591,000
2,956,000
2,333,000
1,055,000
48,000
$ 3,453,000
8,745,000
6,065,000
3,535,000
242,000
$4,040,000 $ 7,083,000 $22,040,000
569,000
591,000
$5,300,000
OS to 25S
$5.3 to 6.6
952,000
1,205,000
$ 9,240,000
01 to 255
2,752,000
3,719,000
$28,511,000
OS to_25t
$9.2 to 11.6 $28.5 to 35.6
15 percent applied to contactors, furnaces, and modification to hydraulics.
Production levels for systems expected to be affected by regulation.
-------
11-22
Figure 11-2
CAPITAL COSTS
FOR NINE AND EIGHTEEN MINUTE CONTACT TIMES
400
Dollars
Per
Thousand
Gallons
300
200
100
0 100 200 300
Average Day Flow in Maximum Month (MGD)
T
B
S
-------
11-23
The project costs shown range from 50 to 70 percent higher
than the previous estimates for most of the categories at nine
and eighteen minute contact time. The variation in relative
increase from one system size to another is in part due to using
standard sizes for system components. This is especially true
for furnaces, where the increases between standard sizes are
signficant and where the design constraint has been imposed
that enough furnaces be installed that each system could meet
its peak month regeneration level with the largest furnace out
of service. As shown in Table II-3, this combination of factors
results in furnaces which would operate at less than 50 percent
utilization over the year.
The project cost subtotals shown in the tables above are
intended to cover all capital costs for GAG treatment at a
water system which has no unusual local problems or constraints.
The site specific additional costs which would be required in
New Orleans, for example, must be identified on a case by case
basis and added onto these standard cost estimates. In review-
ing the comments received by EPA such site specific costs ac-
counted for a range of additional costs from 0 to 25 percent of
the total project cost.
The major reasons for capital cost revisions to the indi-
vidual components in the project cost are:
• To add inflation from 1976, the date of the
earlier estimates, to 1978
• To increase the contactor and carbon volume
to provide the desired contact time at a flow
equal to that of the average day in the maxi-
mum month instead of the average day of the
year
• To provide for a larger number of furnaces
so that the maximum month daily regeneration
volume could be processed with the largest
furnace out of service
• To add an allowance for contingencies of
15 percent on certain major cost components,
namely contactors, furnaces and modifica-
tions to hydraulics
• To increase the design fee of 6 to 8 percent
in the earlier estimates to a level of 15
percent to include legal, financing and all
engineering design fees an'd the cost of pilot
testing.
-------
11-24
The design changes (e.g., increased carbon
and contactor volume and expanded furnace
capacity) account for approximately one-third
of the increased costs. The remainder of the
difference is about equally split between con-
tingencies, increased fees and accounting for
inflation.
Operating and Maintenance Costs
The review of O&M costs has resulted in only minor revisions
to the fuel costs in earlier published estimates. The revised
costs as shown in Tables II-6 and II-7 below have been inflated
to 1978 dollars and are only marginally higher than they were
previously.
Table I 1-6
REVISED 04M COSTS (1978 dollars)
•
50
Average Production in 1981
Capacity in 1981 (mgd)
Assumed Average Day in Max.
Population Served
Contactor Operating Costs
18 Minute Contact Time
Day Regeneration Cycle
75,000-
100,000
(mgd) 16.3
27.0
Month (mgd) 20.0
92,700
S 94,000
Population
100,0000-
1,000,000
50.4
75.6
60.0
263,200
$ 196,000
Served
Over
1 million
256.3
359.3
300.0
1,193,000
S 926,000
Regeneration Ooerating Costs
a. Labor 96,000
b. Maintenance Labor and Materials 117,000
c. Operating Supplies 12,000
d. Fuel Costs 107,000
e. Carbon Replacements 162,000
f. Insurance for Furnace 23,000
g. Laboratory Analyses 20,000
96 ,000
175,000
17,000
315,000
487,000
35,000
20,000
192,000
4.90,000
49,000
1,615,000 \
2,483,000
98,000
20,000
Total
$631,000 $1,342,000 $5,373,000
-------
11-25
Tab.le 1 1-7
REVISED 04M COSTS (1978 dollars)
50
Average Production in 1981
Capacity in 1981 (mgd)
Assumed Average Day in Max.
Population Served
Contactor Operating Costs
9 M1 nute- Contact Time
Day Regeneration Cycle
75,000-
100.000
(mgd) 16.8
27.0
Month (mgd) 20.0
92,700
S 94,000
Population
100,0000-
1,000,000
50.4
75.6
50.0
253,200
3196,000
Served
Over
1 mi 11 ion
256.3
359.8
300.0
1,193,000
S 925,000
Regeneration Ooerating Costs
a. Labor
b. Maintenance Labor- and
c. Operating Supplies
d. Fuel Costs
a. Carton Replacements
f. Insurance for Furnace
g. Laboratory Analyses
Total
96,000
Materials 79,000
3,000
54,000
31,000
16,000
20,000
3448,000
96 ,000
140,000
14,000
153,000
244,000
28,000
20,000
5896,000
192,000
367,000
37,000
308,000
1,242,000
74,000
20,000
53,566,000
-------
III. NATIONAL ECONOMIC IMPACT
OF THE REGULATION
The changes in unit cost estimates described in the pre-
ceding chapter also alter the national economic assessment of
the proposed regulation. This chapter presents the modified
estimate of national economic impact relating to both the
trihalomethane regulation and the treatment requirement
regulation.
The major revisions to the national economic assessment
reflect inflation to 1978 dollars and revised cost estimates
for granular activated carbon (GAG) treatment. Some of the
cost revisions such as inflation, contingencies and a greater
allowance for design and legal fees are applicable to the
other treatment methods considered as well. These revisions
have been incorporated at the same-levels estimated for GAG
costs into the cost estimates for treatment with ozone,
chlorine dioxide and chlorination with ammoniation.
There are several areas in which no changes were made from
the previous analysis. One such area is the expected compliance
status of water systems with the proposed regulations. The
national costs shown below are based on the same assumptions
which were previously published regarding the number of systems
which are likely to be affected by the regulation (86 by the
trihalomethane regulation and 50 by the treatment requirements,
of which 15 would be affected by both simultaneously). Table
1II-1 below shows the breakdown of the number of systems ex-
pected to be affected by each portion of the regulation.
-------
III-2
Table III-l
ESTIMATED NUMBER OF SYSTEMS
AFFECTED BIT THE REGULATION
Regulation
Total Number of
Systems Affected
Number of Systems
Cost- Impacted
Number of Systems
to Install GAC
Source: Economic
THM
Only
71
36
11
Analysis of
Organic Contaminants
Treatment
Requirement
Only Both Total
35 IS 121
35 15 36
35 15 61.
Proposed Regulations on
in Drinking Water, op. cit.,
p. III-l.
An area which has been slightly revised is the mix of
treatment strategies which is likely to be selected by water
systems to comply with these regulations. The previous analy-
sis was based on the best estimate that 11 systems would in-
stall GAC to meet the trihalomethane regulation alone. A more
extreme analysis was also presented based on 28 systems install-
ing GAC only for the THM regulation. That latter case has been
omitted from the present analysis as being unrealistic, since
only 36 systems are expected to be cost-impacted by the THM
regulation alone. As before, the analysis assumes that GAC
treatment would be installed by all 35 systems affected only
by the treatment requirement, and by all 15 systems affected
by both regulations.
The actual requirements of GAC systems in terms of con-
tact time and carbon regeneration frequency are, as noted in
the preceding chapter, uncertain at this time. The actual
requirements can be determined only through site specific pilot
testing and will vary from system to system. The current anal-
ysis is based upon the same assumptions in this area which
have been used in the previous economic analyses: that 9 min-
ute empty bed contact time with 60 day regeneration would be
appropriate for those systems installing GAC solely to meet
the trihalomethane regulation, and that from 9 to 18 minute
contact time with 60 day regeneration frequency would be
required for systems installing GAC to meet the treatment
requirement.
-------
III-3
Finally, the present analysis contains no reappraisal of
monitoring costs to meet the regulation, of the effects of the
regulations upon supplying industries, or of the cost of the
interim requirement that systems affected by the treatment re-
quirement replace their existing filter media with GAG unless
otherwise directed by their State.
NATIONAL COSTS OF THE REGULATIONS
The aggregate national cost of the regulations has been
evaluated in terms of capital expenditure requirements, opera-
tions and maintenance expenses and annual revenue requirements.
The national costs reflect averages of conditions at individual
systems across the country. In order to estimate these national
figures, average cost estimates for various components of the
treatment systems were used with a cost for site specific impacts
attributed to about one-half of the affected systems.
As noted in Chapter II, several system level factors could
actually make costs higher or lower than the average costs
used to characterize individual systems. These factors would
naturally affect the national aggregate figures as well. More
specifically, the following factors could increase system
level and national costs:
• Multiple plants
• Local or site specific costs of redesigning
an operating plant
Factors which could decrease national costs include:
- • -~ Multiple raw water sources where some plants
of a utility would not be out of compliance
• Refined engineering design which may improve
furnace utilization and better tailor other
system elements to local needs
• Less expensive furnace types (e.g., fluidized
bed)
• Operating practices which more strictly follow
the letter of the regulation and seek to meet
long run average standards rather than con-
tinuous maximum concentra-tion levels
-------
III-4
• Selection of filter media replacement rather
than the construction of contactors by some
systems.
The capital expenditure requirements of the regulations
are shown in Table III-2 below. The high estimate of these
costs is $831 million, of which about 15 percent is for the
trihalomethane regulation only, almost 60 percent is for the
treatment requirement only, and the remaining 25 percent is
for systems affected by both regulations. This estimate is
based upon two key assumptions:
• Approximately 30 percent of the systems affected
only by the THM regulation would utilize GAG
treatment.
• The GAG systems installed to meet the treatment
.requirement or both regulations are designed to
yield 18 minute empty bed contact time.
Table III-2
NATIONAL CAPITAL EXPENDITURES
THM AND TREATMENT REQUIREMENT REGULATIONS
(minions of 1978 dollars)
Systems affected by:
THM Regulation Only
Treatment Requirement Only
3oth Regulations
Total
Revised
Estimates''
$113
5348-496
$155-222
$616-331
Costs represent the following assumptions:
for the THM regulation, 11 systems use SAC;
for the treatment requirement and those systems
affected by both regulations, a 9 to 18 minute
GAC contact time is required.
Note: All cost estimates include a 10 percent
allowance for average sita specific costs
in excess of the unit cost estimates pre-
sented in Chaoter II.
-------
III-5
Lower national costs could result if some or all of the
GAG systems built to meet the treatment requirement were de-
signed at lower contact times. Specifically, if 9 minute con-
tact time were sufficient to yield 60 day carbon regeneration
cycles, the national capital cost would be approximately $616
million.
The annual total operations and maintenance (0/M) expenses
and revenue requirements are shown in Table III-3 below.
Table III-3
NATIONAL 0/M AND REVENUE REQUIREMENTS IN 1981
THM AND TREATMENT REQUIREMENT REGULATIONS
(millions of 1978 dollars)
Systems affected by:
—--—•—-Annual QSM Expenses
THM Regulation Only
Treatment Requirement Qnly
Both Regulations
Total
—Annual Revenue Requirements in
THM Regulation Only
Treatment Requirement Only
Both Regulations
Total
Revised
Estimates*
S 16
S 32-48
S 14-22
S 27
3 67-98
S 30-44
$124-169
Costs represent the following assumptions:
for the THM regulation, 11 systems use GAC;
for the treatment requirement and those systems
affected by both regulations, a 9 to 13 minute
GAC contact time is required.
Mote: All cost estimates include a 10 percent
allowance for average site specific costs
in excess of the unit cost estimates pre-
sented in Chapter II.
-------
III-6
The costs in the table refer to the same range of assump-
tions described for the capital expenditures projection. The
operations and maintenance expenses could range from $62 to
$86 million under these various assumptions.
The annual revenues required to cover the financing and
operating expenses for all 121 systems estimated to be affected
by the regulations are projected to be $124 to $169 million in
1981.
PER CAPITA AND CUSTOMER COSTS OF GAG
The local cost impacts which would be felt by residential
customers of water systems which install GAG treatment are
measured in terms of both the total per capita cost and the
annual average family's residential bill increase. These costs
vary significantly depending upon the size of the water system
(i.e., the population served), the design parameters (notably
contact time) and the presence or absence of site specific
additional costs. Representative cost impacts in 1981 for
various assumptions are shown in Table III-4 on the following
page. Those figures reflect costs only for systems using GAG
treatment; the costs for customers of systems using other
treatments to comply with the trihalomethane regulation would
be much lower.
