EPA-670/2-74-068
AUGUST 1974
Environmental Protection Technology Series
                                  AN ASSESSMENT OF
                                        WET SYSTEMS
            FOR  RESIDENTIAL REFUSE  COLLECTION:
                                      Summary Report
                                  National Environmental Research Center
                                    Office of Research and Development
                                   U.S. Environmental Protection Agency
                                            Cincinnati, Ohio 45268

-------
                                     EPA-670/2-74-068
                                     August 1974
        AN ASSESSMENT OF WET SYSTEMS

     FOR RESIDENTIAL REFUSE COLLECTION:

               Summary Report
                     By

                 P. M. MEIER
                  J. KUHNER
                R. E. BOLTON
           Curran Associates, Inc.
      Northampton, Massachusetts  01060
           Contract No. 68-03-0183
         Program Element No. 1DB064
              Project Officers

              Donald Oberacker
Solid and Hazardous Waste Research Laboratory
                     and
                Harry Bostian
Advanced Waste Treatment Research Laboratory
   National Environmental Research Center
           Cincinnati, Ohio  45268
   NATIONAL ENVIRONMENTAL RESEARCH CENTER
     OFFICE OF RESEARCH AND DEVELOPMENT
    U.S. ENVIRONMENTAL PROTECTION AGENCY
           CINCINNATI, OHIO  45268

-------
                    REVIEW NOTICE






     The National Environmental Research Center—




Cincinnati has reviewed this report and approved its




publication.  Approval does not signify that the




contents necessarily reflect the views and policies




of the U.S. Environmental Protection Agency, nor




does mention of trade names or commercial products




constitute endorsement or recommendation for use.
                          ii

-------
                             FOREWORD
     Man and his environment must be protected from the
adverse effects of pesticides, radiation, noise and other
forms of pollution, and the unwise management of solid
waste.  Efforts to protect the environment require a focus
that recognizes the interplay between the components of our
physical environment—air, water, and land.  The National
Environmental Research Centers provide this multidisciplinary
focus through programs engaged in

     •  studies on the effects of environmental contaminants
        on man and the biosphere, and

     •  a search for ways to prevent contamination and to
        recycle valuable resources.

     This report presents a summary of research directed toward
development of utilizing existing sanitary sewer systems to
transport ground solid waste.  Individual subcomponents were
designed, tested, and the overall economic feasibility of this
approach was determined.
                             A. W. Breidenbach, Ph.D.
                             Director
                             National Environmental
                             Research Center,  Cincinnati
                               iii

-------
                                ABSTRACT


This summary is a synopsis of a series of four technical reports evalu-
ating the technical and economic feasibility of wet systems for residen-
tial refuse collection prepared by Curran Associates,  Inc., under Contract
68-03-0183 to the EPA Solid Waste Research Laboratory, Cincinnati, Ohio.

The most promising wet system alternative is identified as a system using
individual household grinders in low density areas, vacuum collection and
neighborhood grinders in high density areas, dilute slurry transport of
ground refuse in the existing sanitary sewer system, and joint treatment
of refuse and sewage at an expanded treatment facility that includes
anaerobic digestion for methane generation.  However,  the economic
feasibility of even the most promising alternative is  doubtful because of
the high cost of grinding, and hydraulic transport of  ground refuse in
existing sewer systems may be feasible only if both metals and glass are
excluded.  A conventional collection of the nongrindable and bulky consti-
tuents of residential refuse will still be needed, albeit of a diminished
frequency.  The total 1973 cost of such a wet system is estimated at
$113-196 per household per year, of which $80-$105 is  for grinding.  This
compares to a total cost for refuse and sewage collection and disposal
under existing concepts of up to $115 per household per year.  Projections
for 1988 indicate that the wet system cost might rise  to $130-$228,
against a range of $61-$143 under conventional concepts.  However, if the
cost of the grinder is excluded (on the basis that individuals would
purchase grinders for their convenience benefits), the 1973 wet system
costs (to the public sector) would be only $33-$91, thus possibly less
than existing public sector costs of about $43-$115 per household per year.
                                   iv

-------
                                CONTENTS
                                                                   Page
Abstract                                                             iv
List of Figures                                                      vi
List o£ Tables                                                     vii
Acknowledgments                                                    viii
Sections
   I  Conclusions                                                     *
  II  Recommendations                                                 ^
 III  Introduction                                                    ^
  IV  Grinding                                                        9
   V  Collection and Disposal of Nongrindables                       18
  VI  Hydraulic Transport                                            23
 VII  Treatment                                                      30
VIII  Sludge Handling                                                46
  IX  Environmental and Socioeconomic Impacts                        63
   X  Total System Cost Analysis                                     79
  XI  Recommendations for Further Research                           82
 XII  References                                                     86
XIII  Publications                                                   92
                                    v

-------
                            FIGURES


No._

 1   Preliminary Design for a Household Grinder                  1°

 2   Basement Installation of a Household Grinder                H

 3   Limiting Refuse Concentrations for Hydraulic                26
     Transport in Sewers

 4   Grinding Water Requirement                                  27

 5   Existing Process Configuration at the Springfield           33
     Wastewater Treatment Plant

 6   Planned Process Configuration at the Springfield            35
     Treatment Plant

 7   Computer Model Representation of the Recommended            40
     Treatment Process Configuration for Ground Refuse

 8   Dry Sludge Solids as a Function of Population Size          48
     and Refuse Generation

 9   Flow Diagram of Sewage Treatment Plant Handling             49
     100 gal/cap/day

10   Flow Diagram of Sewage Treatment Plant Handling             50
     100 gal/cap/day plus 2.25 Ib/cap/day

11   Sludge Handling, Alternative 1                              54

12   Sludge Handling, Alternative 2                              55

13   Sludge Handling, Alternative 3                              56

14   Total Cost of Sludge Handling (Alternative 1, 2 and 3       58
     as a Function of Population; for Average Refuse Generation
     Rate of 2.25 Ib/cap/day
                                vi

-------
                             TABLES
No.                                                            Page

 1   Cost Estimates for Installations of Refuse  Grinding         13
     System

 2   Cost Comparisons Between the Foster-Miller  and              14
     Meier-Nebiker Concepts

 3   Estimated Rates of Flow and Strengths of Wastewater         34
     from All Contributors

 4   Present and Projected Waste Concentrations                  37

 5   Seasonal Variations Due to Ground Refuse Addition           41

 6   Treatment Costs for Sewage Alone; Planned Bondi             43
     Island Treatment Plant

 7   Treatment Costs for Sewage Containing Ground Domestic       44
     Refuse

 8   Total Cost of Sludge Handling  (Alternative  1) in $/ton      59
     Dry Solids for Various Populations and Various Raw
     Refuse Generation Rates

 9   Total Cost of Sludge Handling  (Alternative  2) in $/ton      60
     Dry Solids for Various Populations and Various Raw
     Refuse Generation Rates

10   Total Cost of Sludge Handling  (Alternative  3) in $/ton      61
     Dry Solids for Various Populations and Various Raw
     Refuse Generation Rates

11   Total Annual System Cost per Household Comparisons for      80
     Springfield
                               vii

-------
                             ACKNOWLEDGMENTS


This study was conducted by Curran Associates, Inc., Engineers and Plan-
ners, Northampton, Massachusetts, under the direction of Dr. Peter M.
Meier.  Dr. Roger Bolton and Mr. Jochen Kuhner played major roles in the
conduct of the study and other contributions are due to Messrs. James
Martel, Don Cooper, George Fisette and Michael Joyce.

The contribution to the study of a Review Panel was substantial, both in
review of the early drafts of this report and in the opening of many in-
teresting perspectives during the five discussion meetings held at peri-
odic intervals in the course of this study.  The panel members consisted
of Mr. Donald Oberacker and Dr. Harry Bostian of the EPA National Envi-
ronmental Research Center, Cincinnati, Ohio; Professor Bernard B. Berger,
Director of the Massachusetts Water Resources Research Center; Dr. Roger
Bolton, Professor of Economics at Williams College; Mr. Thomas G. Sanders,
independent consultant and one of the initiators of research interest
into Wet Systems for Refuse Collection; Dr. Donald D. Adrian, Professor
of Civil Engineering, University of Massachusetts; Mr. Herbert B. Wyndham,
Jr., Vice President, Malcolm Pirnie, Inc., New York; Mr. Adi Guzdar,
Manager, Engineering Studies Division, Foster-Miller Associates, Waltham,
Massachusetts; Dr. Iraj Zandi, Professor of Civil and Mechanical Engi-
neering, University of Pennsylvania; and Mr. Wylie C. Hubbard, Superin-
tendent of the Department of Public Works, City of Springfield, Massa-
chusetts.   Dr. Boyd Riley and Dr. Rudi Schmut, independent consultants,
Mr. Al Fisk of Foster-Miller Associates, and Mr. Oscar Albrecht of the
EPA National Environmental Research Center in Cincinnati also attended
some of the panel meetings or gave substantial assistance.

The writers especially acknowledge the assistance of Mr. Wylie C. Hubbard,
Superintendent of the Department of Public Works, City of Springfield,
Massachusetts, and his staff, without whose cooperation the case study
could not have been satisfactorily conducted.  In particular, we acknowl-
edge the contribution of Mr. Lawrence J. Dowd, Deputy Superintendent of
the Division of Solid Wastes, who provided major assistance in the collec-
tion of data and information.  Others whose assistance we acknowledge in-
clude Mr. Paul Sullivan, Chief Foreman, and Mr. Donald Ferris, Foreman,
both of the Division of Solid Waste, Springfield; in the Springfield Sewer
Division, Mr. James Moyland, Deputy Superintendent, Mr. John Risinski,
Chief Sewage Treatment Plant Operator, and Mr. Fred Mascole, Working
Foreman; in the Engineering Division, Mr. Joseph Cote, Office Engineer,
and Mr. Tony Masuch, Engineer; Mr. George E. Sweeney, Manager of the
Springfield Water Department; Mr. John Godfrey, Civil Engineer in the
Springfield Department of Public Works; Mr. Edward Beaudry in the person-
nel office of the City of Springfield; Mr. James Fitchet, Chief Accountant
of the Department of Public Works, Springfield.  Finally, Mr. David
Goodman of Harry Goodman, Inc., Springfield, Massachusetts, provided
valuable information on recycling.
                                    viii

-------
The authors benefited greatly from the comments and suggestions of
the EPA Project Officers, Mr. Donald'Oberacker and Dr. Harry Bostian,
and the assistance of Mr. Robert Smith and Mr. Richard Eilers, also
of the EPA National Environmental Research Center in Cincinnati, was
invaluable in the development of the treatment plant computer model.

These many contributions notwithstanding, the responsibility for the
accuracy and opinions expressed in this report is solely that of the
writers.
                                 ix

-------
                                SECTION I

                               CONCLUSIONS


The major conclusions of this case study can be summarized as follows:

     1.  Implementation of a wet system for residential refuse col-^
         lection in the case study area of Springfield, Massachusetts,
         or any other urban area, is at least a decade distant.

     2.  The ability of existing gravity sewer systems to successfully
         transport the glass and ceramics fraction of refuse remains
         questionable.  Maximum flow velocities encountered in many
         sewer systems are below the necessary scour velocities to
         prevent excessive deposition of ground refuse.  Metals must
         definitely be excluded.  The frequency of collection of non-
         grindables is likely to be higher than might be expected
         because of seasonal variations and the substantial quanti-
         ties of bulky wastes not amenable to grinding.  The adequacy
         of available grinding water without substantial additions of
         fresh water is also questionable.

     3.  To the extent that realistic cost estimates can be made in
         the face of the technical uncertainties that still exist
         in the Springfield, Massachusetts case study area would rise
         from $39 per year to $110-$140 per year under a wet systems
         implementation.  However, if the cost of the grinder is ex-
         cluded, costs tp the public sector under the wet system
         might decrease to as low as $35-$40 per year,

         A consideration of national conditions is more favorable
         to the wet system, showing a total cost of $113-$196 per
         household per year, of which $80-$105 is for grinding.
         This compares to a total cost for refuse and sewage collec-
         tion and disposal under existing concepts of up to $115 per
         household per year.  Projections for 1988 indicate that the
         wet system cost might rise to $130-$228, against a range of
         $61-$143 under conventional concepts.

     4.  Given the premise of technical feasibility, any implemen-
         tation that does occur is likely to be gradual, not unlike
         the spread of garbage grinders and trash compactors, and
         will depend on the development of a domestic grinder pur-
         chased by individual households at a cost consistent with
         consumer acceptance and its perceived domestic convenience
         benefits.  Existing grinding concepts are not regarded as
         practically feasible.

-------
5.  Analysis of the socioeconomic impacts of a wet system
    implementation shows the resultant reduction of the refuse
    collection labor force in Springfield to be a minor prob-
    lem, due to other employment opportunities within the
    public works sector and because of the gradual nature of ,
    system implementation.

6.  The most important impacts of ground refuse additions on
    wastewater treatment facilities will be on preliminary,
    primary and sludge handling facilities.  Conventional
    secondary and advanced waste treatment processes do not
    appear significantly affected.   The major uncertainty re-
    lates to the possibility of introduction of toxic,  nonbio-
    degradable compounds (mercury,  chlorinated hydrocarbons,
    etc.).

7.  The presence of numerous combined sewers in many urban
    areas, including the Springfield, Massachusetts case study
    area, is a serious barrier to the wet systems concept.
    This problem apart, our investigations show that an appro-
    priately modified treatment facility can produce effluents
    of comparable quality to conventional effluents at plants
    treating only sewage.

8.  The cost of dewatering and disposal of the sludge
    generated at a sewage treatment facility handling ground
    refuse is in the range of $15 to $25 per ton of dry solids,
    and substantially below that of a treatment facility
    handling only sewage solids.  The major determinants of
    total per ton cost are per capita refuse generation rates
    and the number of persons served by the facility.  Economies
    of scale and the predominance of a more readily dewaterable
    primary sludge are the major reasons for the relatively low
    unit costs.  The sludge handling facilities of a plant
    handling an average residential refuse generation rate cor-
    respond to the size of the facilities handling only sewage
    sludge generated by 9 to 12 times the number of inhabitants.

-------
                              SECTION II

                           RECOMMENDATIONS
The presently available factual evidence is still insufficient to make
definitive judgments on the technical feasibility of the wet systems
concept.  This study has therefore identified and recommended a se-
quence of bench-scale investigations designed to resolve these uncer-
tainties.  Successful completion of the following studies, presented
in order of priority, are regarded as prerequisites to further invest-
ments at a demonstration scale:

     1.  Study of hydraulic transport of ground refuse under unsteady
         flow and load conditions representative of sewer laterals
         in residential areas.

     2.  A field survey of refuse generation, water consumption and
         sewage generation to establish the daily correlation of
         refuse and sewage generation for individual households,
         a necessary prerequisite for the development of realistic
         grinder design specifications.

     3.  Evaluation of the wet systems concept from an energy
         consumption viewpoint.

     4.  Evaluation of potentially hazardous and non-treatable
         components in ground refuse - sewage mixtures, using
         advanced instrumentation methods on simulated treat-
         ment plant streams.

     5.  Development of alternative grinding concepts, including
         a laboratory study of mixing characteristics of refuse
         slurries and domestic sewage and actual construction
         and testing of a prototype grinder.  The inability to
         develop a grinder within the cost limits established
         in the present study and the specifications developed
         under task 2 would be sufficient to abandon the concept
         as economically or technically unrealistic.

     6.  Re-evaluation of the economic and overall technical
         feasibility of the concept using the new information
         available from the above studies.

-------
                               SECTION III

                              INTRODUCTION

                       The Concept of Wet Systems
                   for Residential Refuse Collection


The wet  system for residential refuse collection of prime interest to
this research involves the use of the existing sewer system to transport
domestic refuse from the .individual home to the point of treatment and
disposal, thereby eliminating conventional house-to-house collection by
trucks.  This idea depends on three technical prerequisites:  the fea-
sibility of reducing refuse in the home to a size suitable for hydraulic
transport, the feasibility of transport of ground refuse in an existing
sewer system without deleterious effects, and the feasibility of joint
treatment of refuse and sewage at an existing or modified treatment plant,
Implementation of the idea requires also economic feasibility, which de-
mands that a rigorous cost-benefit analysis shows the proposed system to
be economically favorable over other existing and proposed systems.  Each
of these prerequisites will be examined at some length in the sections
that follow.

From the viewpoint of solid waste management, the major incentive for
analyzing such a concept is the high cost of collection.  The increasing
costs are due primarily to an increase in per capita waste generation
and an increase in labor wage rates.  As noted by Clark,  the latter has
been far in excess of inflationary trends as the increasing tendency for
collective bargaining has had a substantial effect on the wage rates for
individuals employed in local solid waste management functions.  Al-
though productivity of collection in the last decade has increased sub-
stantially, there are limits to further increases under the constraints
of existing solid waste management concepts; substitutions of capital
for labor can be accomplished only to the point of having a one-man crew
on a truck so large it can no longer efficiently navigate the streets
and, consequently, only basic changes to existing practice offer any hope
to reduce the cost of solid waste collection.
                            Previous Research

The concept of wet systems for residential refuse collection has been
previously studied by several university investigators and government
funded research contractors.  There is also a considerable body of re-
search on pipeline systems for refuse transport that have as their prime
focus not the collection of refuse, but the transportation of refuse from
a few centrally located collection centers to the site of disposal.  Much
of this is directly applicable to the goals of this study.

Research on Wet Systems for Residential Refuse Collection.  The first to
investigate the feasibility of using sewers for refuse transport and
joint refuse-sewage treatment was the Los Angeles County Sanitation

-------
District, which conducted experiments with ground refuse in large inter-
ceptors in July 1963,2  It was concluded that the idea was probably
feasible, but not pursued further because of anticipated problems at the
treatment plant, the most important of which being a projected 25-fold
increase in sludge volume.

The next to study the idea \vas the ASCE Combined Sewer Separation Project.
In a report published in February of 1968, the idea of including ground
refuse in pressurized sewers was examined, but no experiments were con-
ducted. ^  Independently in 1968, Nebiker and Meier^ analyzed the idea
from a social and environmental viewpoint, and they concluded on the
basis of some preliminary cost/benefit estimates that the idea should be
further pursued.

In 1970, Etzel  suggested the use of hydropulpers, widely used by the
pulp and paper industry, as a means of refuse size reduction prior to
discharge into the sewers.  Etzel also suggested some radical changes at
the sewage treatment plant, including the use of advanced waste treatment
processes such as reverse osmosis to treat the high solids loadings that
would result.

The first detailed research study was conducted by Foster-Miller under
contract to EPA.^  The objectives of this study were to evaluate the
feasibility of hydraulic transport of ground refuse in existing sewer
lines by experimentation in a section of straight sewer pipe, and to
evaluate the feasibility of treating a sewage-refuse mixture at a conven-
tional sewage treatment plant by laboratory analyses of sample refuse-
sewage slurries.  However, because the hydraulic experiments used water
rather than sewage, and because limited treatment experiments were con-
ducted, technical feasibility was not unequivocably demonstrated.   In a
second phase, Foster-Miller developed a preliminary design for a household
refuse grinder.^  Experiments during this second phase showed that refuse
was amenable to size reduction in small, low-powered slurry grinders and
that the few components that were not easily ground constituted a rela-
tively small weight and volume fraction of the total domestic refuse.
                    g
McKinney and Mahloch  examined the costs of the concept in late 1971 and
concluded that at the present time, conventional collection was more
economical.  They stressed, however, that increased labor costs for ref-
use collection, improved wastewater treatment methods, and the successful
development of grinders*  would very likely make grinding and discharge
to the sewers the preferred method.

In 1971, the Brookhaven National Laboratory   planned a three-year,
$800,000 study for the Atomic Energy Commission to demonstrate the fea-
sibility of the concept, followed by a $2.2 million study to investigate
resource recovery at the treatment plant.  The Brookhaven concept en-
visaged the use of neighborhood pulpers and transportation in sewers to
a treatment plant which would recover paper fiber, glass, and metals and
use pyrolytic treatment for the remaining constituents.  The goal was an
effluent of sufficient quality for aquifer-recharge.  This planned pro-
ject, however, was never funded.12

-------
A second contract was let by EPA to Foster-Miller to study the feasibility
of hydraulic transport of refuse in sewer appurtenances and other poten-
tial problem areas in sewer systems (siphons, manholes, T-junctions, bends,
etc.).13  Again, water rather than sewage was used as the transport medium.
However, the results of this research did indicate some potential problems,
and will be discussed further in later sections of this report.

Finally, in late 1972, Courtney and Sexton compared the sewer transport
concept against pneumatic and pressurized alternatives.  They concluded
that sewer transport appeared particularly suitable for existing urban
areas.1^

Other  Pertinent Research.  The area of research that is most directly re-
lated  to the wet system for refuse collection is that of pipeline trans-
port of refuse.  However, the emphasis has previously been on hydraulic
transport from centrally located collection points to the ultimate dis-
posal  site, as an alternative to truck or rail haul, rather than elimina-
ting house-to-house collection.

Of all the alternative pipeline systems considered to date, slurry
transport has received most attention.  This calls for refuse to be ground
and mixed with sewage or water and conveyed under pressure as a slurry of
between 6 and 12 percent solids content.  However, several other pipeline
concepts have been considered, including hydraulic transport of briquettes
and slugs.

The slurry pipeline idea has been pursued by Zandi and co-workers at the
University of Pennsylvania for many years, and extended to include a
vacuum system to bring refuse from the individual home to the neighborhood
stations where the refuse is ground and injected into a slurry pipeline
network.16  This combination is therefore a direct competitor to the pro-
posed system, in that it completely replaces conventional collection.
Pipeline transportation of shredded solid waste was also investigated by
researchers at the Western Company, who proposed the use of friction-
reducing polymer additions to increase pipeline capacity.17

In the area of grinding, we must turn to recent developments in the pulp
and paper industry, for the latest generation of hydropulpers appears to
be eminently suited to the size reduction of paper-rich solid waste.
The experience of the paper industry is also very important to an assess-
ment of treatability.   For many years it has worked in treating cellulose
and fiber-rich wastes by technologies similar to sewage treatment, and
with special emphasis on recovery. °  In particular, screening is a
technology well developed in the paper industry that is suitable as a
modification to existing treatment plants for solids removal in prelimi-
nary treatment.

The feasibility of treating refuse slurries by technology adapted from
the paper industry is currently under study at a demonstration project
at Franklin, Ohio.19  Designed by the Black Clawson Company for 150 tons
of municipal refuse per day, the objective of the treatment plant is to

-------
recover metals, glass and paper.  The first step in the treatment process
is to pulp the incoming refuse in a hydropulper, using recycled process
water and secondary effluents.  Although a sewage-refuse slurry would re-
quire some substantial modifications in the treatment process configura-
tion, the proposed system of wet collection of residential refuse appears
to be entirely compatible with these latest efforts toward resource re-
covery from municipal solid wastes.
                       Detailed Technical Reports

The details of the technical, environmental and economic studies conducted
under this contract have been published in four separate technical "reports
which are available from the National Technical Information Service,
Springfield, Virginia  22151.  The first report undertook a preliminary
assessment of the economic and technical feasibility of the concept with
a view to establishing the order of magnitude of costs involved and iden-
tifying the major areas of technical uncertainty,20  A second report ana-
lyzed the economics of processing and disposal of refuse sludges, pub-
lished separately in view of its possible interest to an audience not
necessarily also interested in the wet systems concept.2^  The third re-
port describes a detailed evaluation of the wet systems concept in the
Springfield, Massachusetts Case study area to develop specific conclusions
and cost estimates for the conditions encountered in an area typical of
the Nation's urbanized regions.22  And finally the fourth report documents
the computer program developed for the assessment of treatment costs.2^
Again the motivation for a separate publication is the possible interest
of the subject matter to an audience not necessarily interested in wet
systems for residential refuse collection.
                           The Case Study Area

The choice of the Springfield Standard Metropolitan Statistical Area
(SMSA) in Massachusetts as the case study area, and the city of Spring-
field as its major focus, is not dictated solely by its proximity to the
contractor's home office and its representative size.  Located in the
picturesque Connecticut River Valley in western New England, considerable
environmental awareness exists among its inhabitants, evidenced for exam-
ple by its relatively advanced refuse recycling efforts.  The SMSA has
recently completed a planning program for both solid waste disposal and
wastewater treatment and its wastewater treatment plant is in the process
of being upgraded from primary to secondary treatment.  Thus the case
study addresses an area that is in a very active stage of environmental
improvement, and thus displays an interest in the evaluation of innovative
concepts.  Indeed, the willingness of the city officials to assist the
contractor in his investigations, especially the superintendent of the
city's Department of Public Works and his deputies in charge of the ref-
use and sewer divisions, has been of valuable assistance.  Moreover,
Springfield exhibits many of the problems that must be addressed in a
realistic evaluation of the concept, including the presence of a signifi-
cant proportion of combined sewers and significant seasonal variations.

-------
In retrospect, perhaps the major apparent disadvantage of the choice of
Springfield as a case study area is the fact that the wet system imple-
mentation has had to be compared to an existing collection system of
high cost-efficiency.  But we have thereby avoided the pitfall of over-
estimating the benefits of a wet system implementation by comparison to
an existing system that could easily be improved by other means.   We
recognize, of course, that there are undoubtedly areas of the country
where many of the problems of a wet system encountered in Springfield do
not exist or are of a much lesser magnitude.   On the other hand,  the
most important consideration has been to ensure that all  the important
questions have been addressed, so that the research priorities that  this
study seeks to define have in fact been based on an evaluation of the
the widest possible range of conditions.

-------
                               SECTION  IV

                                GRINDING*

                               Introduction
Although the original concept called  for  a refuse grinder  in every resi-
dence, the existence of substantial economies of scale made a  considera-
tion of alternative grinding concepts potentially attractive.  Grinders
installed on a one-per-structure and  one-per-block or neighborhood were
examined, but rejected; the former because of serious questions as to
ownership and maintenance responsibilities, the latter because of the
very high cost of the vacuum collection system required to convey the
refuse from the  individual home to the central grinding station.  Indeed,
very high residential densities would be  required to justify the block
or neighborhood  approach, densities found only in the older, central
parts of urban areas where the practicability of installing vacuum sys-
tems in old housing was open to serious question (apart from also being
those areas showing high incidence of combined sewers).

The investigations therefore returned to  the one-per-residence approach,
for which Foster-Miller had prepared  a preliminary conceptual  design.24
An artist's impression of this design is  shown on Figure 1.  However,
several questions arose as to the actual  feasibility of the approach
suggested by Foster-Miller, which centered on re-use of all domestic
wastewater except that from water closets.  This wastewater re-use would
be attained by a plumbing separation  feeding all non-fecal wastewater
into a storage tank, the installation concept being shown  on Figure 2.

First, because of power limitations,  the  grinding of ferrous metals at
this scale was considered unrealistic.  However, the exclusion of steel
and heavy gauge  aluminum was not judged a serious restriction, because
it would be necessary anyway because  of difficulties in hydraulic trans-
port in-gravity  sewer systems.  Because the design results in  nongrindable
materials staying within the grinder, and cleaned during the freshwater
flushing cycle following a grinding,  the  metal nongrindables would be
free of putrescible matter, and of relatively small volume; laboratory
tests indicated  that even tin cans would  be considerably compressed.
Storage until infrequent conventional collection would thus be feasible.
Finally, the nongrindable metals would be clean  (paper labels  and paint
would be removed during grinding, for example), and thus represent a
much better raw  material for metal recovery than raw refuse from which
nonmetals must first be separated.
       For more  detailed discussions,  see  Meier,  Kuhner and Martel,
(Ref.  20), Chapter II  and Meier,  Kuhner and Bolton (Ref.  22),  Chapter II.

-------
 -Water Gauge
    -Locking Handle
       -Lifting Handle
Water Nozzle
                                                   40"
                                                                   Lid
                                                                      Shredder Ring
                                                                  Breaker Bar
                                                                                                                    Motor
Figure 1:   Preliminary Design  for a  Household Grinder (Source:   Fisk  $ Guzdar,  op. cit., p.* 83).

-------
To
Sewer



&.
r

IXI

X
/— •
V


               ** * * *^
               t'
Fecal Waste  from
Upper Floors
                        Fecal Waste from
                        First Floor
                       Clean-Out
                          Connecting all
                          Non Fecal - Containing
                          Wastes  from All     I
                          Floors               1
Vent
                            Grinder
                                                 Overflow
                                                 to
                                                 Drain
                / /               Pump

                  Anti-Siphon
                  " Valve
                                                                   Tank

                                //A -
                                           >
                                                                         > >   -
    Figure 2:  Basement Installation of a. Household Grinder (Source:  Fisk § Guzdar, op. cit., p.  32).

-------
 Second,  the sufficiency of the quantity of reuseable wastewater necessary
 for grinding at a solids concentration at which no deleterious effects
 would occur during subsequent sewer transport (See below,  Section VI),
 was found to be open to question.   The subject is a complex one that did
 not lend itself to definitive resolution because of lack of data.  How-
 ever, the need to use some amount  of fresh water, for either grinding or
 flushing (at the end of the grinding cycle)  or both,  was clear.

 The most serious deficiency related to cost.   Apart from the cost of the
 grinding unit itself, the cost of  replumbing an existing residence to
 provide  the grinding unit with its source of reused wastewater is pro-
 hibitive, as shown on Table 1.  These costs  were estimated by a plumbing
 contractor in the Springfield Case Study Area as bid price estimates,
 and are  generally comparable,  although about $200 higher,  to the Foster-
 Miller estimates conducted two years previously.   An alternative con-
 cept was also examined, whereby the grinding would occur at high con-
 sistency (say 5-10 percent refuse  by weight),  pumped to  the sewer in a
 pressurized building connection, and relying on the flow already present
 in the sewer for dilution to the 0.5 to 1.0  precent concentration re-
 quired for hydraulic transport.  Grinding at high consistency would re-
 quire less water for a given weight of refuse.   However,  the cost of the
 pressurized building connection, even though lower than  under the waste-
 water reuse concept,  was still found to be prohibitive.  Table 2 shows
 an analysis of the cost of these two alternatives,  which,  at over $100 per
 household per year, completely outweigh all  other costs  of a wet system
 imp1ementat ion.

 For the  purposes of further investigation, it  was therefore concluded that
 some grinding concept would need to be developed that required neither a
 plumbing separation nor a pressurized building connection.   But this still
 left open the question of the  cost of the grinding unit  itself, and the
 cost at  which such an appliance  might be considered a reasonable propo-
 sition.
              The Refuse Grinder as a Domestic Appliance

The  first approach  in analysing a refuse grinder in terms  of  a domestic
appliance is to study cost trends of similar household appliances;  sim-
ilar to a refuse grinder in either use, such as the garbage grinder;
in consumer appeal, such as the refuse compactor; or in similarity  to
electromechanical sophistication and probable industrial production tech-
nique such as the automatic dishwasher.  Unlike the washing machine, all
three appliances still have connotations of convenience rather than ne-
cessity, and, also unlike washing machines, are many years from market
saturation.