The cost per capita is simply the total annual revenue re-
quirement for a water system divided by the population it
serves. This provides an upper bound in the possible cost of
such treatments to individual consumers if no costs were allo-
cated to non-residential customers. By this measure the util-
ization of GAG treatment in systems which have no special site
specific costs will result in -costs of approximately $5. to $11
per capita for systems with 9 minute contact time and $9 to $15
per capita for systems with 18 minute contact time. In those
special cases where site specific problems could increase cap-
ital costs by as much as 25 percent, the corresponding per
capita cost ranges would be $6 to $12 for 9 minute contact
time and $10 to $17 for 18 minute contact time.
Actually the increase in water rates will usually be less
than this per capita cost because some of the costs will be
borne by non-residential customers of each water system. The
other figures in Table III-4 are estimates of the likely in-
crease in annual water rates for an average.family of three
-------
III-7
assuming that non-residential customers pick up the same pro-
portion of GAG costs that they do of other system costs. On
this basis GAC treatments at systems which have no significant
site specific problems will result in annual increased water
bills of approximately $7 to $16 per family -assuming 9 minute
contact time and $11 to $23 per family assuming 18 minute
contact time. The annual cost per family in cities which do
have substantial site specific problems (at the level of 25
percent increased capital costs) would range from $8 to $19
in the 9 minute case and from $13 to $26 in the 18 minute
case.
Table III-4
ANNUAL PER CAPITA AND CUSTOMER COSTS OF GAC, 1981
INDIVIDUAL WATER SYSTEMS
(1978 dollars)
System Size (Population Served)
75,000-
100,000
—Annual Cost Per Capita*-
100,000-
1 Million
Standard cost, without sita
specific additional costs:
• 9 minute contact time
• 18 minute contact time
Cost with 25 percent extra capital
cost for site specific items:
• 9 minute contact time
• 18 minute contact time
$10.80
$15.30
$12.30
$17.40
$ 7.00
$10.00
$ 7.90
$11.30
Annual Residential Customer Bill Impact*
Standard cost, without site
specific additional costs:
t 9 minute contact time
* 18 minute contact time
Cost with 25 percent extra capital
cost for site specific itarns:
t 9 minute contact time
* 13 minute contact time
$15.20
$23.00
$13.50
$25.10
$10.50
$15.00
$11.90
$17.00
Over
1 Million
Assuming all costs allocated to residential customers.
#
For a family of three, assuming that non-residential cus-
tomers pick uo the same orooortion of GAC, costs that they
do other system costs.
$ 5.40
$ 3.50
$ 5.00
$ 9.50
$ 7.10
$11.40
$ 7.90
$12.70
-------
IV. FEASIBILITY OF FINANCING GAC TREATMENT
INTRODUCTION
The previous chapter presented the national economic im-
pacts of the proposed regulation on the water utility industry
including an estimate of the total amount of capital expendi-
tures required. Water utilities faced with the task of in-
stalling GAC treatment will need to go to the nation's capital
markets to finance these expenditures. This chapter examines
the ease or difficulty which individual water systems can be
expected to encounter when they seek to raise the requisite
funds on a system level to provide an indication of the in-
dustry's ability to finance the regulations on a national
level.
The analysis of the industry's probable ease of access to
financing is based upon evaluations of a sample of individual
water systems. The sample was analyzed using financial ratios
routinely used by investment bankers and credit analysts. Al-
though some of these systems may not actually be required to
install GAC treatment, this analysis assumed such would be the
case so as to assess each system's ability to raise the funds
that could be required by EPA's regulations. In addition, a
high GAC cost scenario was examined as well as a low cost case.
Each system was then categorized as to the ease or difficulty
it is expected to encounter raising the money under each cost
scenario.
Although the analysis is intended to be indicative of the
financial capability of utilities which might be affected by
these regulations, it is not intended to be definitive regard-
ing the specific utilities examined. No projections of future
capital requirements were developed. Also, additional infor-
mation which is generally available to credit analysts (e.g.,
management ability, concentration of industry, demographic
trends, etc.) was employed in the analysis only when it was
readily available from sources such as Moody's Credit Report.
A brief description of the methodology follows below. More
detail may be found in Appendix C.
METHODOLOGY
To evaluate the water utility industry's ease of access
to the capital markets, a sample of water systems was analyzed
-------
IV-2
The sample consisted of systems serving populations of 75,000
or more which could potentially be affected by the regulation
by virtue of their location and raw water quality. The sys-
tems finally analyzed are 27 of the 30 water utilities which
were listed as potentially affected by the regulation in EPA's
January 25, 1978 press release- announcing the proposed organ-
ics regulation. Three cities that were included in the January
list were not examined here due to the lack of information ob-
tainable during the time available for analysis. It is impor-
tant to note that although all systems examined are assumed to
use GAC for the purposes of this analysis, such widespread use
of GAC by these cities would probably not be necessary.
To determine the financial impact of the organics regula-
tions it is necessary to determine the impact on the utility's
ease of access to the capital markets. An acceptable proxy
for the utility's ease of financing these expenditures is the
bond rating that would be assigned to the supporting issue by
rating services. These ratings are often used by investors
as a source of information about the quality and relative risk
of the bonds being sold.
TBS examined the1 same financial ratios which the invest-
ment community uses as the quantitative input to their bond
rating analysis. The form of the ratios differs slightly
depending on the form of ownership of the utility (municipal
or investor-owned) and the form of the financial instrument
(revenue bond or general obligation bond). The most important
ratios relate to the ease or probability of timely payment of
debt obligations. Other ratios reveal information concerning
the operating and financial characteristics of the utility.
The specific ratios are detailed in Appendix C.
The analysis assumes that utilities will continue to
use whatever forms of financing they have used most recently..
Municipally-owned systems would generally finance with either
water revenue bonds or with "self-supporting" general obliga-
tion bonds which are financed, first, against the revenues of
the water system and, second, against the full faith and credit
of the municipality. • Investor-owned systems are assumed to
finance with a mix of debt and common equity which is close to
their current capital structure (for this analysis no preferred
stock financing has been assumed).
The ratios were first computed using the most recent avail-
able financial information for the utility (1976 data in most
cases). The same ratios were then computed showing the impact
-------
IV-3
of low-cost and high-cost GAG treatment scenarios. The two
scenarios were chosen to reflect the range of costs which might
be experienced by individual systems. In this- analysis one
range of unit costs was used for all systems, with the smaller
systems (i.e., those serving 75,000 to 100,000 people) having
unit costs closest to the high end of the-range and the largest
systems having unit costs close to or even below the low end of
the range. The specific cost ranges for each of the 27 cities
are listed in Appendix C.
It is important to note that the methodology used to com-
pute the ratios in the impacted cases assumes that rate in-
creases would be implemented as required to exactly cover the
direct costs of the additional O&M expense and the additional
financing costs associated with the GAC treatment. If these
increases are not obtainable, then the impact of financing GAC
treatment addition (or any other capital expenditure) would
be severe in most cases. If an even greater increase in rev-
enues were obtained, for example to maintain interest coverage
requirements, the financial impacts would be ameliorated. Sys-
tems faced with very large increases in revenue requirements
and which must receive regulatory approval before the instal-
lation of new rates, will have more difficulty financing the
regulations than those with small revenue increments and/or
no regulatory impediments. This issue, however, could not be
explicitly addressed on a system level in this report.
The ease of access to the capital markets of the utilities
was evaluated on the basis of the ratio analysis described above,
More specifically, the judgment of the capability of financing
was linked to the following criteria (which are listed in order
of priority) :
• Debt Service and Interest Coverage ratios
under the two scenarios compared to the
current level of the ratio, industry aver-
ages, and the bond covenant requirements
of the utility
• The current financial stength of the sys-
tem as evidenced by its bon-d rating
• Operating and debt ratios compared to his-
torical levels and industry averages.
Other factors such as information on the municipality's finan-
cial strength, the magnitude of required revenue increases and
planned capital expenditures also entered into the judgment
where the information was available and pertinent.
-------
IV-4
Using these criteria, it is possible to identify utilities
which may encounter difficulties when seeking to finance GAG
treatment addition in the capital markets. Utilities experi-
encing such difficulties would be flagged if their current fi-
nancial strength were weak or if their financial ratios dropped
to levels which are low compared to the industry. Moreover,
those utilities whose ratios may deteriorate to levels incon-
sistent with current strong bond ratings are also called out.
Based on this judgment of ease of access to capital markets,
the systems were divided into the following categories:
• Those which would be able to finance GAC
treatment addition with little or no
difficulty.
• Those which would be able to finance the
treatment but with some effect on their
financial standing. In order to maintain
their financial strength, these systems may
require either revenue increases greater
than those necessary to cover the direct
cost of GAC treatment or some other action
tailored to their needs.
• Those which would be expected to face major
financial barriers under present conditions.
LIMITS TO THE ANALYSIS
It is important to call attention to the limits to the TBS
analysis discussed in this chapter. For example, the judgmental
nature of the final step in the analysis—the actual determina-
tion of the financing capabilities of the utilities—is acknowl-
edged. It is possible that others•examining the same ratios
calculated by TBS would reach slightly different conclusions
regarding the financing ability of the utilities.
TBS also readily points out that these interpretations are
not meant to be definitive for any particular city included in
the sample. Those interested in a definitive analysis of any
specific city would conduct a more extensive analysis than that
presented here. Close interaction with the actual municipality
and the utility would be required before any absolute judgment
of the credit worthiness of the entity can be made. For ex-
ample, information concerning the current level and history'of
rates, management ability and the structure of outstanding
debt would be required for completeness. Also, the analysis
-------
IV-5
does not specifically address the likelihood of the system
receiving a. rate increase sufficient to cover the direct costs
of GAG. This lack of precision is acknowleged by the form of
the findings presented here; the financing capability of the
utilities is described in general groupings of ability rather
than by more specific measures such as expected bond ratings.
The judgments made in our analysis are, however, indicative of
the general financing capability and impacts of financing on
utilities which would be required to add GAG treatment.
FINDINGS
The TBS analysis described above led to the following
major findings:
• Presuming that rates are increased to cover
the *annualized capital costs and the 0/M
expenses associated with GAG treatment ad-
dition, nearly all of the 27 water utilities
examined would be able to finance GAG treat-
ment under either the low cost or high cost
scenario, some with little or no difficulty
and others with some effects on their
financial standing.
• Under the low cost scenario, 21 of the 27
utilities would be expected to encounter
little or no difficulty when attempting to
finance GAG. Five utilities would be ex-
pected to encounter some difficulty financ-
ing the investment; for example, a decline
in bond rating or the supplanting of
-planned capital.expenditures. Only one
system would appear to face major diffi-
culties if it is required to install GAG
under present conditions.
• Under the high cost scenario', 12 of the 27
utilities would be expected to be able to
finance GAG through normal financial chan-
nels with little or no difficulty. Twelve
would be expected to encounter some diffi-
culty and three utilities would be expected
to encounter serious barriers to financing
under present conditions.
-------
IV-6
The results are displayed in Table IV-1 and are discussed
in more detail below. A complete listing of the financial ratio
analysis may be found in Appendix C.
Table IV-1
ABILITY TO FINANCE GRANUALAR ACTIVATED CARBON TREATMENT
UNDER A RANGE OF COST ASSUMPTIONS3
(# systems)
Should Be Able to Finance Through
Normal Capital Market Channels
Under Normal Market Conditions
May Experience Major
Difficulties Unless
Present Circumstances
Improve
Cost Scenario
Low GAC Cost
High GAC Cost
With Little or No
Expected Difficulty
21
11
With Some Difficulty
5
13
The cost assumptions employed in this analyses are ment to be illustrative of the
wide range of costs which might be incurred by affected utilities; the low cost and
high cos assumptions do not refer to specific design-related scenarios. Specific-
ally, low GAC Cost Scenario assumes GAC capital cost of $150,000/MG average daily
production and annual operating cost of S10,000/MG .average daily production. High
GAC cost scenario assumes GAC capital cost of 5400,'OOO/MGD and operating cost of
$20,000/MGO.
Systems Which Can Finance
With Little Or No Difficulty
As shown in Table IV-1, approximately 21 systems, or 78
percent of the total, should be able to finance the low cost
GAC treatment costs with little or no difficulty. Under the
high cost scenario this number drops to 11 systems, or about
40 percent. These systems are in strong financial condition
based on the criteria identified above.
For example, one system in this category would need to
raise approximately $18 million under the low cost scenario.
Its debt service coverage ratio would decline from a current
level of 2.1 times to approximately 1.8 times, assuming that
the capital and operating costs are passed on to the rate-
payer. This slight drop in the ratio should not endanger the
system's Aa rating by Moodyls Investors Service and is well
above Louisville's debt service covenant of 1.2 times.
-------
IV-7
Systems Which Could Finance With Some Difficulty
Between 20 and 50 percent of the systems were judged to
be able to finance the GAG treatment requirement but may ex-
perience some effect on their bond rating or other financing.
These systems would be faced with a trade-off between higher
water rates and financial strength. If they increase water
rates only enough to cover the direct capital and operating
costs of the GAG facility, their financial ratios would de-
cline to a level where the credit rating agencies might down-
grade their rating. Even if the ratios do not decline sig-
nificantly after the GAG treatment cost is absorbed, it may
be that other planned capital expenditures might need to be
postponed. The higher interest costs associated with a lower
credit rating and the possible non-discretionary nature of
some capital projects may provide the utility with a strong
incentive to pursue some alternative course of action.
A wide variety of utility actions are possible which may
mitigate the adverse1 effects of the financial burden:
• Revenue increases greater than the direct
capital and operating costs of the GAG
might be sought.
• Other means of financing a portion of the
GAG might be considered (e.g., special
assessments, general obligation bonds,
etc.).
• Staging of the financing of other capital
projects to allow a gradual phasing in
and lessening the impacts in any single
year could be attempted.
•—-Gradual phasing-in of -the- GAG treatment
requirement might be possible.
• Relief under the Act might be sought for
economic reasons which would delay the
installation of GAG until revenues could
be raised sufficiently.
The most likely and probably best course of action that
a utility might pursue would be a combination of the above.
In particular allowing for a slight decline in financial
ratios, seeking a moderate increase in rates, and the phasing
in of both GAG and other capital projects appears to be a
prudent strategy which would ameliorate the adverse effects.
-------
IV-8
Investor-owned water utilities may have a somewhat more
difficult financing challenge to install GAC treatment than
will municipal systems because of restrictions on water rates
imposed by state regulatory utility commissions. Virtually
all of these systems are required to obtain commission approval
for rate increases. The problem facing many investor-owned
•systems will not be that of obtaining the increases once treat-
ments are installed and operating, but of financing them during
construction. Some states such as Indiana will not allow con-
struction expenditures to be included in the rate base until
the equipment is "used and useful." Other states, such as
Pennsylvania, are allowing such expenditures to be included, at
least when they are required by a regulatory authority and do
not contribute to the productive capacity of the plant (e.g.,
pollution control expenditures). In some cases the financing
for GAC is significant enough that the additional carrying
costs and the effect on the financial ratios would probably
force an investor-owned system into special financing arrange-
ments or into altering the phasing of other capital projects.
Sy-stems were expected to encounter some difficulty fi-
nancing GAC treatment addition for several reasons. Some sys-
tems would see their debt service coverage ratios decline below
1.7 after the GAC treatment is financed. These systems may be
concerned that such a decline would cause a downgrading of their
credit rating and attendant higher interest costs. Other sys-
tems are placed in the second category because they serve cities
with plans for large capital expenditures in the future which
may need to be rescheduled if the system must install GAC.
Still others are included because they serve municipalities
with weak credit standings which may indicate difficulty in
obtaining the necessary revenue increases.
An example of a system which may be concerned about its
credit rating is one for which, under the low impact scenario,
its debt service coverage ratio would decline from its current
level of 3.4 to 1.9 times. This new lower level would still
be fairly strong in the industry and should allow the system
to maintain its Aa rating. Under the high cost scenario, how-
ever, the addition of a $51 million GAC treatment facility
would cause the ratio to decline to 1.4 times, if revenues
were increased only enough to cover direct capital and oper-
ating costs of the facility. Although this decline might
jeopardize the utility's present credit rating, it should not
prevent it from being able to raise the money. Moreover,
these adverse financial effects could be lessened if the
system could'obtain a rate increase in excess of the direct
capital and operating costs of the GAC facility.
-------
•IV-9
A typical system with large capital expenditures planned
has a debt service coverage ratio which is currently only at
about 1.31 times. Although the ratio would only drop slightly
(to 1.23) after the $96 million GAG investment required under
the high scenario, the already low ratio is some cause for con-
cern. The city's plans to raise $300 million over the next five
years, primarily for additional sewage treatment facilities, ex-
acerbate this concern. However, the $96 million GAG investment
under the high cost scenario, would represent only a 4.1 percent
increase in the system's debt as a percent of net assets and
could therefore be absorbed with little or no displacement of
other expenditures.
In contrast, another system has a rather strong debt service
coverage ratio for its self-supporting general obligation bonds.
These bonds, although they are general obligations of the city
it serves, are also secured by a first lien on the water rev-
enue earned. Moody's has recently downgraded its rating on the
bonds from A to Baal, due to the poor financial health of the
city itself. Although this rating drop would cause the water
system to pay a higher rate of interest on future debt issues,
it will not prevent it from issuing debt secured by increased
revenues to finance GAG treatment if costs remain in the range
of the low cost scenario.
Systems Which May Experience
Major Difficulties
The third category of water systems contains one system
in the low impact case and three in the high impact case. One
system is included under both cases not only because its fi-
nancial capability is only marginal at present (BAA rating
by Moody's, the lowest, in this sample) but a.lso because re-
cently passed legislation in the state puts all future mu-
nicipal financing into a very uncertain state of affairs.
The other systems are characterized by low credit ratings
(Baa 1 in one case), low Debt Service Coverage ratios (1.3 and
1.4) or a weak municipal finance situation.
Possible remedies to their tenuous situation include: (1)
revenue increases in excess of those required to pay the direct
capital and operating costs of GAG; (2) a gradual phasing-in of
the GAG treatment requirement to allow the systems to finance
the project in stages while its revenues climb; (3) some form
of local or state guarantee of the revenue bond--perhaps in the
form of a self-supporting general obligation; (4) an alternative
treatment other than GAG; or (5) qualification for relief on
economic grounds under the Safe Drinking Water Act.
-------
APPENDIX A
ANALYSIS OF GAG COSTS FOR SELECTED CITIES
TBS
-------
APPENDIX A
INTRODUCTION
During the public comment period following the proposal
of the regulations for trihalomethanes and synthetic organics,
EPA has received numerous comments from water systems related
to economics. TBS and ERGO, along with the staff of EPA's
Office of Drinking Water, decided to analyze in detail the
costs developed by New Orleans, Louisville, and Indianapolis.
These systems had expressed a strong desire to discuss the cost
differences, and they in one way or another seem to typify many
of the problems that might be incurred implementing granular
activated carbon treatment. Their costs were developed in de-
tail and their estimates varied significantly from the na-
tional unit costs which TBS and ERGO had developed. Two of
the systems, New Orleans and Indianapolis, were visited while
Louisville's estimates were discussed over the telephone.
NEW ORLEANS
Three meetings were held with the staff of the New Orleans
Sewerage and Water Board during April, May, and June. At the
first meeting, the differences between the TBS/ERGO and New
Orleans assumptions and costs were discussed. Since many of
the cost differences seemed either to be site specific or to in-
volve crucial differences in assumptions, Gannett Fleming Corddry
and Carpenter, Inc., a consulting engineering firm, was included
to help resolve some of the differences. The purpose of the
second meeting was to visit each of the New Orleans plants and
to determine the local constraints to the implementation of GAG
treatment. The third meeting was held to discuss Gannett Flem-
ing Corddry and Carpenter Inc.'s estimates as well as TBS/ERCO
and New Orleans revisions to their respective analyses.
Capital Costs
New Orleans has two water supply treatment plants that
have a combined design capacity of 253 MGD: 237 MGD at the
New Orleans (Carrollton) plant and 16 MGD at the Algiers
plant. Average daily flow at these two plants is 125 and 10
-------
A-2
MOD respectively. The combined peak month average daily flow
rate, as estimated by the flow rate on the average day in the
maximum month over the last five years, is 166 MGD, 150 of
which is at the Carrollton plant.
The first cost estimate for GAG treatment developed by the
New Orleans Sewerage and Water Board staff was prepared in the
spring of 1978 and was cited by the staff during EPA's public
hearings on the proposed regulation. That estimate totalled
$89.8 million in capital cost. In explaining their cost es-
timate, the New Orleans staff cited substantial site specific
problems which they would encounter in installing GAG. No
specific cost estimate for New Orleans had been prepared by
EPA, but when the cost parameters from the national analysis
were simply applied to the production volumes in New Orleans
and escalated for inflation to 1978, the resulting cost esti-
mate was only $21.7 million.
The difference in cost estimates was significant enough
that EPA wanted the difference reconciled and the site specific
problems examined, both for perspective on the local impact of
the proposed regulations in New Orleans and for potential re-
vision of the national cost estimates. The difference in
costs was so great that there had to be substantial differ-
ences in assumptions as well as site specific complications
in installing the treatment at these plants.
EPA and TBS decided to retain a consulting engineering
firm experienced in the design of water treatment facilitites
to help reconcile these differences as an independent third
party. The firm of-Gannett Fleming Corddry and Carpenter, Inc.
(GFCC) was retained for the purpose of reviewing the New Orleans
water treatment plants and developing independent preliminary
cost estimates for the use of granular activated carbon at
those plants. It was hoped that this independent analysis
would help reconcile the cost differences in this specific
case. It was also felt that the review process might provide
useful feedback on some of the national cost assumptions as a
by-product.
The first meeting to reconcile costs was held in New
Orleans on April 24, 1978. Participants represented the New
Orleans Sewerage and Water Board, EPA's Cincinnati Municipal
Environmental Research Laboratory, TBS, ERCO, and Gannett
Fleming Corddry and Carpenter, Inc. A number of technical
assumptions were agreed upon at that meeting for purposes of
cost comparison so that all parties could develop comparable
estimates in order to identify and r-econcile differences. These
-------
A-3
assumptions were not necessarily agreed to as the proper basis
for final GAG system design, but were developed to provide a
common basis for discussing costs. These assumptions are listed
in Table A-l below.
Table A-L
TECHNICAL ASSUMPTIONS
FOR PURPOSES OF COST COMPARISON IN NEW ORLEANS
Plant design capacity
Average daily flow
Average daily flow in maximum-
month
Carbon fill and contactor
sizing
Hydraulic capacity _ •>.
Contactor design
Empty bed contact time
Contactor bed depth
Carbon density
Carbon cost
Carbon loss rate during
regeneration and handling
Regeneration cycle
Suffer stock of carbon
Fees for engineering design,
legal and financing
Land acquisition
Cost basis
Carrollton 237 MGO/Algiers 15 MGD
Carroll ton 125 MGO/Algiers 10 MGO
Carroll ton- 150 MGO/Algiers 16 MGD
Carroll ton 150 MGD/Algiers 16 MGO
Carroll ton 237 MGO/Algiers 16 MGO
Filter-type, in-ground units
20 minutes
10 feet
28 Ib./cubic ft.
55ei/lb.
7 percent
50 days
7 percent of initial fill
15 percent of capital costs j
• - i
Carrollton none/Algiers one parcel
1978 dollars
Two sets of cost estimates were prepared on the,basis of
these assumptions, one by the Sewerage and Water Board (SWB)
and one by GFCC. The process included a second meeting for
GFCC to visit both New Orleans water treatment plants and to
review site specific issues. GFCC developed a detailed pre-
liminary cost estimate based on the technical assumptions and
on local conditions at the two plants. The firm developed
independent cost estimates for most of the construction and
equipment items. The GFCC estimates are reproduced in their
entirety as Appendix B of this document.
-------
A-4
A third meeting was held in early. June with representatives
of the New Orleans SWB, GFCC, TBS and ERGO to review the re-
sults of the two efforts. The outcome was that there was sub-
stantial agreement on costs in the $55 to $65 million range
for -designs based upon the common technical assumptions. A
summary comparison of the estimates is provided below in Table
A-2. Significant differences appear only in the cost of fur-
naces. In that case SWB and GFCC agreed on the number and size
of furnaces required, and simply used different sources for
cost estimates. GFCC used a quote from a furnace manufacturer
while SWB relied upon data from a January 1978 article in
Chemical Engineering.1
Table A-2
COMPARISON OF GFCC AND SWB
COST ESTIMATES ON COMMON TECHNICAL ASSUMPTIONS
(costs rounded to nearest thousand 1978 dollars)
Carbon, including inventory
Contactors, filter type
Transfer pumps
Carbon transfer equipment
Carbon storage
Regeneration furnaces and building
Local-chlorine tank and station,
change in existing piping, de-
molition, storage tanks
Land aquisition
Total Construction Costs
Engineering and other fees
Total Project Costs
SWB* GFCC^
S 5,100
21,100
7,500
1,400
15,100
6,400
2,500
S 5,100
17,900
7,200
1,500
500
7,600
7,300
800
Sewage and Water Board (SWB) of New Orleans estimates
obtained in June 19, 1978 meeting.