An analysis of the trends in price of these appliances over the last
twenty years shows dramatic decreases; not only decreases  in  terms  of
inflation adjusted dollars but even in terms of actual selling prices.
                                   12

-------
                Table 1:   Cost Estimates  for  Installations of Refuse Grinding System



                     Construction Type                                     Estimated Total Cost


New Construction 2 Story Slab                                               $ 500.00 - 550.00
New Construction 2 Story Basement                                             S50.00 - 600.00
New Construction 1 Story Slab                                                 400.00 - 450.00
New Construction 1 Story Basement                 '                            450.00 - 500.00
Old Construction 2 Story Slab                                                1300.00 -1400.00
  Included in above:
                    Excavating concrete floor and patching                    150.00
                    Excavating pit and concrete fill                          300.00
                    Cut and patch and paint walls                             175.00
                    Remove and relay carpets                                   75.00

Old Construction 2 Story Basement                                             925.00 -1025.00
  Included in above:
                    Excavating concrete floor and patching                     75.00
                    Cut and patch and paint walls                             .175.00

Old Construction 1 Story Slab                                                1100.00 -1200.00
  Included in above:
                    Excavating concrete floor and patching                    150.00
                    Excavating pit and concrete fill                          300.00
                    Cut, patch and paint walls                                100.00
                    Remove and relay carpets                                   75.00

Old Construction 1 Story Basement                                             725.00 - 825.00
  Included in above:
                    Excavating concrete floor and patch                        75.00
                    Cut, patch and paint walls                              $ 100.00

-------
Table 2:  Cost Comparisons Between the Foster-Miller
             and Meier-Nebiker Concepts
Foster-Miller Meier-Nebiker
(Grinding at 1% Solids (Grinding at 7% Solids
with Wastewater Reuse) Discharge via Pres-
surized Building Con-

Assumed
Item Life Cost
Grinder 15 $180
Installation 15 75
Distr. and Sales
Costs 15 191
Storage Tank 30 116
Installation 30 75
Distr. and Sales
Costs 30 143
Building Con-
nection 30 0

Plumbing
Modification 30 520-1,050
Total Initial Costs 1,300-1,830
Water
Power
Repairs
Total Annual Cost
nection)
Annual Cost
at i = 6% Cost
$18.50 $200
7.70 75
19.65 19.1
8.40 0
v 5.40 0

10.40 0
0 450-500

37.75-
76.20 0
108-146 916-966
1.50
2.50
16.00
128-166

Annual Cost
at i = 6%
$20.60
7.70
19.65
0
0

0
32.60-
36.30
0
80-84
3.00
2.50
16.00
101-105
                           14

-------
For example, in 1952, the automatic dishwasher rated best by Consumer
Reports cost $354; in 1969, the cost had decreased to $280, a decrease
of $74.Z5  If one adjusts for inflation, say by use of the consumer price
index, the decrease in real cost was $210.  Similar decreases were estab-
lished for the other appliances mentioned above.

The conclusion from such a comparison is that the initial selling price
of a newly introduced appliance will fall (if not in current dollars,
certainly in inflation adjusted dollars), and that whatever the initial
cost that might be hypothesized, this would fall as manufacturing refine-
ments, competition and economies of scale becpme operative with increas-
ing production levels.

The Appliance Industry Viewpoint.  In view of the aforementioned simi-
larities between the refuse grinder, the trash compactor, and other simi-
lar appliances, a discussion of the appliance industry viewpoint on
compactor sales is clearly pertinent to an assessment of the prospects
of a refuse grinder.  The appliance industry sees trash compactors as
an item having strong growth potential in the upcoming years.  The trash
compactor is the first new major domestic appliance in years, having
been introduced only in 1970, and is therefore felt to have "novelty
value."  In reply to critics of the trash compactors' comparatively low
sales volume, it is often pointed out that it took dishwashers many years,
literally decades, to become the strong selling item they presently are.

The existing price level of around $200 is felt to be a strong selling
point.  This is considered low, and the prospects for even lower prices,
as previously documented, are good.  It is also noted that despite hin-
dered consumer appeal of dishwashers at the time, some 15 years ago, that
their price was in excess of $300, sales did ultimately accelerate as
price decreases became possible.  Indeed, one can identify a psychologi-
cal barrier at a retail sales price of around $300, and any major appli-
ance selling for less is considered likely to gain broad consumer appeal.

In view of the presently identifiable market for trash compactors, namely
residents of relatively high-income, suburban single-family home areas,
speciality advertising that has maximum impact on sales is regarded as
very effective.  Retailers are pressing the manufacturers for additional
advertising in this direction.  And yet another perceived advantage is
the feeling that because ecology and the environment will become of in-
creasing concern to the householder, advertising can make the compactor
appear to have a strong.positive affect on the environment - thus ap-
pealing to the "environmentally conscious" consumer.

Implications for the Refuse Grinder.  Whatever the shortcomings of the
industry viewpoint toward trash compactors, and especially with regard to
its exaggerated environmental benefits, it must nevertheless be concluded
that some degree of, consumer acceptance already exists, and that this is
likely to grow.
                                    15

-------
Unquestionably,  a refuse grinder would offer much greater  convenience
benefits  than the compactor.  Since refuse would be ground daily,  there
would not be the odor problem that characterises compactors.  The  spe-
cial bags that are required for the compactors would no  longer need  to
be purchased, nor would the compacted refuse need to be  carried  to the
point of  pickup, a task, incidentally, that because of the high  density
of the  compacted product, would certainly be beyond the  many persons un-
able to carry heavy  loads.

There would of course be some relative disadvantages over  the compactor,
including that of higher noise levels and the need to periodically remove
nongrindables.   Nevertheless, it does not appear unreasonable to suggest
that consumer acceptance of an appliance offering unquestionably greater
convenience benefits would be almost assured at an initial  price of  some-
what less than $300, falling gradually as sales increase to perhaps  a
level of  around  $200.

One aspect that  must be stressed, however, is the need for  operational
reliability if consumer acceptance is to be gained.  As  an  appliance that
might require housecalls by either plumbers or electricians in the event
of malfunction,  the potentially high cost of repairs might  prove to  be  a
substantial deterrent, even to the relatively affluent suburban  resident.
                              Conclusions

 If the  similarities of the refuse grinder to the garbage grinder, auto-
 matic dishwasher, and trash compactor are indeed valid, several conclu-
 sions can be drawn from the analysis of recent cost trends.

     1.  A refuse grinder selling at less than $300.00 would gain
         consumer acceptance.

     2.  The cost of the grinder should decrease in the first few
         years of production because of design improvements, econo-
         mies of scale and competition.

     3.  The real cost (inflation adjusted) of the refuse grinder
         should at worst remain stable, but probably decrease with
         time.

The conclusions with respect to the feasibility of the wastewater reuse
concept appear quite unfavorable.  The cost of replumbing existing
housing units is clearly prohibitive, and it is even questionable wheth-
er the additional plumbing costs in new housing would be acceptable.
Although such costs might well be included in the total purchase price,
and hence buried in the mortgage payments, the high interest costs there-
by incurred have considerable bearing on the real social cost of the wet
systems concept.
                                   16

-------
The presently ascertainable  costs  of grinding  have  some  very interesting
consequences for judging the most  likely initial  wet  systems implementa-
tion.  Because of their high cost,  grinders  would inevitably find their
first purchasers in the relatively affluent  suburbs for  installation  in
single-family homes of low residential  density.   Moreover,  such  areas are
very often characterised by  very high conventional  collection costs be-
cause the low costs of an efficient municipal  house-to-house service  are
not available at such low densities.  Many suburban residents spend $1
to $2 a week for a private refuse  pick-up, and consequently there may be
a real economic benefit in addition to  the convenience benefits.   This
motivation exists far less in  the  central urban areas where municipal
collection, as for example in  the  Springfield  Case  Study Area, may cost
as little as $12.00 per household  per year.     It could  of  course be
argued that scale economies  of grinding could  be  exploited  in central
urban areas by virtue of the -possibility of  a  one-per-structure  or a
one-per-neighborhood installation  of grinders,  as noted  previously.
However, again by reference  to the case study  area, almost  70 percent of
the housing units in Springfield are single-family  or duplex residences,
for which only one-per-living  unit installation is  feasible;  and in those
areas where the housing density is sufficient  for,  say,  neighborhood
grinders, the presence of combined sewers makes the concept inoperable.

All the preceding assumes that a grinding concept can be developed that
does not require replumbing  or a. pressurized building connection, yet it
is clear that the details of such  a concept  have  yet  to  be  researched.
In the absence of such concepts, we may conclude  that installation of
existing grinding concepts in  single-family  or duplex housing is economi-
cally infeasible.  The necessity for further research in this area is
obvious, and is again addressed in the  concluding section.

One alternative that might lessen  the impact on the home owner is that  of
public or quasi-public rather  than private ownership. In a manner simi-
lar to the  leasing of hot water heaters from electric utilities  (common
in New England), so might refuse grinders be leased from a  municipal
refuse utility.  This utility  would also be  responsible  for maintenance,
and could readily arrange for  emergency conventional  collection  during
grinder breakdowns.   Indeed, in areas of multi-family housing where
ownership and maintenance responsibilities are defined only with diffi-
culty, such an arrangement appears mandatory.   However,  as  so eloquently
argued by Foster-Miller  in their preliminary grinder  design report, it
is unlikely that anything less than private  ownership would engender
sufficient respect for the appliance, which  in contrast  to  hotwater
heaters, will require continuous care and attention if frequent  break-
downs are to be avoided.  But  again, even definition  of  the utility con-
cept depends first on the development of a grinding concept of reasonable
total cost.
                                    17

-------
                               SECTION V

                COLLECTION AND DISPOSAL OF NONGRINDABLES*

                              Introduction
Although the intent of the wet systems for residential refuse  collection
is to  eliminate conventional collection methods, it is readily apparent
that some residual conventional collection will still be necessary,  al-
beit of a drastically reduced scope.  Even if grinders were to be  capable
of handling metals, large bulky items would still need removal.  Moreover,
many other small constituents in residential refuse are not amenable to
grinding in the small grinders of single homes, especially resiliant plas-
tics,  rubber,  leather and some synthetic textiles.

But the conventional collection and subsequent disposal of only the  non-
grindable constituents will bring some major changes to existing solid
waste  management practice.  Volume will be markedly reduced, and density
markedly increased as the light, bulky constituents (paper, food pack-
aging,  garbage) are easily ground.  The frequency of collection may  be
very low, but  acceptable, because all putrescible materials will be
ground.  The increased density and weight of standard containers would
require trucks to be fitted with hydraulic lifts for the much  heavier
full containers, and may demand that all containers be picked  up from a
rear-house rather than curbside location.  Calculations developed  else-
where  suggest  that the density of nongrindables may be up to 50 percent
higher than the density of uncompacted raw refuse.26  Changes  in truck
design to allow for the more difficult compaction of the collected mate-
rial and the increased axle load would also be necessary; indeed,  the
compactor trucks widely used today may be quite inappropriate  under  the
proposed system.  A major impact of a wet systems implementation would
of course be a dramatic decrease of the collection labor force.  The
socioeconomic  consequences of such a decrease are considered in detail
in Section IX, below; but were found to be much less serious than  might
at first appear, especially with respect to layoffs.

Quantity of Nongrindables in the Case Study Area:

The 1972 city-wide average domestic refuse generation rate for Spring-
field  was established as 2.56 lb/cap/day,27 broken down as follows:

     Bulky refuse, collected by appointment         0.07 Ib/cap/day
     Separately collected garbage                   0.11
     Garbage now ground to sewer                    0.11
      See Meier, Kuhner and Bolton, Ref.  22) ,  Chapter III for details
relative to the case study area, and Meier, Kuhner and Martel, (Ref.20)
Chapter VI for a general discussion of the relationships involved.
                                    18

-------
     Separately collected newspaper
     All other components (regular collection)

                                                    2.56 Ib/cap/day

However, these averages were shown to hide considerable quantity varia-
tion, both by season and by route, and also considerable variation in
composition.  But no data other than an occasional sampling or our
brief survey were available.  Moreover, the figures also include an un-
determined amount of commercial refuse from a downtown collection on
Friday evenings.  Exact calculations to derive the quantity and components
of refuse ground, and not ground, were therefore very difficult because
of lack of appropriate data.  The calculations developed to measure the
impact on wastewater treatment facilities were therefore based on the
assumption that of the 2.56 Ib/cap/day, 10 percent or 0.256 Ib/cap/day
was nongrindable (based on doubling the amount of bulky refuse, as noted
below, plus continued separate collection of newspaper, plus about 4
percent of nonbulky heavy metals, hence:  0.07 + 0.07 + 0.01 + 0.106 =
0.256).  To the extent that glass, ceramics and light metals would not
be ground, the weight percentage of nongrindables would be substantially
higher.

It should perhaps also be noted that the 2.56 Ib/cap/day residential
refuse generation fate for Springfield is consistent with a national
average of 2.25 Ib/cap/day, as established on the basis of the latest
data from the EPA Data Aquisition and Analysis System for Solid Waste
Collection.28

Seasonal Peaks.  Substantial seasonal peaks, and especially the fall peak
of garden wastes in areas such as New England as typified by the case
study area, will have an important impact on the final design of a wet
systems implementation.  If one assumes that the peaks are composed of
largely grindable material, as would be the case for the fall peak, then
the treatment plant would need to be able to handle load fluctuation^
much higher than otherwise necessary.  Moreover, substantial quantities
of fresh water would be necessary to augment the normally available
grinding water supply, which in turn might impose significant demands on
the water system.  On the other hand, if the peaks are composed of largely
nongrindable materials, as would be the case in spring, at which time
relatively bulky attic wastes appear to be the major component, a regular
schedule of nongrindable collection  (based on a more or less constant
generation rate of nongrindable residuals) would be substantially dis-
rupted by strong peaks.  Indeed, aperiodicity in nongrindable collection
might follow unless additional storage space is available.  Perhaps com-
pactors might continue to play an important role here, as the operational
complexities of aperiodic collection schedules are formidable.

Bulky Wastes.  The problem of bulky wastes was also found to be much
greater than originally anticipated, because one must expect that a
significant fraction of the bulky wastes now collected by the regular
collection could not be ground by virtue of its bulk  (rather than by vir-
tue of its composition).  The majority of the bulky waste appears to be
                                    19

-------
of such a size that it can be included in a normal container, or is picked
up by the collection crews as a matter of habit (if not as a matter of
regulation); and thus the existing bulky item collection is reserved pri-
marily only for the exceptionally heavy or bulky material.  But small
furniture, large plastic toys, small appliances, and the like, would not
be readily grindable, and thus the demand for bulky item collection would
undoubtably increase under a wet systems implementation.  At the very
minimum, a doubling of the bulky item collection might be expected.

Unfortunately, we have only our qualitative observations taken during the
route surveys in the case study area to support this contention, and
quantitative data is needed before a more exact judgment can be made.  We
shall return to this point in our recommendations for further research.