*•*
Gannett Fleming Corrdry and Carpenter, Inc. (GFCC)
figures from Appendix B of this report.
While SWB personnel were willing to compare costs on the
basis of the technical assumptions outlined under Table A-l,
SWB personnel do not accept the technical assumptions as the
best estimate of design and operating parameters which would
be used if GAC were actually installed in New Orleans. In par-
ticular, the SWB staff indicated they would not agree to a
R. H. Zanitsch and R. T. Lynch, "Thermal Regeneration Systems
for Activated Carbon," Chemical Engineering, January 2, 1978.
-------
A-5
design in which the desired empty bed contact time was sized
to anything less than the plant design capacities. All parties
recognize that pilot tests will be necessary to establish the
proper design and operating parameters.
While GFCC and SWB were developing their New Orleans cost
estimates, TBS and ERGO were also revising their cost
estimates for the national economic analysis. TBS and ERGO
representatives attended the meetings in New Orleans and also
met separately with GFCC to understand fully the basis for
those estimates.
A number of points raised in the context of the New
Orleans discussions were relevant to the national analysis,
even though neither GFCC nor SWB was directing its efforts
to that end. This process of site specific reconciliation
helped to identify the need for four significant revisions
to the national costs:
• Increasing carbon fill and contactor sizing
to the peak month average daily flow from
the previous basis of annual average daily
flow
• Adding contingencies to all cost estimates
covering equipment and construction
• Increasing the design fee allowance from the
6 to 8 percent level for engineering design
alone to 15 percent to also include pilot
testing, construction inspection, legal fees,
and financing fees
• Including a variable allowance for site spe-
cific costs in excess of.the national unit
costs, for items such as additional pumping,
piping modifications, and more costly site
preparations such as pilings under foundations.
The national unit cost estimates presented in Chapter II
are lower than the SWB and GFCC estimates due to differences
between the national assumptions and the technical assumptions
agreed to for the New Orleans cost comparison. When one selects
unit costs relating to the pumping levels of'the New Orleans
plants from the cost curve in Chapter II (p. 11-22), a total
cost of approximately $32 million results before including any
allowance for site specific additional costs. Including such
local costs, total project costs would be estimated to range
from $32 to S40 million.
-------
A-6
The differences between the cost estimate based on these
national unit costs and the GFCC estimate for New Orleans
based on the specific New Orleans technical assumptions are:
• Carbon density of 26 pounds per cubic foot in
the national analysis versus 28 in the New
Orleans review, accounting for $0.4 million
• Sizing of contactors, initial carbon fill and
buffer carbon stock all based on 18 minutes
of contact time nationally compared to 20
minutes for New Orleans, amounting to $2.7
million including contingencies and fees
• Contactor design, assumed to be above-ground
cylinderical steel tanks in the national anal-
ysis and to be reinforced concrete filter-like
boxes with pile foundations in the New Orleans
analysis, accounting for the largest difference,
$6.6 million including contingencies and fees
• Pumping costs for filtered water transfer to
the contactors—first in terms of the number
of pumping stations, assumed to be one pump
station per plant in the national analysis,
but probably requiring three stations in New
Orleans due to plant layout at Carrollton;
and second in terms of units costs, the
sources of cost data differ—the total vari-
ance in these estimates is $5.3 million in-
cluding contingencies and fees
Operating Costs
No operating costs were submitted by New Orleans in its
comments to EPA on April 5, 1978.
INDIANAPOLIS
One meeting was held with the staff of the Indianapolis
Water Company in June. At the meeting were representatives
of TBS, ERCO, Gannett Fleming Corddry and Carpenter, Inc.,
the Indianapolis ?7ater Company, Black and Veatch (Indianapolis
Water Company's Consulting Engineer), and Baker and Daniels
(Indianapolis Water.Company's attorneys and the attorneys
for the Safe Drinking Water Coalition). The main purpose of
the meeting was to discuss the assumptions used by Black and
-------
A-7
Veatch and the Indianapolis Water Company in developing the
cost estimate which they presented in EPA' s public hearing.
No specific GAG cost estimate was developed by EPA's
contractors in the case of the Indianapolis Water Company.
To the extent that the discussion dealt with differences
between EPA's analyses and the company's, it focused upon
contrasts of the Indianapolis assumptions with the generic
approach employed by TBS and ERCO. A detailed comparison of
costs has not been conducted; the information which follows is
preliminary and intended to highlight the general areas of dif-
ference rather than to be a specific reconciliation of costs.
Capital Costs
Indianapolis has three treatment plants with a combined
design capacity of 190 MGD. Average annual production is 103
MGD. The design capacity at each plant is: White River 130
MGD; Fall Creek 40 MGD; and Eagle Creek 20 MGD. The three rivers
differ in their vulnerability to contamination from synthetic
organics and may also differ in their trihalomethane concentra-
tions. It is not certain whether any or all of the plants would
be affected by the proposed regulations.
Indianapolis' consulting engineer, Black and Veatch, esti-
mated a total capital cost of $45.3 million to install GAG treat-
ment facilities at all three plants. If the national unit
cost estimates of TBS and ERCO (as revised in this report) were
applied to the total Indianapolis production figures, one would
obtain cost estimates of approximately $23 million without any
allowance for local conditions (such as multiple plants), and
$23 to $29 million including local extra costs.
If the national unit costs were applied, to the flows of
the three plants individually, the total cost estimate would
be $33 million without any allowance for local extra costs.
This difference of approximately $10 million from the $23
million estimate is due to duplication of facilities and to
diseconomies of scale resulting from treatment installation
at three locations, two of them relatively small treatment plants
The greatest difference, just over half, is in the cost of re-
generation furnaces, since more furnaces would be required and
each would be smaller than if they were sized to. meet the sys-
tem's combined needs at one location.
A number of reasons account for the difference between the
Indianapolis estimate of $45 million and the estimate of $33
million based on national cost. Those reasons include:
-------
A-8
Sizing. Indianapolis specified carbon and contactor re-
quirements for design capacity and then built in a reserve mar-
gin. As a result the contact time even at design capacity would
range from 20 to 26 minutes at the three plants if the reserve
capacity were used. In comparison tl^e TBS/ERGO costs would
be based on an 18 minute contact time at a flow of 129 MGD,
the estimated average day in the maximum month of the year.
At the annual average flow rate of 103 MGD, the Indianapolis
sizing would yield contact time of over 35 minutes, and the
national estimates would result in contact time of over 22
minutes.
As a result, carbon and contactor volume as specified in
the Black and Veatch estimates exceeds the TBS/ERGO standard
by about 70 percent. The difference in total cost due to this
sizing difference is about $5.3 million in the case of the aggre-
gate cost estimate. This amount consists of $3.3 million for
differences in contactors and $2.0 million for differences in
carbon sizing. Interestingly, at the largest plant, Indianapolis'
estimates for contactor unit costs are lower than the national
figures on a cost per MGD basis.
Carbon Price. Carbon costs of $.60 per pound were used
by Indianapolis versus $.53 per pound in the TBS/ERGO estimate.
This assumption accounts for $0.7 million in addition to the
sizing difference above.
Furnace Costs. The Indianapolis costs were broken down
on a plant-by-plant basis assuming at least two furnaces per
plant for a total of six furnaces. The TBS/ERGO national meth-
odology, when applied on a plant-by-plant basis would result
in the same assumption. There is a significant difference in
hearth loading rates between the Indianapolis and TBS/ERGO es-
timates. The former are based upon throughput of 45 pounds
per square foot of hearth area, which is fairly typical for
wastewater applications, whereas the latter are based on 110
pounds per square foot. Nevertheless, the total furnace costs
are very close when developed on a plant-by-plant basis. In
fact, the costs developed from-the TBS/ERGO national estimates
would be approximately $0.5 million lower than the Indianapolis
costs, before accounting for contingencies and fees.
Pumping Costs. The difference in pumping cost estimates
reflects higher unit costs assumed by Indianapolis for pumping
stations and associated piping. Overall the difference in es-
timates would be a little over $2.8 million, from Indianapolis'
estimate of $6.4 million.
-------
A-9
Site Specific Costs. Indianapolis felt that chlorine con-
tact basins would need to be constructed following GAG adsorption.
The existing clearwells would not be adequate since Indianapolis
maintains a chloramine residual in the distribution system. It
would not be possible to add both chlorine and ammonia in the
existing clearwells where only ammonia is fed at present. Chlo-
rine would also need to be added since prior carbon adsorption
would have removed chlorine. The cost of the contact basins
is approximately $2.5 million.
Contingencies and Fees. The remaining difference of $1.4
million is accounted for by contingencies and engineering, legal,
and financing fees. These items are higher in the Indianapolis
estimate since they have been calculated as a percentage of a
higher base capital cost.
Operating and Maint'enance Costs
Indianapolis estimated combined operating and maintenance
costs of $7.2 million, whereas the national parameters would
yield an estimate of approximately $2.5 million at a single
facility or $3.0 million at a combination of three plants the
sizes of the Indianapolis plants. The differences between the
Indianapolis estimates and these derived from national figures
cannot be reconciled completely because the documentation of
Indianapolis' estimates does not provide sufficient detail.
The reconciliation in the two areas which can be analyzed is
presented below.
Effect of Volume of Water Produced. Indianapolis devel-
oped annual O&M costs on the basis of an annual flow equal to de-
sign capacity (190 MGD). TBS/ERGO assumed that annual O&M costs
should be'based on actual annual production (103 MGD). The
most significant O&M costs, over 70 percent of the total, vary
with the amount of carbon regenerated, which is a function of
water production. If the Indianapolis figures were adjusted
for this difference in flows they would decline by approximately
$2.5 million.
Carbon Loss and Cost of Carbon. Indianapolis assumed a
carbon loss rate of 10 percent versus 7 percent in the TBS/ERGO
estimate and a1- price of $.60 vs. $.53 per pound. After adjust-
ing for the different flow rates, a residual difference of $0.6
million is due to the different assumptions on loss rate and
carbon price.
-------
A-10
LOUISVILLE
Due to time pressures, TBS and ERGO were unable to visit
the Louisville Water Company, although its staff was contacted
over the phone and some of the differences in assumptions were
discussed. The evaluation and comparison of assumptions pre-
sented here is less comprehensive than for the other two systems
as a result, and it may even contain some errors of omission or
misinterpretation. It is the intention of TBS and ERGO to con-
tinue the dialogue with the Louisville Water Company so that
EPA will gain the full benefit of Louisville's research into
GAG design and costs.-
Capital Costs
Louisville has two treatment plants that have a combined
design capacity of 300 MGD, comprised of 240 MOD at Crescent
•Hill and 60 MGD at Harrods Creek. The average daily flow is
140 MGD at Crescent Hill and 20 MGD at Harrods Creek. Louis-
ville estimated a combined capital cost for GAG installations
at its two plants of $81.3 million in 1978 dollars.
If one applied the national cost estimating methodology
(as revised in this report) without any site specific costs,
an estimate of approximately $35 million would result. This
system cost is based on an average daily flow of 160 MGD, de-
sign capacity of 300 MGD and an estimated average day in the
maximum month of 200 MGD. With a range of extra costs for
local conditions, the national costs would imply a total of
from $35 to $43 million for Louisville.
If the national unit costs were applied to the flows of
the two plants individually, the total cost estimate would be
$39 million- exclusive of any site-specific costs. The remaining
difference of $42 million ($81 million minus $39 million) is
acounted for by the following items:
Contactor Sizing. Louisville's costs for contactors and
carbon were based on an 18-minute contact time at 70 percent
of design capacity, plus a margin for units out of service for
carbon removal and replacement. The result is that Louisville
would estimate contactor requirements of 460,000 cubic feet of
effective contactor volume (i.e., GAG fill). In contrast, the
national TBS/ERCO parameters would yield an estimated effective
contactor volume of 334,000 cubic feet. This difference com-
pletely accounts for the variation in contactor cost estimates
of $17.7 million developed by Louisville and $12.9 million es-
timated from the national figures. The direct implication of
-------
A-ll
this result is that the unit costs of the independently prepared
Louisville and TBS/ERGO analyses agree.
Carbon Costs. Louisville estimated $9.0 million for car-
bon costs, including wasteage and the initial placement of GAG
in the contactors. The national figures, however, would indi-
cate a cost of only $4.9 million. There are five factors which
seem to account for the difference: (1) the sizing, as with
the contactors above, amounts to additional carbon costs of
$2.4 million; (2) the price per pound of 60^ instead of 53£
accounts for $0.7 million; (3) the assumption on density of
carbon at 29 instead of 26 pounds per cubic foot represents
another $0.6 million; (4) the wasteage and initial placement
of carbon, after adjustment to a smaller initial fill size,
represents $0.7 million which is not explicitly in the national
estimates; (5) the buffer stock which is in the national esti-
mates and not in the Louisville analysis adds an off-setting
$0.3 million.
Furnace Costs. Louisville estimated total furnace costs
of $23.4 million while the TBS/ERGO parameters would yield
a cost of $8.3 million. Some of the difference is due to fur-
nace sizing assumptions. The TBS/ERGO analysis assumed furnaces
would be sized to handle peak month daily needs with one furnace
out of service, while Louisville's estimates were based on meet-
ing needs at design capacity with some reserve margin. The
most significant difference in furnace costs, however, stems
from the fact that Louisville's furnaces are over three times
as expensive as the furnaces in the TBS/ERGO analysis per pound
of regenerating capacity. This is mostly due to a much lower
throughput rate in the Louisville analysis. Louisville's costs
were developed from the article in Chemical Engineering.2
Ozonation Costs. Louisville assumed the 60 day regenera-
tion cycle could only be maintained if ozonation preceded carbon
adsorption otherwise trihalomethane breakthrough would occur.
This cost amounted to $5.7 million.
Site Specific Costs. Louisville included some $15.5 million
in site specific costs for such items as land acquisition, site
preparation and clearwell addition. Louisville is served by two
treatment plants and this leads to higher costs for most capital
items and higher site specific costs than if the system were
served by one large plant. This cost amounts to an add-on of
almost 23 percent over the total of the other capital costs.
o
R. H. • Zanitsch and R. T. Lynch, Ibid.
-------
A-12
Pumping costs. The estimates for pump stations and asso-
ciated piping which were subject to wide variations in the New
Orleans analysis were not significantly different in Louisville.
The TBS/ERGO estimates would be approximately $4.0 million vs.
$5.0 million for Louisville.
Fees and Contingencies. Costs for professional fees were
included in both the Louisville and the national estimates.
TBS/ERGO calculated fees for engineering, legal, and financing
services at 15 percent of total capital costs whereas Louisville
added 6.5 percent to its total costs for fees. The two fee
estimates—$5.1 million from the national figures and $5.0 mil-
lion from Louisville—are nearly the same since the base capital
costs in the Louisville estimate are considerably higher. An
allowance for contingencies was included in the national anal-
ysis, and would amount to $3.8 million in the case of the two
Louisville plants. The estimates prepared by Louisville did
not include a line item for contingencies.
Operating and Maintenance Costs
Louisville estimated a total operating and maintenance cost
of $6.3 million. Based on costs for individual plants with
average daily flows of 20 MGD at Harrods Creek and 140 MOD at
Crescent Hill, the TBS/ERCO unit costs would yield an annual
0/M cost estimate of $4.0 million. A number of factors account
for the difference as described below.
Carbon Losses. Louisville assumed a 10 percent loss rate
versus 7 percent in the TBS/ERCO analysis. At a carbon cost of
$.53 per pound and a density of 26 pounds per cubic foot-this
difference alone is $0.7 million.
Carbon Cost and Density. Louisville assumed a carbon cost
of $.60 per pound and a density of 29 pounds per cubic foot.
Compared with the TBS/ERCO assumption of $.53 per pound and 26
pounds per cubic foot this difference accounts for $0.6 million
per year for carbon replacement.
Ozone Generating Cost. Louisville assumed an annual power
cost of $0.3 million to generate ozone while the national 'ERGO/
TBS estimates did not include ozonation as a treatment that
would be necessary with GAC.
-------
A-13
Fuel Usage and Costs. Louisville assumed a fuel cost of
$.034 per pound while the TBS/EECO estimates included a cost of
$.024. The difference is mainly accounted for by a higher as-
sumption of the necessary Btus to regenerate a pound of carbon.
Louisville assumed 8000 Btu while TBS/ERCO assumed 5000 Btu.
This difference amounts to $0.5 million.
Labor Costs. Louisville assumed some 60 people were
necessary to operate the GAC treatment process including lab
technicians. This is approximately two times the number of
people implied by the national TBS/ERCO parameters and ac-
counts for the remaining difference of $0.4 million in cost
totals.
-------
APPENDIX B
ESTIMATED COST FOR
GRANULAR ACTIVATED CARBON (GAG)
FACILITIES—NEW ORLEANS WATER
TREATMENT PLANTS
-------
APPENDIX B
In order to reconcile the differences between EPA's
GAG cost estimates and those provided by the New Orleans
Sewerage and Water Board, TBS subcontracted with Gannett
Fleming Corddry and Carpenter, Inc. to provide an indepen-
dent preliminary capital cost estimate. The GFC&C, Inc. re-
port detailing the resultant construction and project cost
estimates is reproduced in its entirety in this Appendix.
-------
ESTIMATED COSTS
FOR
GRANULAR ACTIVATED CARBON (GAC) FACILITIES
NEW ORLESNS WATER TREATMENT PLANTS
June 1978
Prepared by
GANNETT FLEMING CORDDRY AND CARPENTER, INC .
Harrisburg, Pennsylvania
-------
c_ GANNETT FLEMING CQRDDRY AND CARPENTER. INC.
ESTIMATED COSTS
FOR
GRANULAR ACTIVATED CARBON (GAG) FACILITIES
NEW ORLEANS WATER TREATMENT PLANT
CONTENTS
I. INTRODUCTION
II. PURPOSE AND SCOPE OF WORK
HI. DEVELOPMENT OF COST ESTIMATES
IV. SUMMARY OF DESIGN GUIDELINES
V. TABLES 1 through 10
TABLE 1 - COST ESTIMATES FOR GAC FACILITIES
TABLE 2 - GAC INITIAL FILL
TABLE 3 - CARBON CONTACTORS
TABLE 4 - REGENERATION FURNACES
TABLE 5 - FILTERED WATER TRANSFER SYSTEM
TABLE 6 - CARBON TRANSFER EQUIPMENT
TABLE 7 - BUFFER CARBON STOCK
TABLE 8 - BUFFER CARBON STORAGE FACILITIES
TABLE 9 - LOCAL ITEMS
TABLE 10 - ALTERNATE DESIGNS OF CARBON CONTACTORS
-------
GANNETT FLEMING CORODRY AND UARPENTER. ir«i.
ESTIMATED COSTS
FOR
GRANULAR ACTIVATED CARBON (GAC) FACILITIES
NEW ORLEANS WATER TREATMENT PLANTS
I.. INTRODUCTION
Previous preliminary estimates prepared by Temple Barker and
Sloane (TBS) and the Sewerage and Water Board of New Orleans (SWB)
resulted in a wide difference in the estimated project cost of the granu-
lar activated carbon (GAC) facilities required at the two water treatment
plants owned and operated by SWB to comply with the proposed Interim
Primary Drinking Water Regulations for the Control of Organic Chemical
Contaminants in Drinking Water" issued by the Environmental Protection
Agency in the Federal Register dated February 9, 1978 , It was decided,
therefore, that it would be best to have a consulting engineering firm,
experienced in the design of water treatment facilities, make an investi-
gation of the specific site conditions at both plants in New Orleans and
prepare independent construction and project cost estimates for the re-
quired GAC facilities.
Gannett Fleming Corddry and Carpenter, Inc., (GFC <£C , Inc.), a
consulting engineering firm with its home offices in Harrisburg, Pennsyl-
vania, was retained by TBS to develop these independent cost estimates.
II. PURPOSE AND SCOPE
The purpose of this assignment is to establish a realistic level of
cost for the required GAC facilities. Prior to initiating this assignment,
it was clearly understood by all parties that the intent was to have
Gannett Fleming Corddry and Carpenter, Inc., prepare independent cost
estimates and not attempt to verify either of the previous estimates.
Prior to preparing the cost estimates, it was agreed that the
estimates would be based on available data, data developed tn the prelim-
inary investigations, and on the engineering experience and professional
judgement of the staff of Gannett Fleming Corddry and Carpenter, Inc. It
was further agreed that no detailed investigations, foundation exploration,
hydraulic analyses, pilot studies or actual designs would be undertaken
as a part of the assignment.
The general physical arrangement of the existing water treatment
facilities was to be considered, however, in the development of any pre-
liminary concepts and layouts on which the estimated costs were based.
The proposed Interim Primary Drinking Water Regulations require
the use of an interim control measure where treatment systems using
-------
GAG would not be operational until 2 to 3-1/2 years after the effective
date . The scope of this assignment did not include the development of
costs to satisfy this interim control measure.
Similarly, the scope of this assignment did not include the de-
velopment of.operating costs or annual costs.
HI. DEVELOPMENT OF COST ESTIMATES
1. General.
Prior to preparing cost estimates for installing granular activated
carbon (GAG) facilities at the Carrollton and Algiers water treatment
plants in New Orleans, meetings were held with representatives of the
staff of the Sewerage and Water Board. During these meetings, certain
basic design guidelines were discussed and generally agreed upon for
purpose of cost comparison. Also, information was obtained on the exist-
ing physical features of the water treatment plants and on the present
operation of the plants. Onsite inspections were made of both plants.
The cost estimates were developed, therefore, taking into consider-
ation data obtained from the onsite inspections, design guidelines agreed
upon with the staff of SWB, for purpose of cost comparison, preliminary
concepts for the required facilities, published data,' quotes from equipment
manufacturers, costs of other water treatment facilities previously designed
by GFC&C, Inc., and the engineering experience and professional judge-
ment of the staff of GFC&C , Inc.
2. Preliminary Layouts.
Drawings showing the overall plant layouts at both facilities were
obtained from SWB. These drawings were utilized to establish a general
layout for the required GAC facilities in accordance with the preliminary
concepts developed by the SWB staff.
Items taken into consideration were: sizing of the carbon contact-
ors; sizing of the filtered water transfer pumping station and connecting
piping; sizing of the furnaces; physical location of the required facilities;
routing of proposed piping; foundation conditions; housing of furnaces;
architectural treatment of buildings; connections to existing piping, etc.
3. Development of Costs.
Cost estimates for the furnaces and filtered water transfer pumps
were developed, utilizing quotations from equipment manufacturers. Unit
costs per square foot of the reinforced concrete carbon contactors at the
Carrollton Plant were developed by making a preliminary layout of the
contactors, sizing a typical contactor and assigning appropriate costs to
each of the general elements that are considered a part of the carbon con-
tactor. (See TABLE 3). The unit cost per square foot of contactor
-2-
-------
GANNETT FLEMING CQRDDRY AND CARPENTER. INC.
developed for the Carrollton Plant was also used in developing the total
carbon contactor costs for the Algiers Plant. Estimated costs for the
remainder of the work items, including outside piping, pumping stations,
concrete structures, steel tanks, mechanical, electrical and ventilating
work, were based on the cost of similar items of work taken from projects
previously designed by GFC&C, Inc.
Information obtained from the staff of SWB included the cost of
demolition of certain existing structures, purchase of property, and other
miscellaneous items. The staff of SWB also provided valuable input in
the development of costs for piling, pumping facilities, and outside pip-
ing from their past experience at the specific sites.
Based on preliminary investigations, it was assumed that the carbon
contactors would be mostly aboveground to conform with existing hydraulic
conditions, and that certain outside piping would be constructed above- .
ground due to the limited space available for installing underground piping.
4. Limitations.
Although specific costs were developed for each respective item
as outlined in the attached tables, the costs must be considered to be
approximations. The costs are not based orr actual quantities determined
from detailed designs and necessarily do not take into account all the de-
tails and requirements of a complete installation. As noted hereinbelow,
however, a contingency factor was incorporated into the estimated costs
to cover these uncertainties.
Costs were developed for gravity-type carbon contactors having
a bed depth of 10 feet and providing 20 minutes empty bed contact time.
These criteria resulted in a surface loading of 3.74 g.p.m. per square
foot of contactor. In addition, the component cost of the proposed con-
tactors were then modified to determine comparative costs for a gravity-
type contactor with a surface loading of 5 g.p.m. per square foot. The
comparison indicates that using a higher surface loading results in a
significant cost savings. No attempt was made, however, to determine
the most economical contactor sizing, contactor arrangement or contactor
type. Any economical comparison of contactor type should include steel
upflow and steel gravity units. -«- ":~ - ~ - • --• ----- - -
All cost estimates are based on 1978 cost levels.