Commercial Establishments.  With regard to the case study area, the city
of Springfield seems to feel very strongly about providing solid waste
collection service to commercial establishments, as evidenced by the
city's willingness to pickup commercial refuse on residential routes
without charge.  It seems reasonable to assume, therefore, that the city
would also be willing to let such commercial establishments install,
grinders, to the extent that the nature of their refuse would be com-
patible with sewer transport.  Because the collection of nongrindables
would be infrequent, inclusion of commercial establishments in the latter
collection would not replace the need for regular weekly collection of
commercial refuse were grinding not allowed.

If commercial establishments were allowed to use refuse grinders, rigorous
enforcement of ordinances specifying prohibited materials would be neces-
sary.  A permit granted only after analysis of a commercial establish-
ment's refuse, with the permit fee applied to the cost of sampling and
analysis of the refuse, might be one way of regulation.  Because the new
water quality legislation requires those who add industrial wastes to the
sewerage system to pay an equitable portion of treatment costs, it is
probable that commercial establishments would likewise need to pay some
form of user charge.  Grinders at most commercial establishments would
obviously depend entirely on fresh water for grinding water, and hence
metering of the freshwater inflow to the grinder could be used for an
equitable user charge since the quantity of water used will be propor-
tional to the weight of refuse ground and introduced to the sewer.

Emergency Collection.  An important consideration in assessing the accept-
ability of the wet systems concept is that of its flexibility to accomo-
date failures.  The capital intensity of the wet system would appear at
first glance to compare poorly with the existing system whose labor-inten-
sity and the possibility of incentive and overtime schemes makes it
relatively easy to handle unforeseen refuse peaks or equipment failures;
witness for example, the existing arrangements in Springfield where crews
who have finished early are assigned to other routes in cases of breakdown
or exceptional peaks.
                                    20

-------
There are two sources of potential  failure  of the  wet  system.   The first
is that of grinder breakdown,  and clearly an emergency collection, avail-
able on demand, would need to  be available  for such  situations-.   However,
some form of incentive would be necessary to expedite  repair of the
grinding equipment, and a nominal pick-up fee imposed  after one or two
weeks of emergency collection  might be  appropriate.

The second, and potentially more serious  problem,  would follow  a sewer
blockage.  If a sewer blockage did  occur, some mechanism would  be neces-
sary to inform upstream residents to  cease  use of  their grinders until
repairs were complete, as the  continued load of refuse solids to a situa-
tion of reduced flow velocities would clearly aggravate the problem.

These problems, coupled with the possibly aperiodic  collection  of non-
grindables and the by appointment bulky item collection, would necessitate
both an intensive public information  campaign as well  as a  unified opera-
tions center for refuse collection  and  sewer maintenance.

Frequency and Cost of Collection.   Using  relationships for  the  frequency
and cost of collection of nongrindables as  developed in the Phase I re-
port, 29 it was estimated that  between three and four regular collections
would be needed in the Springfield  case study area per year (in  addition
to increased frequency of the  by appointment bulky waste collection), and
that a typical annual per household collection cost  would fall  from $17.50
to $2.50.30  Because of the present lack  of detailed data,  a more precise
definition of the collection frequency  is not yet  possible.

Because of the drastically reduced  collection frequency for nongrindables,
and especially in view of the  possible  aperiodicity  resulting from sea-
sonal peaks, it may be necessary to send  some form of  media reminder to
householders a few days in advance  of the nongrindables collection.
Clearly, it would be unreasonable for a householder  to remember  the days
of collection if spaced at such long  intervals.  Such  notice might  entail
considerable administrative effort, and would probably presuppose a cen-
tralized and computerized system run  in conjunction  with the operational
control of the collection vehicles.
                                 Disposal

 Disposal methods would also  be  affected significantly.  Because most of
 the nongrindable materials are  incombustible,  incineration of nongrind-
 ables  is not  a  feasible proposition.   On  the other hand, recovery of
 metals from the ungrindables will  doubtless be aided  in the  absence of all
 light, putrescible matter.   But landfill  disposal is  the most likely op-
 tion,  until such time  as  resource  recovery and recycling of  metals becomes
 more widespread.  Unfortunately, the  economies of scale in conventional
 disposal methods will  have a negative impact on the disposal of nongrind-
 ables  under a wet systems implementation;  a 90 percent reduction in volume
 will bring less than a 90 percent  cost reduction, and the per ton cost of
 disposal will increase.
                                    21

-------
It has been argued that the cost of existing solid waste collection could
oe substantially reduced if the volume of domestic refuse were reduced by
tne^elimination of relatively bulky glass containers resulting from a ban
on  no-deposit no-return" bottles.   But this would also benefit the wet
system, as ground glass is one of the major refuse categories that can be
expected to cause problems in sewer transport (see below,  Section V).
                                   22

-------
                               SECTION VI

                          HYDRAULIC TRANSPORT*

                               Introduction
The key to the feasibility of the wet  systems  concept  is the ability of
an existing sewer system to transport  ground refuse without adverse ef-
fects.  Unfortunately, the literature  and  experience dealing specifically
with the ability of sewers to transport  solid  waste is extremely limited,
and only three studies were considered of  any  consequence to our analysis.
One mentioned earlier was a result  of  depositing very  coarsely ground ref-
use in large sanitary sewers performed in  1963 by the  County Sanitation
Districts of Los Angeles County.2   The other two were  studies conducted
by Foster-Miller Associates for  the Environmental Protection Agency.31'32

A detailed review of this literature shows that two conditions must be
present for hydraulic transport  in  an  existing system  to be judged feas-
ible;33 a scour velocity of about 2.5  ft/sec is necessary to prevent accu-
mulation in the sewer of the heavier components such as glass and metals,
and the concentration of refuse  must be  below  the so-called "limiting con-
centration" for the sewage-ground refuse mixture to have hydraulic proper-
ties similar to sewage alone.  This limiting concentration is also a func-
tion of flow velocity, increasing with increasing velocity and depth of
flow, and lies between about 0.5 and 1.5 percent refuse solids by weight.

Although the Foster-Miller experiments covered steady-state conditions in
both straight sewer pipes and in a  variety of  appurtenances (including
manholes and siphons), their conclusions of feasibility must be put in the
light of actual rather than designed conditions in a sewer system.  Some
preliminary analysis showed that although  both conditions are generally
met in interceptors, there might be substantial difficulty in meeting them
in lateral street sewers.  Consequently, the focus of  the case study in-
vestigation with respect to hydraulic  transport was to establish the mag-
nitude of maximum velocities in  typical  street sewers, and to establish
the relationship between the quantity  of freshwater needed for grinding
and the limiting refuse concentration  in the sewer.  Obviously, by adding
sufficient freshwater to the grinding, the limiting refuse concentration
can always be met; the problem,  however, is that such  freshwater additions
may be undesirable both from the viewpoint of  increasing total freshwater
consumption as from the viewpoint of increasing flows  at the treatment plant,


                      Scour Velocity Investigations

The laboratory studies conducted by Foster-Miller Associates have deter-
mined that for the particle size distribution  emanating from their pre-
liminary grinder design, a minimum  velocity of 2.5 ft/sec (0.76 m/sec) is
needed to maintain self-cleansing conditions in sewers.  Since steel cans
     *For more detailed discussions,  see Meier,  Kuhner  and Martel,
 (Ref. 20), Chapter  III and  Meier,  Kuhner and  Bolton,  (Ref. 22),  Chapter  IV.
                                    23

-------
have already been eliminated due to excessive power requirements for
grinding, high grinding costs and problems in appurtenances, especially
siphons, this velocity is required to transport the glass, ceramic, and
aluminum components only.  No determination was made as to how often or
how long this scouring velocity should be present.  However, in order to
avoid excessive accumulation of debris, it would not be unreasonable to ^
require daily scouring of settled materials.   Otherwise, deposited materi-
als may combine with grease or other cohesive materials to form a cake re-
sulting in a further loss in hydraulic properties.

Assuming unobstructed flow conditions, there are two occasions where
sewers are incapable of providing the required 2.5 ft/sec (0.76 m/sec)
scour velocity.  The first occasion is where sewers are designed according
to the minimum requirements of most health departments which specify a
"minimum full-conduit velocity of 2 fps."34  Since maximum flow velocities
are achieved at full or near full flows, these sewers cannot be expected
to flow faster than 2.0 ft/sec (0.61 m/sec).   The second occasion is where
sewers are not flowing full enough to attain a 2.5 ft/sec (0.76 m/sec)
velocity.

Investigation of the Springfield sewerage system revealed that sewers
laid at minimum slope (to produce a full pipe velocity of only 2.0 ft/sec)
are not uncommon.  It also revealed, through a sewer survey, that even in
sewers where the design slope is great enough to produce velocities in
excess of 2.5 ft/sec (0.76 m/sec), the quantity or depth of flow is insuf-
ficient.  The highest maximum velocity reached during a 7 day survey of 7
lateral sewers in August 1973 was 1.62 ft/sec (0.5 m/sec), and the lowest
maximum velocity occuring in a 24 hour period was 0.74 ft/sec (0.225
m/sec).  The conclusion that glass and other high specific gravity materi-
al would accumulate in such sewers is inescapable.

One obvious alternative to resolve this problem would be the elimination
of glass, ceramic and aluminum material from the list of grindables, along
with steel cans, heavy gauge plastic and heavy gauge aluminum.  However,
this would increase the collection cost of nongrindables and would seri-
ously reduce the convenience to the user.  On the other hand, instead of
prohibiting the grinding of glass and other high specific gravity materi-
als , it may be feasible to simply flush accumulated materials from the
affected sewer lines at regular intervals.  The frequency of flushing
would depend on the rate of accumulation and the effect of the accumulated
deposits on the hydraulic properties of the pipe.  Without actual experi-
ence or field test results, one can only assume a likely frequency.  But
based on some calculations which indicate an accumulation of about 2.2
inches of glass in a typical 8-inch lateral sewer per year, an intensive
flushing program would probably be required.  Discussions with municipal
officials with great experience in sewer maintenance indicated the impos-
sibility of accurate prediction in this area in the absence of actual
data.

Another alternative to eliminating glass as a grindable item is to rede-
sign the sewer system to achieve the minimum scour velocity of 2.5 ft/sec
                                    24

-------
(0.76 m/sec) at the  shallower depths common in sewer flows.   This would
mean steeper sloped  sewers  and a more frequent use of lift stations.   This
might be considered  feasible for newly developing and urban renewal"areas
that could readily implement new design standards.  However,  some sample
designs and cost  analyses  quickly showed that a tripling of construction
and operating cost was  involved and this alternative, too, was rejected
as a feasible proposition.


                      Grinding Water Investigations

The sewer survey  showed that street sewers rarely, if ever,  flow greater
than half full.   This is important because the limiting refuse concentra-
tion is strongly  influenced by the flow depth.   Foster-Miller defined
limiting refuse concentration as that concentration at which  the friction
resistance is the same  as  that for water and beyond which the friction
resistance increases rapidly.35  Consequently,  if refuse mixtures above
the limiting concentration are introduced into the sewerage system, re-
duced flow characteristics  would result.  Figure 3 shows the  relationship
between limiting  refuse concentration and flow depth.   For the depths  of
flow normally experienced  in street sewers (between .1 and .3 d/D or  d =
2 inches) it appears that  the limiting concentration should be in the
vicinity of 0.5 percent, instead of 1.0 percent as originally thought
(the latter figure being used by Foster-Miller as a basis for the grinder
design).  However, in the  larger interceptors where greater depths of
flow can be expected, the  limiting refuse concentration may be as high as
1.5 percent if flowing  full or near full.

Since the ground  refuse must flow through the street sewers of residential
areas before discharging into an interceptor, the lower limiting refuse
concentration of  0.5 percent should govern the dilution water requirements
at each grinder installation.  This is unfortunate because the lower  lim-
iting refuse concentration (0.5 percent) is made to apply in  an area where
the least amount  of  dilution water is available.  In street sewers, only
residential flows and some infiltration can be expected whereas in the
larger interceptors, dilution water is available from industrial and  com-
mercial sources as well.

The relationship  between the amount of grinding water required and the
percent solids content  for various refuse generation rates is shown in
Figure 4.  This graph was  developed from the equation:

                                 _ L(l-fl-X)
                               q "  8.34X

where q is the gallons  of grinding water required per capita  per day,  L
is the pounds of  refuse per capita per day, fi is the moisture content of
the refuse  (assumed  to  be  20 percent) and X is the grinding  consistency
 (0
-------
   5-1-
   4--
in

JC.
O
3
u_
u_
O
I
a.
Id
o
   3--
   2--
   I --
                                                                                  6LASS
                                                                              TRANSITE
                                                                              (ASBESTOS CEMENT)
                                                                  6 INCH  DIAMETER PIPES
                              0.5                       1.0
                           LIMITING REFUSE CONCENTRATION,
                                                                                 1.5
                                                                percent
Figure 3:  Limiting Refuse Concentrations  for Hydraulic Transport in Sewers
           (Source:  Guzdar  §  Rhee,  op.  cit., p.  98)

-------
                                POUNDS OF REFUSE PER CAPITA PER DAY
                                    (INCLUDING MOISTURE)
                          234
                          PERCENT SOLIDS CONTENT
Figure 4:   Grinding Water Requirement

                                    27

-------
is 2.56 Ibs/cap/day  (1.14 Kg/cap/day), the amount of grinding water  re-
quired is approximately 48 gals/cap/day (182 £/cap/day).

Because of this high grinding water requirement, Foster-Miller suggested
that household nonfecal wastewater be stored and reused for grinding.  For
the purpose of this analysis, nonfecal wastewater includes discharges  from
all plumbing fixtures except the toilet.  Therefore, the total household
water consumption required to allow refuse grinding using only nonfecal
wastewater is the sum of the household grinding water requirement and  the
household fecal wastewater component.  The household grinding water  re-
quirement can be determined by multiplying the number of persons per
household times the per capita grinding water requirement determined from
Figure 4.  The household fecal wastewater component can be calculated
based on an average toilet use of 4 uses/cap/day at 5 gals/use (19 £/use).
For an average household with 3 persons per household, the household water
consumption rate needed to grind refuse is approximately 204 gals/HH/day
(3 persons per household x (48 + 20) gals/cap/day), which is substantially
above actual consumption rates.

The water consumption records were examined for homes in two neighborhoods
in Springfield, one of 36 single-family homes,  the other of 50 duplexes.
Analysis showed that the single family home area would require an 8.5
percent increase in water consumption in the winter and a 39 percent in-
crease in the fall.  The duplex area would require a 3.7 percent increase
in the winter and a 12 percent increase in the  fall.  The seasonal depen-
dance is dictated by the seasonal fluctuations  in refuse generation, as
noted in Section IV, above.   Considering that over 65 percent of the
housing units in the city of Springfield are either single-family homes
or duplexes, this would represent a large increase in total water demand.
Moreover, while increased flows could probably be accommodated in the  -
lateral sewers without difficulty (because, as  shown by our survey,  they •>
rarely flow at more than half depths),  there may be problems in intercep-
tors that are more likely to reach flows nearer their design value,  and
of course, the total inflow at the sewage treatment plant would also in-
crease.  A conclusion from this analysis is that much of the garden wastes
collected in the fall, which is largely responsible for the higher refuse
quantities in that season, cannot be ground for reasons of unacceptable
freshwater demands, and would therefore need a  special conventional coir
lection.
       Hydraulic Transport at Higher Than Critical Concentrations

The large quantity of dilution water required to maintain hydraulic flow
characteristics similar to water raises the question as to the impact of
depositing refuse in sewers at concentrations higher than the limiting
refuse concentration (0.5 percent).   If higher refuse concentrations of 1
percent and more were hydraulicly transportable, substantially less dilu-
tion water would be required.  From Figure 4 it can be seen that at 1 and
2 percent refuse concentrations, dilution water requirements can be re-
duced substantially.  This in turn -would mean that a much larger percen-
tage of homes would possess the necessary dilution water and very little
                                    28

-------
additional fresh water would be needed.   Also,  the laboratory experiments
of Foster-Miller have shown that  flows at higher refuse  concentrations  (2
percent and above) have the ability to maintain heavier  particles  in  sus-
pension.  This would eliminate the  need  to attain a specific  scour veloci-
ty to remove settled materials.   However,  these heavier  particles  would
likely settle out as the mixture  is diluted further downstream.