No pilot tests were run to determine basic design criteria such as
the optimum empty bed contact time, bed depth, surface loading, regener-
ation cycle, etc. These criteria can be determined only through adequate
pilot testing.
The general layout of the propo-sed facilities as developed by the
SWB staff was adopted for use in preparing the cost estimates. Alternate
layouts were not developed to determine the most economical arrangement
and location of the required facilities.
-3-
-------
GANNETT FLEMING CORDDRY AND CARPENTER. INC.
No costs were included for fuel storage or for a fuel transfer
system between the fuel storage facilities and the furnaces.
Foundation conditions were assumed, based on the information
available from the SWB staff.
No hydraulic analyses were made to determine exact head losses
in the connecting piping between structures. Also, the existing piping
was not uncovered to verify its location or size.
It was assumed that wastewater from the carbon handling facilities
and from the scrubbers on the furnaces will be discharged into existing
wastewater pipe systems and that no specific treatment of these waste-
waters will be required.
5. Contingencies.
A contingency factor of 15 percent was incorporated into the cost
estimates for constructing the GAG facilities. No contingency was added,
however, to the estimated costs of the initial carbon fill and the buffer
stock.
The contingency factor is intended to cover the unknowns and un-
certainties related to preparing estimates without benefit of proper pilot
tests, foundation exploration, verification of underground piping and con-
duits, detailed designs, construction drawings, detailed quantity takeoffs,
etc. The contingency factor is also intended to cover general problems
associated with plant expansions and modifications, maintaining electri-
cal service during construction, keeping the remainder of the plant in
operation during construction, connecting to existing piping and conduits,
etc.
6. Alternatives.
A general review of the preliminary concept for the layout of the
GAG facilities as provided by SWB indicated that there are several areas
where alternate layouts may result in cost savings. These potential alter-
nates include the following:
• Purchase additional property and utilize the athletic field at the
Carrollton Plant for the construction of the GAG facilities in lieu
of demolishing the existing filtered water storage reservoir, re-
placing the lost finished water storage capacity with six -
3.3 million gallon steel storage tanks, and demolishing the exist-
ing abandoned sedimentation chamber.
• Regenerate spent carbon from both plants at the Carrollton Plant.
Under this concept the furnaces at the Carrollton Plant would be
adequately sized to handle the carbon regeneration requirements
of both plants, and the spent and regenerated carbon would be
trucked back and forth between plants.
-4-
-------
GANNETT FLEMING- COROORY AND. CARPENTER.- INC—
Provide only one furnace rather than two at the Algiers Plant.
Again, under this concept, adequate capacity would be built
into the furnaces at the Garrollton Plant to provide standby
capacity for the Algiers Plant.
Consider alternate carbon contactor designs which may provide
more cost effective sizing. Preliminary investigations indicated
a savings of approximately $2,700,000 could be realized if the
basic criteria was changed from a contactor bed depth of 10 feet
(which results in a surface loading of 3.74 g.p.m. per square
foot of contactor area) to a surface loading of 5 g.p.m. per
square foot (which results in a contactor bed depth of 13 .4 feet).
Collect all of the filtered water at Carrollton Plant in one location
and provide one filtered water transfer pump station in lieu of the
two indicated on TABLE 5.
-------
GANNETT FLEMING COROORY AND CARPENTER. INC.
IV. SUMMARY OF DESIGN GUIDELINES
The following design guidelines were used in the preparation of the
cost estimates.
Design Itama
* Hydraulic capacity (rated
plant capacity)
*Contactor design flow (average
day of maximum month)
*Empty bed contact time
^Contactor bed depth
*Unit weight of carbon
*Cost of GAG
*Regeneration cycle
*Buffer stock GAG
Contactor surface area
Contactor surface loading
Number of contactors
Initial GAG fill
Buffer stock of GAG
Carbon regeneration furnace
capacity
Number of regeneration furnaces
with one unit out of service
and 60-day regeneration
cycle
Regeneration furnace size
Carbon regeneration:
Cairollton Plant
237 m.g.d.
Algiers Plant
16 m.g.d.
16 m.g.d.
150 m.g.d.
20 minutes
10 ft.** '
28 lb/ft3
$0.5S/lb.
50 days
7% of initial fill
27,852 ft2 2,971 ft 2
3.74 g.p.m./ft2**
40 4
278,520 ft3 29,710 ft3
19,496 ft3 2,080 ft3
90-120 lbs/day/ft2
of hearth area
32.5 ton/day
7 ton/day
60 days at average annual
daily flow
101,20.0 Ibs/day
10,800 Ibs/day
-------
GANNETT FLEMING CORDDRY AND CARPENTER. INC.
Design Items Carrollton Plant Algiers Plant
Carbon regeneration: (Cont'd.)
60 days at average daily
flow of maximum month 129, 600 Ibs/day 13, 800 Ibs/day
45 days at average daily
flow of maximum month 172,800 Ibs/day 18,400 Ibs/day
Chlorine contact time at
maximum daily flow 20 minutes
Filtered water transfer pumps -
connected capacity with largest
unit out of service 237m.g.d. 16m.g.d.
Carbon contactors Gravity flow, reinforced-concrete
boxes similar to conventional gravity
filters.
* Design guidelines discussed with and generally agreed upon by the staff
of the Sewerage and Water Board of New Orleans (SWB) for the purpose of
cost comparison.
** As an alternative design, an estimate was also prepared based on a
contactor surface loading of 5 g.p.m./ft2 of contactor surface area
which resulted in a bed depth of 13.4 feet.
-------
GANNETT FLEMING CORDDRY AND CARPENTER. INC.
TABLE 1
COST ESTIMATES FOR GAG FACILITIES
CARROLLTON AND ALGIERS WATER TREATMENT PLANTS
NEW ORLEANS
GAG initial fill $ 4,746,742.00
Carbon contactors 17,877,340.00
Regeneration furnaces 7,625,000.00
Filtered water transfer system 7,245,000.00
Carbon transfer equipment 1,495 ,000. 00
Buffer carbon stock " 332,270.00
Buffer carbon storage facilities ' 644,000.00
Local items (demolition of structures, land
acquisition, chlorination facilities,
chlorine contact tank, and storage tanks) 8,112 ,735 . 00
TOTAL CONSTRUCTION COSTS $48,078,087.00
Fees (engineering, legal, topographic surveys,
pilot tests, pilot test facilities, financial
work, foundation exploration, administration,
inspection, regulatory agencies' approval, etc.) 7,221,9.13 .00
TOTAL PROJECT COSTS $55,300,000.00
-------
GANNETT FLEMING CQROORY AND CARPENTER. INC.
TABLE 2
GAG INITIAL FILL
Key Assumptions:
a. Empty bed contact time
b. Unit weight of granular activated carbon (GAG)
c. Cost of GAG
d. Contractor design flows' (average day of
maximum month
Carrollton Plant
Algiers Plant
" * TOTAL
Sizing and Costs:
Contactor volume
flow (o.p.m.) x 20 min.
7.48 gal/ft3
Carrollton Plant
Algiers Plant
TOTAL
Carbon weight
(volume x 28.1bs/ft3)
20 minutes
28 lbs/ft3
$0.55 per Ib.
Carbon costs
(weight x $0.55/lb.)
150 m.g.d.
16 m.cr.d.
166 m.g.d.
278,520 ft3
29,710 ft3
308,230 ft3
Carrollton Plant 7,798,560 Ibs.
Algiers Plant 831,880 Ibs
TOTAL 8,630,440 Ibs,
Carrollton Plant $4,289,208.00
Algiers Plant 457,534.00
TOTAL $4,746,742.00
-------
GANNETT FLEMING CORDDRY AND CARPENTER. INC.
TABLE 3
CARBON CONTACTORS
Key Assumptions:
a. Hydraulic capacity (rated plant capacity)
Carrollton Plant 237 m.g.d.
Algiers Plant 16 m.g.d,
b. Carbon contactor design flows (average
day of maximum month)
Carrollton Plant 150 m.g.d,
Algiers Plant 16 m.g.d,
c. Carbon contactor bed depth 10 ft.
d. Gravity flow
e. Concrete box similar to conventional filter units
f. Approximate size of boxes
20 ft. x 35 ft. x 21 ft. high (Carrollton Plant)
Siz ing and Cos ts:
a. Total carbon contactor volume
Carrollton Plant 278,520C.F. .
Algiers Plant 29,710 C.F.
TOTAL 308,230 C.F.
b. Bed areas
Carrollton Plant 27,852ft2
Algiers Plant 2,971 ft2
c. Surface loading 3 . 74 g.p.m./ft2
-------
GANNETT FLEMING CORDDRY AND CARPENTER. INC.
TABLE 3
(Cont'd.)
Sizing and Costs: (Cont'd.)
d. Number of carbon contactors
Carrollton Plant
Algiers Plant
e. Unit cost approximation for Carolton Plant:
1. Excavation and backfill
2 . Cast-in-place concrete piles
3. Concrete, Class A
4. Reinforcing steel
5. Superstructure
6. Miscellaneous metals
7. Filter equipment
8. Filter bottom, gravel and sand
9. Inside piping
10. Instrumentation and control valves
11. Dampproofing and painting
12
Heating, ventilation, and air
• conditioning
40
4
$ 5.00/ft2
84.00/ft2
115.00/ft2
51.00/ft2
20.00/ft2
4.00/ft2
10.00/ft2
13.00/ft2
51.00/ft2
122.00/ft2
9.00/ft2
of Contactor
of Contactor
of Contactor
of Contactor
of Contactor
of Contactor
or Contactor
of Contactor
of Contactor
of Contactor
of Contactor
13. Electrical work
Subtotal
Contingencies
Total
f. Cost of contactors:
Carrollton Plant 27,842 ft2 -x $580.
Algiers Plant* 2 , 971 ft2 x $580 .
TOTAL**
* Unit cost developed for Carrollton Plant used to
Algiers Plant
** Contingency factor applied to unit cost item.
6.00/ft2 of Contactor
16.00/ft2 of Contactor
$506.00/ft2 of Contactor
74.00
$580.00/ft2 of Contactor
00/ft2 = $16,154,160.00
00/ft2 = • 1,723,180.00
$17,877,340.00
determine costs for
-------
GANNETT FLEMING ' CORDDRY AND CARPENTER. INC.
TABLE 4
REGENERATION FURNACES
Key Assumptions:
a. Regeneration cycle - 60 days
b. Multiple hearth type units
c. Furnace costs include instrumentation, dewatering and quench
tanks, pollution control equipment, installation, startup,
and testing
d. Foundation conditions will require a pile and concrete slab
support system
e. Units will be housed to increase furnace life
f. Housing costs include architectural treatment, ventilation,
plumbing, and electrical work
g. Furnace capacity criteria - flow equals average day of maximum
month:
1. 60-day regeneration cycle with one furnace out of service
2. 45-day regeneration cycle with all units operating
h. Carbon regeneration:
Carrollton Plant 10, 800 Ibs/day
Algiers Plant 13,800 Ibs/day
Total " 18,400 Ibs/day
1. 60 days at average annual flow 101,200 Ibs/day
2. 60 days at average daily flow of
" maximum month 129,600 Ibs/day
3. 45 days at average daily flow of
maximum month 172,800 Ibs/day
i. Spent carbon loading rates - 90 to 120 Ibs/day/ft2 of
hearth area
-------
GANNETT FLEMING COROORY AND CARPENTER. INC.
TABLE 4
(Cont'd.)
Key Assumptions;- (Cont'd.)
j. Number of furnaces
Carrollton Plant
Algiers Plant
k. Wastewater from scrubbers on furnaces will be
discharged into existing wastewater systems
Sizing and Costs:
Carrollton Plant:
Furnace:
3 - 32 .5 ton/day units
@ $1,375,000.00 =$4,125,000.00
Foundation:
45 ft. x 12 ft. concrete
base on piles
Housing facilities:
45 ft. x 120 ft. x 50 ft.
high
Algiers Plant:
Furnace:
2-7 ton/day units
@ $850,000.00
Foundation:
30 ft. x 60 ft.
Housing facilities:
30 ft. x 60 ft. x 50 ft
high
110,000.00
475,000.00
= $1,700,000.00
75,000.00
= 145,000.00
Subtotal
Contingencies
TOTAL
3
2
$4,710,000.00
1,920,000.00
$6,630,000.00
995,000/00
$7,625,000.00
-------
GANNETT FLEMING CORDORY AND CARPENTER. INC.
TABLE 5
FILTERED WATER TRANSFER' SYSTEM
Key Assumptions:
a. Hydraulic capacity (rated plant capacity)
Cairollton Plant 237 m.g.d.
Algiers Plant 16 m.g.d.
b. Pumping capacity - hydraulic capacity with largest unit out of
service
c. Number of transfer pump stations
Carrollton Plant - 2 (each station handles one-half the flow)
•s, :
Algiers Plant - 1
d. Number of pumps in each transfer pump station
1. Carrollton Plant - 5 vertical mixed-flow units; 2 units
equipped with variable-speed drives
2. Algiers Plant - 3 vertical mixed-flow units; 2 units equipped
with variable-speed drives
e. Costs include transfer pump facilities; piping between existing
filtered water piping and transfer pump facilities; piping to
carbon contactors gallery; and piping from the carbon con-
tactor gallery to connections with high-service pump sta-
tion inlet piping
f. Costs include piping tor backwash of carbon contactors
g. Wastewater from backwashing will be discharged into existing
wastewater piping
h. • Costs for transfer pump facilities include instrumentation,
metering, controls to match pumping of high-service units,
reinforced-concrete wet well, pile foundation, architectur-
ally treated superstructure, plumbing, ventilating, and
electrical work.
-------
GANNETT FLEMING CORODRY AND CARPENTER. INC.
TABLE 5
(Cbnt'd.)
Costs;
Carrollton Plant:
a. Pumping station
2 stations at $1,750,000.00 $3,500,000.00
b. Outside piping 1,650,000.00
$5,150,000.00
Algiers Plant:
a. Pumping station $ 800,000.00
b.. Outside piping 350,000.00
1, 150,000.00
Subtotal $6,300,000.00
Contingencies 945,000.00
TOTAL $7,245,000.00
-------
GANNETT FLEMING CORODRY AND CARPENTER. INC.
TABLE 6
CARBON TRANSFER EQUIPMENT
Key Assumptions:
a. Carbon transfer equipment includes all piping and mechanical items
between the GAG contactor and the top of dewatering bin above
the furnace and between the furnace quench tank and the GAG
contactor.
b. Each GAG contactor contains its own carbon transfer system
equipment.
c. Wastewater from carbon transfer facilities will be discharged into
into the existing wastewater piping system.
Carrollton Plant:
Transfer equipment (includes piping,
pumps, specialty items, etc.) $ 900,000.00
Additions to GAG contactors 240,000.00
Algiers Plant:
Transfer equipment 100,000.00
Additions to GAG contactors 60,000.00
Subtotal $1,300,000.00
Contingencies 195,000,00
TOTAL' $1,495,000.00
-------
GANNETT FLEMING COROORY AND CARPENTER. INC.
TABLE 7
BUFFER CARBON STOCK
Key Assumptions:
a. 7% of initial fill
b. Unit weight of GAG 28 lbs/ft3
c. Cost of GAG $0.55 per Ib
Sizing and Cost:
Volume:
CairoUton Plant:
278,520 ft3 x 0.07 = 19,496 ft3
Algiers Plant:
29,710 ft3'x 0.07 = 2,080 ft3
Carbon Cost:
Carrollton.Plant:
19,496 ft3 x 28 Ibs/ft3 x $0.55/lb =$300,238.00
Algiers Plant:
2,080 ft3 x 28 Ibs/ft3 x $0.55/lb = 32,032.00
- TOTAL $332,270.00
-------
GANNETT FLEMING CORDDRY AND CARPENTER, INC.
TABLE 8
BUFFER CARBON STORAGE FACILITIES
Key Assumptions:
a. Storage tanks - aboveground steel tanks
b. Capacity of each tank
C arrollton P lant 20,000 ft3
Algiers Plant 2,000 ft3
c. Number of tanks at each plant 2
d. Handling facilities for loading and discharging makeup carbon
are included in the cost of Buffer Carbon Storage Facilities.
Sizing and Costs:
Carrollton Plant:
2 storage tanks
Accessories
Foundations
Algiers Plant:
2 storage tanks
Accessories
Foundations
Subtotal
Contingencies
TOTAL
$200,000.00
100,000.00
40,000.00
$340,000.00
$110,000.00
80,000.00
30,000.00
$220,000.00
$560,000.00
84,000.00
$340,000.00
$644,000.00
-------
(jANNETT t-UEMING I—UKUUKT Anu
TABLE 9
LOCAL ITEMS
Key Assumptions:
a. Demolition of existing structures at Carroilton Plant taken from the
cost estimates developed by the Sewerage and Water Board.
b. Costs for property purchase at Algiers Plant taken from summary
tables in the Sewerage and Water Board cost analysis.
c. Chlorination facilities were sized for feeding 1.5-p.p.m.
d. Chlorine contact basins will be constructed of reinforced concrete
and will be constructed underground.
e.. Chlorine contact time - 20 minutes.
f. Six water storage tanks-re'quired to replace treated water storage
facility removed to make room for carbon contactors.
g. Type of water storage tanks - aboveground, steel.
h. Capacity of each steel water storage tank - 3.3 million gallons.
Costs:
Demolition of existing structures $1,200,000.00
Purchasing property 830,000.00
TOTAL $2,030,000.00
Chlorination facilities: -
Carroilton Plant:
Chlorination building $ 70,000.00-
Railroad siding 20,000.00
Miscellaneous items (elec-
trical, instrumentation,
accessories, etc.) 30 ,000 . 00
$ 120,000..00
-------
GANNETT FLEMING CORODRY AND CARPENTER. INC.
TABLE 9
(Cont'd.)
Costs: (Cont'd.)
Chlorination facilities: (Cont'd.)
Algiers Plant:
Chlorination building $ 40,000.00
Miscellaneous items (elec-
trical, instrumentation,
unloading facilities,
accessories, etc.) 20,000.00
$ 60,000.00
Subtotal $ 180,000.00
Contingencies 27,000.00
TOTAL $ 207,000.00
Chlorine contact basins:
Carrollton Plant $1,270,561.00
Algiers Plant 238,824.00
Subtotal $1,509,385.00
Contingencies 226,350.00
TOTAL 1,735,735.00
Six steel, finished water storage
tanks (Carrollton Plant):
Cost per tank - $600,000.00
Six tanks $3,600,000.00
Contingencies 540,000.00
TOTAL 4,140,000.00
-------
GANNETT FLEMING CORDDRY AND CARPENTER. INC.
TABLE 9
(Cont'd.)
Summary Costs of Local Items:
Demolition
Property Purchase
Chlorination facilities
Chlorine contact basins
Six steel, finished water
storage tanks
TOTAL
$1,200,000.00
830,000.00
207,000.00
1,735,735.00
4,140,000.00
$8,112,735.00
-------
GANNETT FLEMING CORDDRY AND CARPENTER. INC.
TABLE 10
ALTERNATE DESIGNS OF CARBON CONTACTORS
Key Assumptions:
Same assumptions as outlined in TABLE 3, except as follows:
a. Surface Loading
b. Carbon contactor bed depth
c. Number of carbon contactors:
Carrollton Plant
Algiers Plant
d. Approximate size of contactors
(Carolton Plant)
5 g.p.m
13.4 ft.
30
3
./ft2
20 ft. x 35 ft. x 25.5 ft. high
Unit cost approximation for Carrollton Plant:
1. Excavation and backfill
2 . Cast-in-place concrete piles
3. Concrete, Class A
4. Reinforcing steel
5. Superstructure
6. Miscellaneous metals
7. Filter Equipment
8. Filter, gravel and sand
9 . Inside piping .-,,_
10. Instrumentation and
control valves
11. Dampproofing and painting
12. Heating, ventilation, and
air conditioning
13
Electrical Work
Subtotal
Contingencies
TOTAL
$ S.00/ft2 of Contactor
102.00/ft2 of Contactor
140.00/ft2 of Contactor
62.00/ft2 of Contactor
20.00/ft2 of Contactor
4.00/ft2 of Contactor
12.00/ft2 of Contactor
13.00/ft2 of Contactor
56r.00/ft2 of Contactor
. 126.00/ft2 of Contactor
10.00/ft2 of Contactor
6.00/ft2 of Contactor
16.00/ft2 of Contactor
$572.00/ft2 of Contactor
86.00
$658.00/ft2 of Contactor
-------
GANNETT FLEMING CORDDRY AND CARPENTER. INC.
TABLE 10
(Cont'd.)
Sizing and Costs:
Carroll-ton Plant 20,833 ft2 x $658.00/ft2 = $13,708,114.00
Algiers Plant* 2,222 ft2 x $658.00/ft2 = 1,462,076.00
TOTAL** $15,170,190.00
* Same unit cost developed for Carrollton Plant used to determine costs
for Algiers Plant.
** Contingency factor applied to unit cost item.
-------
APPENDIX C
CAPITAL MARKETS
RATIO ANALYSIS FOR SELECTED
WATER SYSTEMS
-------
APPENDIX C
The analysis of the ability of water systems to obtain
capital financing as presented in Chapter IV is based upon
ratio analysis of selected water systems. The financial ratios
studied for those cities are presented in this appendix.
The cities which were studied are those listed in EPA's
January 25, 1978 press release announcing the proposed organics
regulation. Adequate data for detailed analysis was obtained
for 27 of the 29 systems listed then as potentially violating
the proposed regulation.
As EPA noted at the time, some of these cities might not
actually require additional treatment to comply with the pro-
posed regulation. Also, other cities as yet unidentified may
be affected by the regulation.
Finally, this analysis is based upon the costs of granular
activated carbon treatment as the most expensive treatment and
therefore the most rigorous financing challenge, although not
all of these systems would actually be expected to require GAG
treatment.
The ratios .chosen for analysis are those regularly used
by financial analysts in the bond rating field. The ratios
used to examine revenue bonds and self-supporting general
obligation bonds include:
• Operating ratio: operating and maintenance
expenses (excluding depreciation) divided by
total operating revenues.
• Net take-down: net revenues (gross revenues
less operating and maintenance expenses)
divided by system gross revenues.
• Interest coverage: net revenues available
for debt service divided by interest require-
ments for the year. This ratio indicates
the ability of the utility to meet its fixed
charges. A high level of this ratio is an
indicator of financial strength.
-------
C-2
• Debt service coverage: net revenues available
for debt service divided by principal plus
interest requirements for the year. Bond
holders prefer a high level for this ratio
which indicates the ease or probability of
timely payment of obligations.
• Debt service safety margin: system gross
revenues less operating and maintenance ex-
penses and less current debt service divided
by system gross revenues. A high value of
this ratio is favorable.
• Debt ratio: net debt (gross debt as shown
on balance sheet less bond principal reserve)
divided by sum of net utility plant plus net
working capital. This ratio indicates the
degree to which the firm is leveraged. The
higher the ratio, the less the opportunity
to carry additional debt.
For general obligation bonds (bonds secured by the full
faith and credit of the entity) the relevant measures include
• Debt per capita: determined on a direct
basis (only the debt of the municipality)
and on an overall basis (including the
debt of overlapping entities such as
school boards).
• Debt to assessed value: determined on the
basis of full market value. This measure
is also calculated on a direct and overall
basis.
These ra'nges of values (including quartiles) for these
measures among water utilities are included in Table A-l.
-------
C-3
Table C-l
RANGES OF VALUES OF KEY FINANCIAL RATIOS
FOR MUNICIPALLY-OWNED WATER UTILITIES
(1975-1976 data)
Operating ratio
Net take-down
Interest coverage
Debt service coverage
Debt service safety margin
Debt ratio
Source: Moody 's Investors
Low
23.3
11.1
0.86
0.36
2.1
11.6
Service,
1st
Quartile
51.3
37.1
2.69
1.72
17.4
26.6
Inc.
Median
60.8
43.3
4.28
2.22
23.8
39.9
3rd
QuartiTe
69.5
52.3
5.1S
3.06
31.3
58.5
High
90.7
77.6
14.36
19.31
59.1
94.3
For investor-owned utilities, the following measures
were examined:
• Times interest earned: operating profit divided
by interest charges. This ratio is calculated
both pre-tax and after-tax and gives an indica-
tion of the margin of safety which the company
has to meet its interest payments.
• Times interest and preferred dividends earned:
operating profit divided by interest charges
and preferred dividends. This ratio is cal-
culated after tax and provides an indication
-" oi the safety margin which income provides to
meet the company's fixed charges.
• Debt ratio: net debt (gross debt as shown
on the balance sheet less bond principal
reserve) divided by the sum of net utility
plant plus net working capital. This ratio
indicates the degree to which the firm is
leveraged. The higher the ratio, the less
the opportunity to carry additional debt.
• Return on equity: income available to common
stock ^divided by average common equity for
the year. This ratio provides an indication
of the ease of common stock financing with-
out suffering dilution.