But, Foster-Miller also noted that  head  losses  increase  dramatically  at
refuse concentrations above the limiting refuse concentration.  Con-
sequently, mixture velocities would be less than for sewage alone, or for
refuse mixtures below the  limiting  refuse concentration.  Depending on the
condition of the sewer, this could  cause more frequent blockages.  Also,
the detention of sewer flows may  encourage septic conditions  which would
be more objectionable in residential areas.  Although sewers  can be ex-
pected to flow fuller as a result of the decrease in velocity, this is not
expected to be a problem since most street sewers are overdesigned with
respect to hydraulic capacity as  evidenced in .the sewer  survey, and rarely
flow above half-full.

Thus although there  appear to be  some benefits  to be derived  from  grinding
and transporting refuse mixtures  at higher than critical concentrations,
it is difficult to make judgments of its technical feasibility until  fur-
ther experiments have been conducted.


          Alternatives to  Hydraulic Transport in Gravity Sewers

A number of alternative modes of  hydraulic transport were examined in the
early stages of the  study, but rejected  from further consideration either
for reasons of lack  of data or for  obvious technical reasons  at the scale
of application here  under  investigation.  Briquette and  slug  flow  systems,
investigated by the  Stanford Research Institute for point-to-point pipe-
line transport of refuse^ were rejected for both technical reasons and
because of the obvious impossibility of  providing briquette and slug
forming equipment at a scale smaller than neighborhood.    Similarly,
pressure systems were rejected on the basis of  both cost (since a  new
pressure conduit would need to be built  in existing areas) and operational
difficulties.37
                                     29

-------
                               SECTION VII

                               TREATMENT*

                              Introduction


There is a critical lack of information on the physical-chemical charac-
terization and treatability of ground refuse in sewage, and therefore ac-
curate predictions of the impact of ground refuse to an existing treatment
facility, or detailed design specifications for new facilities, are not
easy.  Indeed, the only comprehensive study is that undertaken by Metcalf
£ Eddy38 as part of the initial Foster-Miller contract, although some
limited information is also available from the experiments conducted in
the early sixties in Los Angeles.2

A detailed review of this literature and other sources considered relevant
is contained in Chapter IV of the Preliminary Assessment,20 and we shall
here summarize only the most important conclusions:

     - Existing grit chambers would probably be adequate to handle
       increased load; major problem would be increased entrapment
       of organics and paper fibers.

     - Suspended solids (SS) concentrations into primary settling
       units would increase about 10-fold over existing levels.
       Existing primary settlers would probably be adequate, but need
       much strengthened skimming and sludge scraping devices.  BOD
       and SS concentrations in the effluent would be of the same
       order of magnitude as those encountered at existing treatment
       plants.

     - Screens were found by Metcalf § Eddy to be effective as a
       primary treatment device, especially the Bauer "hydrosieve."

     - The need for equalization of either load or flow is largely
       unknown, because the hour-to-hour patterns of refuse grinder
       operation are still conjectural.

     - Secondary treatment appears to be little affected by the ad-
       dition of ground refuse.  Long-term BOD exhertion studies showed
       no inhibitory effects and identical rate constants.  Phosphorous
       and nitrogen were determined to be adequate, with little change
       over the case of sewage alone.  Heavy metal concentrations were
       also largely unchanged, with no evidence of toxicity.
      See Meier, Kuhner and Bolton, (Ref.  22),  Chapter V and Meier, Kuhner
and Martel, (Ref. 20).Chapter IV for detailed discussions.  The problems
of sludge handling and disposal are discussed also in Section VIII of
this report, below.
                                  30

-------
     -  Theoretical analyses of the maximum mercury concentration that
       might be introduced by residential refuse showed the pos-
       sibility of mercury concentrations in sewage substantially
       higher than those encountered today.  Metcalf § Eddy's measure-
       ments, however, showed no increase.

     -  The impact of ground refuse addition on advanced waste treatment
       was concluded to be marginal.  Tertiary polishing devices (e.g.,
       microstrainers) were the only type of AWT that might need to be
       specified for reasons of ground refuse additions because of the
       difficulty in aerobic degradation of cellulose fibers.  The im-
       pact on phosphorus, nitrogen or dissolved organics removal pro-
       cesses was considered immeasurable in the absence of data.

Because of the difficulties encountered in making generalizations, con-
siderable effort went into an examination of treatment in the case study
area, using mathematical simulation techniques as the major tool.  The
mathematical model utilized for this purpose is described in detail in
a separate report.23  The most immediate shortcoming of using a mathe-
matical model based primarily on design relationships valid for conven-
tional sewage is that the addition of ground refuse may dramatically
change the physical-chemical relationships involved.  Whenever possible,
such changes were incorporated, but the basis for the changes were the
very limited studies discussed above and hence may well be subject to
question.  Nevertheless, in the judgment of the writers and the review
panel, the results generated by the model were consistent with the goals
of the study, and, whatever their shortcomings, certainly an improvement
over mere qualitative and subjective opinions as to the consequences of
ground refuse additions.


          Existing Treatment Facilities in the Case Study Area

At the present time, Springfield utilizes two wastewater treatment facili-
ties, each providing primary treatment for domestic sewage and most of the
industrial wastes in the city.  The smaller facility, Indian Orchard
Treatment Plant, is located in the northeast section of the city on the
Chicopee River.  At present the plant handles only domestic sewage from
the  Indian Orchard section of Springfield, and wastes from the Monsanto
industrial complex are dumped directly into the river.  The nominal capac-
ity  of the plant is estimated at 3.0 mgd  (11,355 m3/day).  The treatment
process consists of screening, grit removal, and sedimentation prior to
discharge of the unchlorinated final effluent.  Settled sludge is digested,
dried on glass-covered drying beds, and used as fill.

The  major portion of Springfield's waste  load is treated at the main
treatment plant located on the Connecticut River at Bondi Island in Agawam.
In addition to Springfield, the plant presently treats wastes from West
Springfield, Agawam, and parts of East Longmeadow.  The treatment facility
is quite old, having been completed in 1940 as a Works Progress Administra-
tion (WPA) project.  It provides units for coarse screening, grit removal,
                                    31

-------
comminution of solids, and primary settling.  Sludge handling facilities
include heated digestion tanks, elutriation tanks, a vacuum filter, and
land disposal site.  There is no chlorination before discharge.  Figure 5
provides a schematic of the present operation.

Although the design capacity of this existing plant is rated at 30 mgd,
flows in excess of this figure are regularly recorded, with a resulting
decrease in removal efficiency.  Flows beyond 45 mgd (170,325 m /day) are
automatically bypassed directly to the Connecticut River.

The primary factor limiting the treatment capacity of the existing plant
is the sludge disposal system.  The digestion process is impeded when
large volumes of sludge are fed to the digesters.  Also, one unit must be
dewatered each year for grit removal.  As a result of the digesters  in-
ability to handle the present sludge volumes, approximately half of the
settled sludge is now bypassed directly to the elutriation tanks.  There
the two sludge streams, digested and raw, are mixed prior to vacuum fil-
tration.

The computer model representation of the existing system further stressed
the inadequacy of the present removals.39  On the basis of the model, it
was estimated that only 48 percent of the BOD is removed by the existing
facility.  The theoretical addition of 2.56 pounds of refuse per capita
per day (1.15 Kg/cap/day) would obviously aggravate the present plant
limitations even further.
                                   32

-------
     BAR SCREEN
/? , GRIT «r78!™
// CHAMBER W
SCREENINGS

GRlf
BY-
PASS


PRIMARY CONNECTICUT
CLARIFIER
1
i
i
(
i
SLUDGE PUMPS
JFIRST STAGE DIGESTER

JSECdND STAGE DIGESTER
i

. ELUTRIATION

VACl
FILT

i
JUM
ER
TRUCK HAUL
. ^
RIVER
LANDFILL
f\

Figure 5:  Existing Process Configuration at the Springfield Wastewater Treatment  Plant

-------
                     Proposed Sewage Treatment Plant


In light of the inadequacy of the present plant, the city of Springfield
has contracted for development of a new treatment facility.  The plan
calls for abandonment of the Indian Orchard facility, and construction of
a new plant at the Bondi Island site.  This new plant would provide sec-
ondary treatment for all the present participant towns and, in addition,
would include the.town of Ludlow and the industrial flow from the Monsanto
Chemical Company.  The rates of flow and strengths of wastewater projected
for each contributor for the design year (1990) are listed in Table 3.
The plant as designed will consist of bar screens, grit separation units
(cyclones), primary settling tanks, aeration tanks, flocculation tanks,
final settling and chlorination facilities.  Pumping units will be pro-
vided to discharge the plant effluent into the Connecticut River during
flood seasons.  The sludge handling facilities will include gravity
thickeners for primary sludge, flotation thickeners for waste activated
sludge, holding tanks for both primary and waste activated sludge, and
wet oxidation facilities to sterilize arid reduce sludge volumes and thus
the size of the vacuum filters.  A process schematic of the proposed
64.1 mgd (264,618 m3/day) waste treatment plant is shown in Figure 6.


           Table 3:  Estimated Rates of Flow and Strengths of
                    Wastewater From All Contributors
City
Ludlow
Monsanto
Indian Orchard
East Longmeadow
Springfield
West Springfield
Agawam
Flow,
Average
2.7
8.8
• 3.5
2.3
31.7
9.0
6.1
,a
mgd
Peak
8.2
19.1
15.7
7.4
91.7
16.9
15.4
BOD .
lbs/day°
4,500
48,500
7,300
3,400
45,700
11,300
11,800
s.s. 1
Ibs/day
3,380
23,500
5,850
4,400
44,500
16,200
14,500
TOTAL                   64.1        174.1        132,500        112,330


Smgd x 3,785 = m3/day

blbs/day x .454 = Kg/day
                                    34

-------
             BAR SCREEN
en
 SKIMMINGS
_C
                                  PRIMARY
                                  CLARIFIER
                                                                  POLYMERS OR ALUM
                                     SLUDGE
                                     PUMPS
                     AERATORS
                                        CYCLONE
CLARIFIERS
                                                                     SLUDGE RETURN
                                                                  AIR
                                                                 FLOAT-
                                                                 ATION
                                           GRAVITY THICKENER
                                           a HOLDING TANKS
                                    COMMINUTER

                                  ^SLUDGE PUMPS
                 CHLORINE CONTACT CHAMBER
                                 TO
                            CONNECTICUT
                                                                               HOLDING TANKS
                                          COMMINUTER
                                     SLUDGE PUMPS

                                           BY-PASS
                                                                  RIVER
                                                                                                  TRUCK HAUL   LANDFILL
                                                                                            I
                    Figure  6:   Planned Process Configuration  at the Springfield Treatment  Plant.

-------
                     Generation of Input Parameters

 In order to estimate what effect the inclusion of ground refuse  would have
 on the proposed plant's performance, it is necessary first to model  the
 plant as it is presently envisioned*2 and then to adjust the input quality
 parameters to reflect the additional load occasioned by 1.79 pounds  of
 ground refuse per capita per day (0.81 Kg/cap/day).

 As noted previously, there is a critical lack of information on  the  char-
 acter of ground refuse in sewage.  For this reason, figures dealing  with
 the physical and chemical characteristics of ground refuse in sewage must
 necessarily be best estimates based on the limited information which does
 presently exist.  Because of the large amounts of solids and floatables
 present in sewage containing ground refuse, and the subsequent problems
 these solids cause in laboratory analyses, all past tests have been  per-
 formed on primary effluents.  In the Metcalf and Eddy study,43 ground
 refuse was added to Boston sewage in varying concentrations ranging  from
 0.25 percent to 1.0 percent based on a per capita daily wastewater genera-
 tion of 100 gpcd (378.5 Jl/cap/day).   In Springfield,  records show that
 about 140 gallons (530 litres) of sewage are generated per person per day.
 Thus, at a refuse generation rate of 1.79 Ib/cap/day (0.81 Kg/cap/day)
 the loading on the treatment plant will average only about .2 percent by
 weight.

 In order to estimate the quality parameters associated with a refuse-
 sewage stream of this strength, it is necessary to estimate the nature of
 the primary effluent and then to work back, using a solids balance to
 obtain reasonable values for the influent flow.   Following this procedure,
 estimated concentrations associated with a 1.79 Ib/cap/day (0.81 Kg/cap/
 day) loading in the Springfield sewerage system were developed.  These
 concentrations, along with those of the sewage flow,  are listed in Table
 4.  The 1.79 Ib/cap/day is based on the assumption that of the total  2.56
 Ib/cap/day generation rate, 10 percent by weight is nongrindable and  20
 percent by weight is moisture.  In the following analysis, therefore,  the
 Ib/cap/day figures used are on a dry weight basis.


          Effects of Refuse on the Proposed Bondi Island Plant

 The imposition of ground domestic refuse on the proposed Bondi Island
 Treatment Plant would affect some processes to a greater degree than
 others.   It was our intent, through the computer model, to highlight
 those plant functions which would be most severely limited.  Once these
 problem areas have been designated,  it is then possible, again with  the
 help of computer simulations, to try various alternative schemes in  order
 to obtain the best quality effluent at the least possible cost.

As might be expected, the increased concentrations of solids and BOD will
 greatly influence the sizing of many plant components, especially sludge
handling processes.   It is estimated that a loading of 1.79 pounds of
                                   36

-------
          Table 4:  Present and Projected  Waste  Concentrations
                                               mg/l,
                                       Present       With
            Parameter                   Sewage     .2% Refuse
Solid Organic Carbon
Solid Nonbiodegradable Carbon
Solid Organic Nitrogen
Solid Organic Phosphorus
Solid Fixed Matter
Solid BOD CSBOD)
Volatile Suspended Solids
Total Suspended Solids (TSS)
Gross Suspended Solids (GSS)
Normal Suspended Solids (NSS)
Heat, BTU/gallon
Grit CGRIT)
Dissolved Organic Carbon
Dissolved Nonbiodegradable Carbon
Dissolved Nitrogen
Dissolved Phosphorus
Dissolved Fixed Matter
Alkalinity, mg/5, as CaCQ^
Dissolved BOD
Dissolved Organic Matter
105.0
30.0
10.0
2.0
73.0
93.0
91.0
164.0
0.0
154.0
616.0
10.0
43.0
19.0
11.0
5.5
259.0
92.0
40.0
160.0
181.0
60.0
12.0
4.0
473.0
113.0
1,041.0
1,514.0
525.0
699.0
5,692.14
290.0
73.0
20.0
20.0
8.25
500.0
250.0
115.0
320.0
ground refuse per capita per day  (0.81 Kg/cap/day) will result in an
average total solids concentration of 1,514 mg/H entering the plant.  This
figure is roughly ten times the present loading.  A great portion of these
added solids will be in the form  of grit from ground glass (290.0 mg/£),
and gross solids from packaging and the like (525.0 mg/fc).  The quantities
of these two constituents place particular constraints on the design and
operation of the grit handling facility.  The proposed plant will feed all
the primary sludge, including the grit, to a cyclone separator.  Although
this scheme should provide adequate removal for all projected grit quanti-
ties resulting from sewage alone, the greatly increased concentrations ex-
pected as the result of ground refuse addition could cause numerous prob-
lems with clogging and inefficient grit separation.  Rather than increase
the capacity of the cyclone separators, a cheaper and more efficient solu-
tion was found to incorporate a conventional aerated grit chamber and grit
washer prior to primary settling.  Such a modification will have the ef-
fect of removing grit prior to primary sludge pumping, and should provide
a more effective operation under  the changed conditions anticipated with
ground refuse additions.
                                    37

-------
As presently envisioned, the proposed plant will utilize two separate
sludge thickeners.  Primary sludge will flow from the cyclone separator
directly to a series of gravity thickeners.  Waste activated sludge, on
the other hand, will be thickened to about 4 percent solids by use of
six air flotation units.  There are holding tanks for each stream of
thickened sludge prior to mixing and injection into the wet air oxidation
unit.  The ratio by weight of dry solids of primary to waste activated
sludge in a conventional wastewater treatment plant handling the sewage
alone is about 1.2:1, or even 0.5:1.    By comparison, the ratio which
could be expected by adding 1.79 lb/cap/day (0.81 Kg/cap/day) ground ref-
use soars to about 12:1, primary sludge to secondary.  At such high ra-
tios, the cost-effectiveness of two separate thickening processes would
appear questionable.  Because the waste activated sludge portion is so
small, it can be mixed with the primary sludge and then thickened together
with little loss of efficiency, but at great cost savings.45  Therefore,
should a wet system of refuse disposal be initiated,  it would be more
economic to thicken both sludge streams together in a gravity thickener.
This is confirmed by the cost analyses presented below.

The wet-air oxidation process operates under the principle that organic
substances may be oxidized under high pressures at elevated temperatures
with the sludge in a liquid state by feeding compressed air into the
pressure vessel.  The process was developed in Norway for pulp-mill
Wastes, but has been revised for the oxidation of raw sewage sludge,
pumped directly from the primary settling tank or thickener.   Combustion
is not complete.  For the Springfield plant, a low pressure oxidation
unit has been specified.  Performance characteristics of such a unit in-
clude a 15 percent reduction of Chemical Oxygen Demand (COD)  and a 30
percent destruction of the insoluble volatile matter.  The end product
of the process is a sterile dewaterable and oxidized  liquid mass.  As a
result of the rather low degree of oxidation and the  destruction of the
volatile solids, a good portion of the solid COD is converted to dis-
solved COD and returned with the supernatant from the vacuum filter to the
head of the plant.  For normal sewage this additional COD loading is not
significant, but when ground refuse has been added to the sludge stream,
the COD of the supernatant can be expected to be of such large magnitude
as to severely affect the aeration processes to which the vacuum filter
supernatant is returned.

This would not be as serious a problem if anaerobic digesters were used
since the digesters convert organic acids to methane  gas and carbon dir
oxide, and thus remove a great deal of the COD from the return stream.
As noted by Kos  et al.,4" there is sufficient evidence to show that
ground refuse can be satisfactorily digested with sewage sludge.  More-
over, contingent on a favorable local market, this technique can provide
a valuable commodity through methane gas production.   For these reasons,
we would recommend that should ground refuse be added to the sewage stream
at Bondi Island, anaerobic digesters would provide a better solution than
wet-air oxidation.  However, if the methane gas would yield less than
$0.80/1000 cu ft, our detailed studies on sludge handling show that the
digester should be omitted, with the thickener underflow fed directly to
holding tanks and vacuum filters. '
                                    38

-------
In addition to those processes already mentioned, many  other plant  func-
tions would need to be enlarged to handle  the  additional  load;  foremost
among these are the sludge holding tanks,  sludge pumps, thickeners  and
the aeration basins.

The computer analyses indicate that  as a result of  ground refuse additions
dissolved BOD loadings into the aerator will increase threefold and solid
BOD would increase by a factor of 1.2.  Total  solids to the aerator would
increase by a factor of about 4.5.   In order to achieve the same degree
of BOD removal as the proposed plant with  just sewage,  the aeration basin
would need to be substantially enlarged.   This seems in sharp contrast to
the primary settler, which would not need  to be enlarged.  The  reasons
for this apparent discrepancy is that using existing settling curves, the
great majority of solids added by the ground refuse would either float to
the surface or settle easily within  the alotted detention period.   This
fact is further detailed by comparing solids concentrations in  primary and
secondary sludge stream.  With ground refuse added, the total solids in
the primary sludge stream increases  elevenfold over that  expected from
just sewage.  On the other hand, the solids in the  waste  activated  sludge
stream remain virtually the same for both  cases.

This increase in primary sludge solids will affect  the  sizing of all sub-
sequent sludge handling processes.   Accordingly, there will be  increased
capacity and cost associated with the sludge pumps, sludge holding  tanks,
thickeners, digesters and vacuum filters.  Specifics regarding  these pro-
cesses, including landfill requirements, are discussed  in detail in the
separate report on sludge handling and disposal.48

Figure 7 shows the computer model representation of the recommended plant;
the notation and abbreviations are detailed in Meier and  Fisette.24  We
have indicated that a microscreen would be used just prior to chlorination.
The effect of the screen would be to compensate for seasonal variations
and to polish the effluent by removing many of the  trace  solids (cellulose)
which might remain.  The result is that the plant as a whole will remove
92 percent of the BOD from the refuse-sewage mixture and  will provide an
effluent which is at least as good as that forecast for the proposed plant
with sewage alone.

In addition to these quality parameters outlined in Table 4, there  remain
a number of variables associated with ground refuse in  sewage about which
we have little or no knowledge.  These include possible effects connected
with grease and oil, toxic materials and heavy metals.  It is important
that a complete investigation into the nature  of ground domestic refuse
in sewage be conducted in order that effects on the various treatment
processes associated with these additional parameters might be  fully ad-
dressed in the preliminary phase of  our investigations.

Seasonal Variations.  As detailed in Section III of this  report, one can
expect substantial seasonal variations in  both the  character and quanti-
ties of refuse generated.  Just as these variations affect present
                                    39

-------
   BSCRNI   GRIT
                COMM
                                        PRSET      SKIM
AEMFS        SPLIT
                                                                   21
                                                                                    MICRO   CHLOft
                                                                       15
                                                                                 27

                                                                                  34
                                 29
                        33
                                                                             MI)T 33
                                                                ABBREVIATIONS ARE
                                                                EXPLAINED IN  Mcitr a Flscttt, op. cit.
                                                     MIX      SHAUL     LANDF
Figure  7:   Computer Model Representation of the  Recommended Treatment Process Configuration
            for Ground Refuse

-------
disposal techniques, they would  also  affect treatment plant operation
should these peaks be ground  into  the sewers.

Table 5 shows the expected  seasonal variations  in influent  quality para-
meters, given that the fractions of each refuse category ground remain
constant throughout the year.50  Clearly,  the peak seasonal load would
be expected in the fall.


       Table 5r  Seasonal Variations  Due to Ground Refuse Addition
*
Parameter
SBOD, ppm
VSS, ppm
TSS, ppm
DBOD, ppm
GRIT, ppm
GSS, ppm
NSS, ppm
Summer
112.25
1002.80
1450.00
112.12
267.50
518.17
665.85
Fall
118.92
1354.00
1850.00
133.00
277.45
628.84
944.69
Winter
108.32
794.97
1213.50
100.26
260.36
464.94
491.73
Spring
112.95
1012.05
1542.30
114.58
354.76
488.65
692 .-73
Annual
Average
113
1041
1514
115
290
525
699
 *See Table 4  for  abbreviations.
 In order to  evaluate  the  effect such a seasonal  peak would have on the
 treatment plant  operation,  the computer model  was  run  first with input
 parameters representative of fall  alone,  and then  with the average values
 for summer,  winter and spring.   The  results  of this analysis indicate
 that some additional  capacity would  be needed  to handle the increased
 sludge volumes,  but that  the major impact would  be on  secondary treatment
 facilities.   It  is estimated that  the fall seasonal loading would result
 in a 44 percent  increase  in total  solids  and 20  percent increase in BOD
 entering the aeration basin.   These  increased  concentrations would there-
 fore require either increased aerator capacity or  subsequent microscreen-
 ing to provide equivalent removal  efficiency.  Microscreening affords the
 cheaper of the two alternatives ($2,000,000  as against $3,000,000 to in-
 crease the size  of the aerator)  and  thus  is  included in our recommended
 treatment scheme.
                               Cost  Analysis

As mentioned previously,  the  intent of this  analysis  is to provide a
treatment scheme which  could  handle the higher concentrations of pollu-
tants occasioned by  ground domestic refuse,  and yet not jeopardize the C
classification of  the Connecticut River at Bondi  Island.  The costs
associated with such a  proposal  are necessarily a prime consideration in
determining the feasibility of the  wet systems concept.
                                    41

-------
According to the construction specifications of Springfield's consult-
ants,51 the Bondi Island plant will require a capital investment of
$36,865,000.  Our computer analysis provides a somewhat lower figure of
$34,460,912.  One reason for this discrepancy is that the computer pro-
gram relies on national construction cost averaged to predict individual
plant costs, yet at the present time Naw England costs exceed the na-
tional average.  Also, in the case of particular plant operations, the
two cost figures may not describe identical processes.  For example, the
consultant's specification of sludge conditioning and filtration costs
may include more than just wet air oxidation and vacuum filtration, which
the computer uses to establish its cost figure.  Moreover, it would be
unreasonable to expect a program that is focussed on preliminary design
to yield identical results with those of the final construction specifi-
cations.  Table 6 provides a breakdown of the capital expenditure required
for analysis.  Using an interest rate of 6 percent and a 25 year amortiza-
tion period, the capital cost would range between 11.50^/1000 gallons
(3^/1000 fc) and 12.25^/1000 gallons (3.25*/1000 £).   Operation and main-
tenance costs are projected to be 5.8^/1000 gallons  (1.5^/1000 &), such
that the total cost of treatment would be 17.3 to 18.05^/1000 gallons
(4.6 to 4.7^/1000 A) of sewage treated.

If the individual plant processes were to be kept as presently planned,
but enlarged to handle ground refuse, the total treatment cost of these
same facilities is estimated at 30.64{/1000 gallons  (8.04^/1000 £), or
slightly less than twice the projected present cost  for sewage alone.
But, as mentioned previously, the introduction of ground domestic refuse
would require modification of a number of plant processes, particularly
those dealing with sludge handling, in order to avoid problems of clog-
ging and inefficient operation.  Therefore, it is more instructive to
compare existing costs for sewage alone with those associated with the
recommended wet systems treatment scheme.

The initial capital cost of this modified treatment  facility (Figure 70
is estimated at $49.4 million.  Operation and maintenance would run in
the neighborhood of 7^/1000 gallons (1.85f'/1000 &),  such that the total
treatment cost would be only 23^/1000 gallons (6^/1000 £) or just more
than 1.3 times the projected cost for sewage alone.   The major cost
savings would result from the replacement of the wet air oxidation unit
with a conventional anaerobic digestion system.  There was no provision
in the cost analysis for income derived from methane gas production.
Additionally, the computer model of the recommended  treatment stream was
run to simulate equivalent removal efficiency for a .4 percent by weight
mixture of ground refuse in sewage, corresponding to a doubling in refuse
generation rate to 3.58 Ibs/cap/day (1.62 Kg/cap/day).  Despite this in-
crease, economies of scale results in an estimated cost increase of only
30 percent, up to 29.9^/1000 gallons (8.0^/1000 A),  -A summary of the
cost data for treatment of ground refuse in sewage is provided in Table 7.
                                    42

-------
             Table 6:   Treatment Costs for Sewage Alone;
                Planned Bondi Island Treatment  Plant
Item
Consultant's
Estimated
Project Cost
Computer
Estimated
Project Cost
Grit and screenings
Primary settling tanks
Secondary treatment
facilities
Sludge pumping
Sludge conditioning and
filtration
Sludge thickening
Chlorination
•
Electrical and
Instrumentation
Yard piping
Site preparation
Land acquisition and fees
TOTAL CAPITAL COST
$ 520,000
4,160,000
12,630,000
650,000
5,980,000
840,000
1,250,000
3,250,000
2,600,000
4,290,000
695,000
$36,865,000
$ 127,270
4,076,472
11,177,441
352,833
4,823,516
2,175,906
890,763
3,250,000*
2,600,000*
4,290,000*
696,711
$34,460,912
$/1000 gallons
TOTAL AMORTIZATION COST
TOTAL 0 § M COST
TOTAL TREATMENT COST
12.25
5.8
18.05
11.50
5.8
17.3
 Reproduced consultant's figures.
                                43

-------
Table 7:  Treatment Costs for Sewage Containing
             Ground Domestic Refuse
Item
Grit 5 screenings
Primary settling
tanks
Secondary treatment
facilities
Sludge pumping
Sludge conditioning
$ filtration
Sludge thickening
Chlorination
Microscreen
Electrical and
Instrumentation
Yard piping
Site preparation
Land acquisition
and fees
TOTAL CAPITAL COST
Proposed
Plant
.2% Refuse
$ 128,225
4,140,375
16,906.792
551,872
13,542,624
4,190,909
889,166
	
4,991,556
3,993,245
6,588,854
695,989

$56,619,607
Recommended
Plant
.2% Refuse
$ 72,951
3,972,556
15,764,124
885,936
7,876,073
1,356,862
828,845
2,388,640
4,991,556
3,993,245
6,588,854
696,221

$49,415,863
Recommended
Plant
.4% Refuse
$ 84,026
3,953,085
20,940,392
1,159,230
13,951,909
2,216,045
812,324
2,017-, 394
5,557.500
4,446,000
7,335,900
693,835

$63,167,640
i/lOOQ gallons
TOTAL AMORTIZATION
COST
TOTAL 0 5 M COST
TOTAL TREATMENT COST
18.84

11.80
30.64
16.0

6.60
23.0
21.10

8.80
29.90
                     44

-------
                                Conclusions

Our results for the cost of treatment  of ground refuse-sewage mixtures
are at first glance at considerable  variance to those  of Metcalf  § Eddy.52
If one excludes land acquisition  and site preparation,  our  calculations
show an increase in capital cost  from  $31.91 million to $42.63 million,
or by a factor of 1.33; whereas Metcalf § Eddy anticipated  an increase
from $1.6 million to $4.1 million, or  by a factor  of 2.5

There are several explanations  for this discrepancy..  First, the Metcalf
§ Eddy estimates are for a facility  of 1 mgd (3,785 m3/day) whereas our
estimates are for a 64 mgd  (242,240  m3/day)  facility.   Second, the Metcalf
§ Eddy estimate was based on  a  refuse  concentration of 0.5  percent by
weight.  Indeed, our calculation  for an 0.4 percent by weight addition of
refuse, corresponding to a 4.5  Ib/cap/day (2.04 Kg/cap/day) generation
rate, shows an increase of 1.72,  which approaches  the Metcalf § Eddy
figure.