-------
C-4
The levels of the relevant measures are shown in Tables
C-2, C-3 and C-4 for the 27 cities for the following cases:
••• Current: ratio analysis based on most recent
available data.
• Low-cost scenario: $150,000/MG average daily
production capital cost, $20,000/MG average
daily production O&M cost.
• High-cost scenario: $400,000/MG average daily
production capital cost, $10,000/MG average
daily production O&M. cost.
The current bond rating and capital cost requirements
are also detailed in these tables.
-------
C-5
Page 1 of 2
Table C-2
CREDIT RATIO ANALYSIS OF MUNICIPALLY-OWNED WATER SYSTEMS
WHICH NORMALLY FINANCE CAPITAL PROJECTS WITH
REVENUE OR SELF-SUPPORTING GENERAL OBLIGATION BONOS
(do 11 an in thousands)
Bond
Rating*
Anarillo, TX
Current Al
Low
High
Charleston, SC
Current A
Low
High
Cincinnati, OH
Current Aa
Low
High
Columbus, OH*
Current Aa
Low
High
Contra Costa, CA
Current Baa
Low
High
Dayton, OH*
Current Aa
-aw
High *
Fairfax, VA •
Current A
Low
High
Jackson, MS
Current Aa
Low
High
Jefferson Parish *1, LA
Current
Low
High
Jefferson Parish i2, LA
Current A
Low
High
Louisville, KY
Current Aa
Low
High
Melbourne, FL
Current A
Low
High
Miami, Fl
Current Al
Low
High
Montgomery, AL
Current Al
.ow
, High
1
Year
of Oata
1976
1975
1977
1977
197S
1975
1976
1975
1977
1977
1976
1977
1976
1977
Capital
Cost0
$ 3,500
$ 9,280
$ 4,300
$12,300
$11,340
$30,240
$15,150
$40,400
$ 3,750
$10,000
$ 5,625
$15,000
$ 3,790
$23,440
$ 4,260
$11,360
$ 6,960
$18,560
$ 3,716
$ 7,240
$18,300
$48,800
$ 1,875
$ 5,000
$10,305
$27,480
$ 4,200
$11,200
Operating
Revenues
$ 8,587
$ 9,063
$ 9,341
$ 7,430
$ 8,158
$ 9,030
$20,334
$21,941
$24,115
$16,775
$18,912
$21,827
$ 5,523
$ 6,111
$ 6,924
$ 7,033
$ 7,731
$ 8,809
$17,590
$18,936
$20,756
$ 4.347
$ 4,951
$ 5,767
$ 5,310
$ 6,366
J 7,815
$ 2,757
$ 3,168
$ 3,735
$23,128
$25,721
$29,230
$ 4,221
S 4.505
$ 4,342
$18,635
$20,172
$22,280
$ 8,360
$ 3,987
$ 9,345
Operating
Ratio
~TCT
54.5
54.2
52.0
37.1
38.0
37.6
76.5
74.4
70.3.
64.7
S2.9
59.2
50.0
58.3
45.0
38.4
34.0
77.7
37.4
37.3
37.4
55.0
54.4
51.5
63.4
60.0
55.0
78.3
74.2
67.3
59.3
58.5
55.5
53.3
53.0
52.0
65.4
53.3
60.8
50.3
52.7
51.0
Net
Take-Oown
nn
45.5
45.3
48.0
62.9
62.3
62.3
24.3
26.4
29.9
40.3
42.0
44.9
48.7
41.7
51.4
15.9
19.9
25.5
62.7
62.2
62.7
45.0
45.6
48.5
49.3
50.2
53.1
34.6
37.1
41.0
38.5
40.0
42.7
51.9
52.0
53.0
34.6
36.2
39.1
46.7
-.. 47.3
49.0
Interest
Coverage
21.1
9.7
5.6
7.3
5.2
4.0
7.2
4.3
2.9
6.6
4.3
3.0
3.3
2.8
2.4
.
„
-
2.2
2.1
1.9
8.2
4.3
4.1
4.0
3.0
2.3
2.0
1.8
1.6
2.1
1.9
1.3
2.6
2.4
2.2
7.5
4.7
3.1
5.3
4.2
3.2
Debt
Service
Coverage
4.3
3.4
2.9
3.8
3.1
2.6
2.2
1.9
1.6
4.9
3.3
2.3
1.9
1.7
1.6
1.9
1.5
1.3
1.3
1.7
1.6
2.2
1.9
1.6
2.4
2.0
1.6
1.4
1.3
1.2
2.1
1.8
1.6
2.3'
2.1
1.9
2.7
2.2
1.9
2.1
1.9
1.7
Debt
Service
Safety
Margin
34.9
32.0
31.0
46.2
42.0
38.0
13.4
12.5
11.4
32.5
29.1
26.5
22.4
20.5
18.3
7.4
5.7
5.9
28.0
25.7
23.4
19.9
21.7
13.6
28.7
24.7
20.8 '
9.5
8.4
' 7.1
20.1
18.2
16.0
29.0
27.3
24.6
21.7
20.1
18.0
24.2
22.5
20.5
Oebt
Ratio
HT
11.5
18.1
27.2
25.0
31.0
39.6
-
.
-
-
.
-
43.1
50.0
52.3
.
.
-
73.0
74.0
76.0
15.7
25.8
33.0
46.1
53.9
64.9
50.0
55.2
62.0
43.0
49.0
56.7
57.3
59.9
63.5
20.4
27.3
37.5
23.5
27.9
34.2
(continued)
-------
C-6
Table C-2
Page 2 of 2
Bond " Year
Rating* of Data
New Orleans^ LA
Current A» 1977
Low
High
Norfolk, VA*
Current Aa* 1977
Low
High
Omaha, NEB
Current Aa 1976
Low
High
Passaic Valley, Nd«-
Current A 1976
Low
High
Philadelphia, PA
Current A 1976
Low
High
Phoenix, AR*
Current Aa 1977
Low
High
Tampa, Ft.
Current A 1975
Low
High
Tooeka, KS
Current Al 1976
Low
High
Waterbury, CT*
Current 8AA1 1977
Low
High
Capital Operating Operating Net Interest
Cost*8 Revenues Ratio Take-Gown Coverage
$11,459 77.8 . 23.3 7.1
$19,400 $14,205 71.8 29.2 2.7
551,600 . $17,900 64.2 36.3 1.9
J 9,6H 51.8 38.Z 22.6
$ 7,337 $10,724 60.3 39.7 6.7
$20,880 $12,224 57.2 42.3 3,7
$13,819 51.9 49.5- 6.5
$11,040 $15,383 51.4^ 47.2 4.2
$29,440 $17,500 50.6. 49.4 3.1
$ 9,167 65.2 32.8 6.9
$ 7,762 $10,345 63.0 38.0 3.8
$20,697 $11,936 59.0 42.0 2.5
$68,534 57.2 42.3 1.7
$36,150 $74,018 56.3 43.7 1.6
$96,400 $81,553 54.3 46.0 1.6
$31,942 47.1 52.9 3,9
$18,240 $34,527 46.1 53.3 3.4
$48,640 $33,025 46.0 54.0 2.3
$ 6,953 65.6 33.5 3.4
$ 7,620 $ 8,114 S2.S 39.3 2.5
$20,320 $ 9,701 57.5 43.8 2.0
/
$ 5,489 53.3 46.7 5.9
$ 3,870 $ 6,066 53.0 47.0 4.2
$10,320 $ 6,857 50.2 49.3 3.0
$ 2,528 45.5 41.5 ~ 3.0
$ 2,940 $ 2,989 41.5 43.8 2.4
$ 7,840 $ 3,626 39.8 54.3 2.0
Self-supporting general obligation bond
J8ond ratings listed in this table in order of decreasing financial strength are: AAA, AA, Al, A, BAA1
H"he cost assumptions employed in this analysis are meant to be illustrative of the wide range of costs
incurred by affected utilities; the low cost and high cost assumptions do not refer to specific design
For specific explanation of cost assumptions, see page C-4.
Debt
Debt Service
Service Safety
Coveraae Margin
T?r~
3.2 16.0
1.8 13.0
1.4 10.4
2.4 22.4
2.0 20.1
1.7 17.6
4.8 39.0
3.2 32.5
2.4 28.S
2.0 19.8
1.7 18.0
1.5 1S.O
1.3 10.1
1.3 9.4
1.2 8.5
2.4 30.9
2.2 28.9
1.9 26.0
1.7 15.3
1.5 13.2
1.3 11.2
3.8 34.2
2.9 31.0
2.2 27.4
1.3 12.7
1.3 10.9
1.2 9.1
, BAA.
which might be
related scenarios.
Debt
Ratio
1ST
13.5
34.9
53.9
7.2
16.5
28.5
21.0
30.1
41.2
75.1
83.0
39.0
45.0
45.5
49.1
54.0
67.4
71.8
fttf
661
74.4
43.0
52.8
63.3
28.3
36.0
45.2
-------
C-7
Moody 's
Rating
OES MOINES. IA
Current Aaa
Law
High
Median
The cost assumptions employed
incurred by affected utilities
Table C-3
CREDIT RATIO ANALYSIS OF MUNICIPALLY-OWNED WATER SYSTEMS
WHICH NORMALLY FINANCE WITH GENERAL OBLIGATION BONOS
Year Capital Debt Per Capita Debt Per Assessed Value
of Data Cost* Sirect Overall Direct
(JOT) (S) .(5) " (i)
1977 $ 241 $ 334 2.3
S 5,565 $ 268 $ 362 2.5
$14,340 $ 314 $ 408 3.0
S 299 $ 424 3.5
in this analysis are meant to be illustrative of the wide range of
; the low cost and high cost assumptions do not refer to specific
Overall
3.2
3.5
3.9
4.3
costs which might be
design related scenarios.
Moody 's
' Rating
AMERICAN WATER WORKS CO.
Current —
Low
High
INDIANAPOLIS WATER CO.
Current A
Low
High
WESTERN PENNA WATER CO.
Current
Low
High
The cost assumptions employed
incurred by affected utilities
Table C-4
CREDIT RATIO ANALYSIS OF PRIVATELY-OWED WATER SYSTEMS
Times Times Times Interest
Year Capital Interest Earned Interest Earned and Preferred Oiv. Debt Return
of Data Cost* Pre Tax After Tax Earned Ratio on Equity
1976 1.33 1.61 1.51 25.0 9.3
$ 3,250 1.38 1.55 1.55 ZS.Z 9.3
$22,000 1.90 1.67 1.57 32.9 9.3
1977 4.65 2.45 2.14 45.5 16.3
$15,450 4.37 2.35 2.10 47.3 15.3
$41,200 4.09 2.24 2.05 • 50.3 16.3
1976 2.19 1.97 1.73 50.2 8.2
$12,165 2.15 1.94 1.73 52.5 8.2
$32,440 2.08 1.38 1.71 56.9 3.2
in this analysis are meant to be Illustrative of the wide range of costs which aright be
; the low cost and high cost assumptions do not refer to specific design related scenarios.
, , _
-------
APPENDIX D
ORGANIZATIONS CONTACTED
DURING THIS REVIEW
-------
APPENDIX D
The following organizations have been contacted by
TBS and ERGO in the period of April-June 1978 during the
re-examination of the economic and financial implications
of the proposed organics regulation.
Bankers Trust Company
Black and Veatch
Calgon
Carborundum
Gulp, Wesner & Gulp
Envirotech
Gannett "Fleming Corddry and Carpenter, Inc.
Indianapolis Water Company
Louisville Water Company
Moody's Investor Service, Inc.
Neptune Nichols
Philadelphia Water Department
Sewerage & Water Board of New Orleans
Shirco
State Street Bank
-------
APPENDIX B
ESTIMATED COST FOR
GRANULAR ACTIVATED CARBON (GAC)
FACILITIES—NEW ORLEANS WATER
TREATMENT PLANTS
-------
APPENDIX S
In order to reconcile the differences between EPA's
GAG cost estimates and those provided by the New Orleans
Sewerage and Water Board, TBS subcontracted with Gannett
Fleming Corddry and Carpenter, Inc. to provide an indepen-
dent preliminary capital cost estimate. The GFC&C, Inc. re-
port detailing the resultant construction and project cost
estimates is reproduced in its entirety in this Appendix.
-------
ESTIMATED COSTS
FOR
GRANULAR ACTIVATED CARBON (GAC) FACILITIES
NEW ORLSfiNS WATER TREATMENT PLANTS
June 1978
Prepared by
GANNETT FLEMING CORDDRY AND CARPENTER, INC .
Harrisburg, Pennsylvania
------- |