The approach taken in our analyses might perhaps be questioned on the
basis that the use of design  relationships based on sewage  inputs are
not necessarily valid for ground  refuse additions.  Whilst  conceding that
this may well be so, we have  at least  used a rational  quantitative basis
for our inferences and cost estimates, which, surely,  must  be preferable
to intuitive qualitative judgments.
                                    45

-------
                              SECTION VIII

                            SLUDGE HANDLING*

                              Introduction


Although the treatment plant simulations described in the previous sec-
tion were judged adequate to determine overall treatment costs, a more
detailed investigation into sludge handling was deemed necessary, because
the unit operation routines in the treatment plant simulation model were
operating at solids concentrations in excess of their original design.
Perhaps the most important specific goal of the detailed study was to
establish the unit cost of sludge processing and disposal.  The cost of
collection and disposal of residential refuse under existing management
concepts lies in the range of $20 to $60 per ton; consequently, if the
cost of just the sludge handling in a wet system is also $20 to $60 per
ton (of dry solids), as would be expected on the basis of existing prac-
tice at sewage treatment plants, then the economic feasibility of the wet
systems concept is questionable, even given that the transformation of
solid waste to sludge solids is not 1 to 1.

It should be noted, however, that our efforts should be interpreted as
only a preliminary effort to analyze a very complex problem, arid will
need to be extended considerably before even a pilot scale demonstration
facility is contemplated.  The major constraint limiting both the scope
of the analysis and its accuracy was the exceptionally limited experi-
mental evidence available to us.  However, we are conficent that our pre-
dictions of total costs are entirely adequate for the primary- purpose of
the contract which is to establish the order of magnitude of cost for
the wet systems concept as a whole, even if we have not been able to
address the myriad of operational complexities in sufficient detail to
make definitive judgments on process selection and operation.


                            Sludge Quantities

The major factor affecting the quantity of sludge produced at a treatment
facility handling ground refuse is. the per capita refuse generation rate.
Other factors to be considered include sedimentation efficiency, the con-
version of dissolved organic matter to settleable biomass and the per
capita generation of sewage spiffs.  In the case of an average raw refuse
generation rate of 2.25 Ib/cap/day (1.02 Kg/cap/day), the addition of
ground refuse to the sewage will result in about a fifteenfold increase
in the quantity of primary sludge.  On the other hand, because of a very
small increase in the BOD in the primary settler effluent due to the
ground-refuse, the volume and dry solids content of the waste activated
     o
      Sludge Processing and Disposal is the subject of a separate report
by Kos, Joyce and Meier, (Ref.  21).
                                   46

-------
sludge will change much  less.   Under the same condition (2 25 Ib/cap/day
of raw refuse) there will  be only a 45 percent increase of the quantity
of waste-activated sludge,  which is still in the range of variation ex-
hibited by existing sewage treatment plants.  The quantity and the con-
centration of the mixture  of primary and waste-activated sludge is there-
fore affected primarily  by variations in the primary sludge.   The amount
of dry solids of the sludge mixture of primary and waste-activated sludge
in tons per year, as a function of population and refuse generation rate,
is shown on Figure 8.  The report by Kos et al.21, should be  consulted  for
the derivation of these  assertions.

In order to dramatize  the  impact of the ground refuse upon the sludge
handling, quantity-concentration flow diagrams of a conventional sewage
treatment plant and a^sewage treatment plant handling ground  refuse in
addition to sewage were  constructed.  Assuming a sewage generation rate
of 100 gal/cap/day  (378.5  J$cap/day), Figure 9 shows the consequent flows
(in the percentages of the incoming sewage) and their solids  concentrations,
Similarly, Figure 10 shows the flow diagram of a treatment plant handling
100 gal/cap/day of sewage  plus 2.25 Ib/cap/day of ground refuse.

Comparison of these diagrams reveals the magnitude of the impact of ground
refuse additions.  For example, the volumetric flow rate of vacuum filter
cake shows an increase from 0.04 to 0.34 percent of the influent volume.
This implies that at a 10  mgd (37,850 cu m/day)  facility,  the filter cake
would increase from 4,000  to 34,000 gal/day (15,140 to 128,690 Vday).
For the total raw refuse generation rates of 1.0, 2.25, or 4.5 Ib/cap/day
(0.454, 1.02, or 2.04  Kg/cap/day), a 2.5, 4.3 and 7.8-fold increase of
sludge quantities must be  expected.  Using a dry solids basis (which is
unaffected by concentration variations of the sludge)  rather  than a
volumetric measure, the  introduction of ground refuse into the sewage
under the same conditions  as above will result in the 4.7,  9.3 and 17.5-
fold increase of dry sludge solids, respectively.  Note that  the dif-
ference between the volumetric and dry solids increases are due to the
increase in concentration  of the primary-secondary mixture.   In summary,
the sludge handling facilities of a sewage treatment plant handling an
average 2.25 Ib/cap/day  (1..02 Kg/cap/day) of ground refuse will correspond
to a conventional treatment plant handling a population that  is approxi-
mately nine times larger.


                              Sludge Quality

Information on the quality parameters of ground refuse sludges is very
limited, consisting of studies conducted by Metcalf & Eddy as part of
the original Foster-Miller work, and isolated information from a few
"studies whose focus was  energy production from solid waste.   In particuT-
lar, there appear to be  no measurements of quantitative sludge parameters
such as coefficient of compressibility or specific resistance, both of
obvious importance to  assessments of dewatering characteristics.  This  is
an area that is in urgent  need of more detailed investigation if the wet
systems concept is to  be further pursued.
                                     47

-------
       7 -r
       6 • •
        5 • •
       4 • •
   CO

   u.
   o
   §
   o
       2 •
                                    500,000



                                 POPULATION
1,000,000
Figure 8:   Dry Sludge Solids as  a Function  of  Population Size and Refuse

           Generation
                                    48

-------
               PRIMARY
               CLARIFIER
       AERATION TANK
                  SECONDARY
                  CLARIFIER
         o
         H
                  CM
                  M
                           Q«I0049%
                           C*
                            0" -89%
                            C»
                       0*140%
                       C*30OOppm
                               0>
                               c«
                      0« 40%
                      C* 10,000 99m
               THICKENER
DIOESTER
VACUUM FILTER
TRUCK HAUL
LANDFILL
DISPOSAL
Figure 9:  Flow Diagram'of Sewage Treatment Plant Handling 100 gal/cap/day

-------
                           PRIMARY
                           CLARIFIER
                                      AERATION  TANK
                                                                                  SECONDARY
                                                                                  CLARIFER
in
o
0=100%
C= I860 ppm
                                                     Q»40%
                                                     C= lO,OOOppm
                                                                                              0 = 99.66%
                                                                                              C=40ppm
                                                                                           Q=.34%
                                                                                            C=25.%
                           THICKENER
                              DIGESTER
VACUUM
FILTER
TRUCK HAUL
                                                                                    t=l
LANDFILL
DISPOSAL
         Figure 10:  Flow  Diagram of Sewage Treatment Plant Handling  100  gal/cap/day plus 2.25 Ib/cap/day

-------
Raw Sludge.  Metcalf § Eddy53 found that  1-hour settling of ground  refuse
in sewage produced a sludge of about  5 percent  solids  content   Much of
the suspended solids, consisting mainly of paper products,  did not  settle
or only very slowly.  However, the dewatering characeristics of this pri-'
mary sludge (including the floatables) appeared to  be  good.  Metcalf §
Eddy tested three devices for dewatering  the  primary sludge produced by
adding ground refuse to sewage, although  tap  water  rather than sewage
was used in these laboratory tests:   a Smith  and Loveless sludge concen-
trator, a Komline-Sanderson vacuum filter,  and  a Bird  centrifuge.   All
three devices worked well, and neither the vacuum filter nor the centri-
fuge required addition of chemicals.  For all three methods tested, they
obtained a sludge cake of at least 25 percent solids content.  The  yield
of a vacuum filter was at least 8 pounds  per  square foot per hour (39.06
Kgs/sq.m/hr), and Metcalf § Eddy judged that  it could  serve as the  only
dewatering unit.

The specific resistance and coefficient of compressibility  of sludge are
quantitative measures of its dewatering characteristics,  but unfortunately
Metcalf § Eddy did not evaluate them, and these are as yet  unknown  for
the sludges obtained when ground refuse is added to sewage.  The vacuum
filtration yield reported by Metcalf  and  Eddy is in the  range of expected
yields of chemically conditioned primary  sludge,  and therefore the  de-
Watering characteristics of primary sludge of a sewage treatment plant
handling ground refuse will be considered to  be the same as those of
chemically conditioned primary sludges.   This assumption is not unrea-
sonable in the light of some recent experiments at  the University of
Texas by Garden and Malina   on the efficacy  of pulped newsprint as a
sludge conditioner.  Pulp additions of about  100 to 125  percent of  the
weight of sludge solids were found to reduce  the specific resistance of
activated sludge to levels comparable to  the  optimal levels attained with
ferric chloride or cationic polymers. It was noted that the conditioning
effect of the newspaper pulp increased as the solids content of the,
activated sludge increased, and it was concluded that  the chemical  condi-
tioners could be replaced by the paper pulp with essentially similar re-
sults.  This means that we might expect paper fiber in the  primary  sludge
to act as a conditioner for the activated sludge fraction.

There is no information on the thickening of  the sludge  that would  be
generated in a sewage treatment plant handling  refuse,   On  the basis of
the apparent similarity between filtration characteristics  of the primary
sludge at sewage treatment plants handling ground refuse and between
ordinary chemically conditioned primary sludge, the thickening of these
two kinds of sludges may also be assumed  to be  similar.

The remaining problem is the effect of adding a small  portion of secondary
treatment sludge upon the character of primary  sludge.   It  is usually re-
ported that the specific resistance or vacuum filter yield  is less  than
for the mixture of primary and activated  sludges than  for primary sludge
alone.  On thecbasis of the  laboratory thickening experiments, Konicek
Kos and Pardus55 found, however, that the addition  of  activated sludge to
                                    51

-------
primary sludge in the ratio occurring in a conventional secondary  treat-
ment plant had little significant effect on the thickening of primary
sludge.

In  light of these facts and because of the very high ratio of primary to
waste-activated sludge at the treatment plant handling refuse, the addi-
tion of waste-activated sludge to the primary sludge will most likely .have
little or no effect upon the thickening and filtration characteristics of
the primary sludge.  Consequently, the characteristics observed for pri-
mary sludge by Metcalf § Eddy may be assumed to hold also for the  primary-
secondary mixture.

Digested Sludge.  Pfeffer, who used the filter-leaf technique, found  that
the residue from digestion of refuse-sewage sludge dewatered very  easily,
without addition of conditioners.56  This is in contradiction to the
findings of Wise,57 who found that the digester residue dewatered  only
with difficulty.  However, the reactor solids concentrations used by Wise
were below 2 percent, whereas Pfeffer's were generally above 6 percent.
In  the absence of a comprehensive study of the residue after digestion
of  refuse-sewage sludges, we must assume that the quality of the digested
refuse-sewage sludges will be the same as that of the ordinary digested
sludge.  Although it is recognized that this assumption is subject to
question, it is unavoidable in the absence of any actual data.  But inso-
far as this asssumtion does imply a need for chemical conditioner addi-
tions prior to dewatering, it is probably conservative.


                                Analysis

As  in the case of treatment, mathematical modelling and simulation was
chosen as the method by which the answers to some of the questions re-
lating to cost would be examined.  In order to keep the simulation efforts
within reasonable time and cost limits,  four likely process configurations
were selected in advance, all purposely involving only traditional tech-
nologies.  This was justified on the basis that cost functions for many
of  the newer processes such as wet oxidation or pyrolysis are not pre-
sently available in the absence of a sufficiently widespread practical
experience.  But more importantly, perhaps, is the inability to make pre-
dictions of process performance.  Whilst it is relatively easy to predict
the consequences of ground refuse addition to, say, thickening or vacuum
filtration, in the absence of even laboratory data, similar predictions
for the more sophisticated technologies would be quite speculative.   In-
deed, we are already at the limit of reasonable prediction by including
digestion as one of the alternatives.

This is in no way to imply that these technologies are inferior or even
more expensive than the traditional ones to which we have addressed our
detailed analysis.  Rather, our goal is to establish a reasonably  accurate
picture of the costs involved in the handling and disposal of ground  ref-
use sludges, in order that the economics of the entire wet systems concept
can be established.
                                    52

-------
Considerations in the Selection of Alternative  Process  Configurating
Because of the dramatic increase in the  ratio of primary  to  secondary'
sludge at a facility accepting ground refuse, it was earlier concluded
that the addition of secondary sludge to the very much  larger quantity of
primary sludge would not alter the latter's physical-chemical characteris-
tics.  Consequently, separate treatment  of these sludges  does not appear
reasonable.  Many new treatment plants provide  different  treatment pro-
cesses for primary and secondary sludge  streams; as noted in an earlier
chapter, the planned additions to the treatment facilities in the Springfield
area will use air floatation thickening  for the waste-activated sludge
and gravity thickening for the primary.   It is  questionable  whether such
a configuration would be beneficial for  handling ground refuse sludges,
especially from the viewpoint of scale economies.  Therefore, the initial
operation for all process configurations is a mixing of the  two sludge
streams.

Because activated sludge is currently the most  widespread secondary treat-
ment method, we restricted our focus to  that particular source of second-
ary sludge.  The assumption of a 1 percent solids concentration of the
waste activated sludge may well be questioned,  but in view of the wide
diversity of concentrations encountered  in practice, and  the fact that
the much higher primary to secondary sludge ratio tends to make subsequent
computations relatively insensitive to the assumed secondary concentration,
the 1 percent figure does not appear unreasonable.

The impact of the addition of advanced waste treatment  processes to a
secondary treatment plant must doubtless be considered  in the very near
future.  And while the sludge production varies considerably with the
process selected, there is as yet insufficient  information upon which to
base a rational choice of advanced waste treatment at a plant handling
ground refuse.  Consequently, we may rationalize our omission of advanced
treatment sludges not only on the basis  that the quantity involved would
be small in comparison to that from the  ground  refuse (that  will settle
as primary sludge), but also that the uncertainties involved are still
too great for the selection of any one process  to have  real  meaning.

Three of the process configurations chosen for  study are  illustrated on
Figures 11, 12, and 13.  The fourth configuration, including both anaero-
bic digestion and incineration was also  examined, but was found to be
quite uneconomic.

Model Representation and Cost Functions.   Mathematical  models for all of
the unit operations considered were constructed for inclusion into the
computer simulation model.  Most of the  cost functions  built into the
individual unit operation models were from the  study conducted for EPA
by Black § Veatch,58 whereas the design  relationships were taken from a
variety of published sources or derived  by the  writers.   Details of the
model representations for each of the unit operations considered can be
found in Chapter III of Kos
                                      53

-------
                                                                                   ALTERNATIVE  I
in
                             CO
                             K

                             i
                                                         FILTRATE
                         GRAVITY
                        THICKENER
SLUDGE
HOLDING
 TANK
VACUUM
 FILTER
TRUCK
 HAUL
LANDFILL
 DISPOSAL
         Figure  11:   Sludge Handling,  Alternative 1

-------
                                                                                         ALTERNATIVE  2
tn
tn
                     GRAVITY
                     THICKENER
DIGESTER
SLUDGE
HOLDING
 TANK
VACUUM
 FILTER
TRUCK
 HAUL
LANDFILL
DISPOSAL
             Figure 12:  Sludge Handling,  Alternative 2

-------
tn
                    I	
                                oc
                                a:
                                a.
                                                                                            ALTERNATIVE  3
bJ
(9
O


in

O
UJ




I


UJ
                                                            FILTRATE
                           GRAVITY

                           THICKENER
     SLUDGE

     HOLDING

      TANK
VACUUM

 FILTER
INCINERATOR
INCINERATOR

 RESIDUE

 DISPOSAL
            Figure 13:  Sludge Handling, Alternative 3

-------
The simulation model was then  run for each process configuration under a
wide range of operating parameters,  sludge solids concentrations  size of
facility, input solids generation rates,  and the details of the results
plotted and analyzed in detail in Chapter IV of Kos  et al.21

Comparison of Alternative  Process Configurations.  The costs of the  three
feasible process configurations are  summarized on Figure 14.  Alternative
1, involving thickening, vacuum filtration and land disposal, is the indi-
cated choice, with a cost  of between $10  - $15 per ton ($11.02 - $16.53
per Kkg) of dry solids for facilities serving in excess of 100,000 popu-
lation.  If a digester is  added (Alternative 2), total costs will be about
$5/ton  ($5.51/Kkg) dry solids  higher, an  increase that is almost offset
by the value of the digester gas produced.  Under the assumed conditions
of a landfill at 10 miles  (16  km) distance, incineration as a means  of
disposal (Alternative 3) is shown to be another $4/ton ($4.41/Kkg) more
expensive.  Digestion together with  incineration was dismissed as an
economically infeasible alternative.

Tables  8, 9, and 10 summarize  the costs of these three alternatives  as  a
function of population size and per  capita refuse generation rate, with
an indication of the corresponding cost for a facility treating only
sewage  sludge.  It is quite evident  that  the unit cost of handling sludge
at a facility handling ground  refuse is substantially lower than if  only
sewage  sludge were handled.   This is  due to the economies of scale re-
sulting from larger absolute quantities involved, and the very different
physical nature of the sludge  to be  dewatered and disposed.

Optimal Solids Concentrations.  Our  analyses show that in those cases
where the total cost is sensitive to solids concentration,  such as those
process configurations involving digesters, an optimum appears to lie  in
the vicinity of 14 percent solids.   If this appears high,  as it doubtless
does in comparison to solids concentrations encountered at existing
treatment plants, one should be reminded  of the drastically different
nature  of the sludge involved.  A feed of such a high solids concentration
is essential whenever digesters are  used, as the cost of digester capacity
would otherwise be prohibitive.  Where digesters are not used,  the decrease
of vacuum filtration cost  resulting  from  higher solids concentration is
balanced by the increase in the cost required to thicken the sludge  to
the higher concentration,  and  no optimum  solids concentration can be
^identified.

Digestion.  Our calculations show that the cost of anaerobic digestion  is
somewhat higher than the resultant value  of methane produced, and that  the
cost of process configurations using landfill for ultimate disposal  is
insensitive to whether or  not  digestion is included, given a methane
value of $0.80 per 1000 cu ft  ($28.50/100 cu m).  A lesser methane value
would make the use of digesters unlikely, whereas an increased value,  or
consideration of the more  intangible benefits of increasing energy sup-
plies given the current energy shortages, such as benefits to the balance
of payments, would tend to make digestion more attractive.   All this, of
                                     57

-------
       70
    9
    $
    oc
    Q
    O
    W
    8
       60
        50-
       40
        30
       20
                       "VALUE OF DIGESTER GAS
        10
                                      500,000

                                   POPULATION
1,000,000
Figure 14:  Total Cost of Sludge Handling  (Alternative  1,  2  and  3)  as  a
            Function of Population; for Average Refuse  Generation  Rate of
            2.25 Ib/cap/day
                                     58

-------
tn
         Table  8;   Total  Cost  of Sludge Handling  (Alternative  1)  in  $/ton  Dry Solids  for Various Populations
                                            and Various Raw Refuse Generation  Rates
Raw Refuse
Generation Rate
Ib/cap/day
0
(Sewage Siudge Only)
0.5
1.0
2.0
2.25
3.0
4.0
POPULATION
10,000
80.75
39.16
30.35
23.92
23.04
21.13
19.51
50,000
41.96
22.79
19.21
16.26
15.99
15.07
14.37
100,000
34.23
19.59
16.72
14.55
14.31
13.66
13.09
250,000
27.68
16.58
14.47
12.97
12.76
12.30
11.93
500,000
23.98
14,95
13.27
12.10
11.95
11.60
11.33
1,000,000
21.03
13.69
12.36
11.47
11.35
11.11
10.92

-------
Table 9:  Total Cost of Sludge Handling (Alternative 2) in $/ton Dry Solids for Various Populations
                              and Various Raw Refuse Generation Rates
Raw Refuse
Generation Rate
lb/cap/day
0
(Sewage Sludge only)
0.5
1.0
2.0
2.25
3.0
4.0
POPULATION
10,000
134.58
56.30
42.71
33.46
32.25
29.64
27,47
50,000
53.94
30.95
26.31
22.93
22.47
21.58
20.70
100,000
43.25
26.52
23.20
20.87
20.54
19.86
19.27
250,000
34.72
22.93
20.68
19.07
18.84
18.36
17.94
500,000
30.57
21.09
19.34
18.07
17.89
17.48
17.09
1,000,000
27.32
19.74
18.31
17.20
17.03
16.61
16.20

-------
Table 10:  Total Cost of Sludge Handling (Alternative 3) in $/ton Dry Solids for Various Populations
                               and Various Raw Refuse Generation Rates
Raw Refuse
Generation Rate
Ib/cap/day
0
(Sewage Sludge only)
0.5
1.0
2.0
2.25
3.0
4.0
POPULATION
10,000
265.74
107.61
77.88
56.89
54.08
48.01
42.92
50,000
87.67
44.47
38.29
31.27
30.30
28.17
26.37
100,000
64.15
37.14
30.99
26.35
25.72
24.37
23,29
250,000
46.28
28.94
25.24
22.75
22.44
22.56
21.80
500,000
37.96
25.30
22.94
21.89
21.64
21.45
21.20
1,000,000
32.36
24.04
22.05
21.26
21.04
20.77
20.64

-------
course, is contingent on the assumption that anaerobic digestion  is  in
fact feasible, an assumption that is yet to be convincingly demonstrated.
As noted in a recent paper on management aspects of methane production
from solid wastes,59 reliability of supply will play a crucial role  in
successful marketing of the recovered energy.  But skeptics who cite the
traditional unreliability of sewage treatment plant digesters as  a key
argument against digesters might be reminded that sophisticated physical-
chemical sludge treatment methods, such as the wet oxidation processes
currently very much in favor, are surely at least as dependent on trained
and experienced operator personnel as are digesters.

The constraints of this study have not allowed as detailed an investiga-
tion of the trade-offs in digestion as will be necessary if digestion
were to be seriously considered in an actual implementation.  In  particu-
lar, the physical characteristics of the digesting sludge require identi-
fication if factors such as mixing cost are to be adequately analyzed.
Other areas not fully explored include the impact of climate (which  has
a substantial impact on heating costs), the relationship between  deten-
tion time and reliability, the impact of recycling the digester super-
natant to the secondary treatment stages, and nutrient addition.  A
comprehensive study focussed on a. complete economic model of digesters
would thus be recommended as a segment of further investigations  into
sludge handling.

Disposal Method.  Of the two disposal methods examined by our studies,
incineration is shown to be preferable to direct landfill disposal only
if the distance to the landfill exceeds 45 miles (72.4 km).  The  magnitude
of this critical distance, of course, is dependent on many contributory
factors, and would tend to increase as the cost of air pollution  control
equipment drives up the cost of incineration, and would tend to decrease
with increased land or unit transportation costs.  But, given the politi-
cal realities as they presently exist, it is questionable whether a  city
that cannot find a landfill site within its own city limits or with  an
immediately adjacent community could ever use a more distant landfill,
regardless of cost.  Thus, all that can usefully be concluded in  general
terms is that both alternatives are technically feasible, and that their
"cost is of comparable magnitude.  Indeed, it would appear that the wet
systems concept offers little significant advantage over conventional
practice with respect to the ultimate disposal problem.  And although ad-
vanced technologies such as pyrolysis might offer a solution, such
technologies are not dependent on a wet collection method.
                                    62

-------
                                SECTION IX

                ENVIRONMENTAL AND SOCIOECONOMIC IMPACTS*

                           Environmental Impact


The conclusions with  respect  to environmental impact are necessarily
qualitative, rather than  quantitative,  and we have not  attempted to at-
tach dollar values to the environmental costs and benefits  identified
And although the conclusion of the analysis is that the net impact will
probably be beneficial, there are one or two potentially serious impacts
which cannot yet be assessed  due to lack of data.   Thus,  even if accepted
methodologies were available  for the dollar measurement of  intangible
environmental impacts, inclusion of these dollar values into the engir
neering cost comparisons  would be presently quite premature.

One of the difficulties in assessing the environmental  impact of the wet
systems concept is the need to assess the totality of environmental quali-
ty.  Thus, air quality might  improve at the same time that  water quality
might diminish, yet there is  no accepted way to effect  the  trade-off in
even qualitative terms.   Is one additional unit of BOD  in a river worth
the decrease of one unit  of particulate emissions?  Clearly, such questions
can only be answered  by reference to a specific situation and perspective,
and the goals and priorities  that govern the evaluation.


                                Air Quality

Air Pollution from Conventional Practice.   There are two  major potential
sources of air pollution  from conventional refuse collection and disposal
practice:  the emissions  from collection trucks, significant especially
in the central city,  and  emissions from incineration or illegal open
burning in dumps.

With respect to emissions from collection trucks,  the conditions under
which they must operate make  for very unfavorable emission  conditions.
Internal combustion engines burn fuel most efficiently  at relatively
constant speeds, and  the  amount of unburned fuel increases  sharply when
a motor vehicle is accelerated or decelerated.   Emissions of hydrocar-
bons, carbon monoxide and particulates  may be 50 to 100 percent greater
for average speeds less than  10 miles per hour (mph)  as compared to
30-35 mph, and especially idling, acceleration and deceleration are as-
sociated with high emission levels.

In-central urban areas where  distance traveled by collection trucks is
slight compared to time spent, the contribution to urban air pollution


     *See Meier, Kuhner,  and  Bolton, (Ref.  22) Chapters VI, VII and VIII
for detailed discussions.
                                    63

-------
from the collection fleet may be in excess of what it would appear to be
if total miles traveled is used as a measure.  For example, in the Spring-
field study area, the miles of travel by the 21 collection trucks may be
conservatively estimated at 500 miles by map measurement.  For each truck,
the measured distance included length of the assigned route plus allowance
for two trips to the landfill site.  Since vehicular travel in Springfield
amounts to over 1.1 million miles per day60 this 500 miles is negligible.
However, considering the substantial time spent idling and the fact that
emissions during idling and speed changes may be greater than cruising
emissions by factors of 2 to 6, the total contribution of the fleet to
automobile air pollution may be on the order of 0.5 percent of the total.

Although neither refuse incineration nor open burning occurs in the Spring
field area, the impact of such practices elsewhere is well recognized.
The costs of municipal refuse incineration are expected by many to
double as enforcement of emission standards becomes widespread,62 and the
impact of open burning was demonstrated all too tragically by the fatali-
ties of a recent multiple car crash on the New Jersey turnpike reportedly
caused by aggravation of fog conditions by smoke from a nearby dump.63

The Impact of Wet Systems on Air Pollution.  Introduction of wet systems
disposal for residential refuse would greatly decrease the number of
refuse trucks on the streets.  Collection by vehicle could be eliminated
for all but those crews necessary to handle the nongrindable fraction of
the refuse.  Estimations discussed previously show that between three
and four pickups per year will be required to handle nongrindables, re-
quiring at most four collection trucks.  Assuming identical unit emis-
sion levels would imply that air pollution from collection truck exhausts
would be decreased 75 percent from present levels.  A corresponding de-
crease should be experienced in support vehicle travel, such as vehicles
operated by foremen and other supervisory personnel.  .On the other hand,
it is expected that bulky item collections, currently involving two crews
a day in Springfield, will require more vehicles for operations in con-
junction with wet systems implementation, perhaps doubling in size.

Wet systems collection would virtually eliminate air pollution from on-
site burning.  The availability and ease of disposal of rubbish into the
grinder should drastically decrease the incentive for on-site burning,
making the enforcement of existing ordinances a more feasible proposition.
But since it is primarily garden waste that continues to present incen-
tives for on-site incineration in residential areas, the practical
feasibility of having to carry such wastes into the house and to the
basement grinder location is somewhat questionable.  Nevertheless, im-
mediate removal of such wastes by grinding may well offset this apparent
inconvenience.

Because the wet system would landfill refuse in the form of a sludge that
may have 70 percent moisture or more, opportunities for open burning are
clearly diminished relative to filling raw refuse.  To the extent that
                                     64

-------
the refuse sludge would be incinerated rather than landfilled, an addi-
tional source of pollutant emissions might be created; but there is no
reason to assume that  control of emissions at incinerators burning sludge
is more difficult or costly than control at incinerators burning raw ref-
11«A .                                                            °
                               Noise Quality

Noise from Conventional  Refuse Collection Practices.   Collection  vehicles
generate noise  from the  truck motor, the compaction mechanism and the
crew during the loading  process.   Solid waste collection trucks travel
their routes rarely at more than walking speed,  and high sound levels are
associated with both stopping, due to the "sigh" of air brakes, and
starting, due to revving the motor as the clutch is released.  The truck
motor must be revved during movement from stop to stop and during the
compaction cycle.   This  produces a loud, low frequency sound of varying
length in the first case, of approximately 30 seconds duration in the
latter, and sound levels of 90 dBA or higher depending on the type and
condition of the truck.   When the motor is engaged for compaction, a
second source of.sound is added to the motor noise.  This comes from the
compaction mechanism itself and sounds associated with the compression
of  loose rubbish.   Although specific references are not available, it
may be that higher estimates of noise production from collection  opera-
tions found in  the literature of 100-110 dBA may be associated with this
peak output level.

The noise resulting from the banging and clanging of  the rubbish  cans
while being emptied may also be quite severe.  This occurs not only when
the collectors  are carrying and emptying the containers,  but also earlier
in  the morning  when residents are bringing their cans to the curbside.
Large steel drums tend to be more offensive since they are often  too
heavy for the collectors to lift and so are dragged out to the truck
causing the high frequency rasping of metal against pavement.  They are
then banged against the hopper lip to assist in emptying.   Less noise is
produced if rubber, plastic or fiber containers are used instead  of gal-
vanized metal.   Plastic and paper bags can be carried or dragged  and
loaded with little or no noise.

It  should be pointed out, however, that the annoyance associated  with a
particular noise level is closely related to its duration, and thus the
aforementioned  impacts are not necessarily as great a nuisance as might
be  supposed.  Nevertheless, a substantial proportion  of complaints under
newly introduced noise ordinances concern refuse collection,05 and efforts
are now underway for the promolgation and enforcement of federal  noise
emission standards for collection vehicles.

Observations in the Study Area.  In the Springfield,  Massachusetts study
                                    65

-------
was significantly different from all other models regarding noise  level
production.  The compaction mechanism in this truck is a cylinder  in which
a corkscrew shaped blade turns continuously.  Since compaction is  con-
tinuous, the truck motor is constantly idling at a high rate.  The blade
gradually forces the waste material deeper into the cylinder, compacting
the rubbish with new rubbish as it is loaded in.  This continuous  circu*
lar path of the rubbish against the corkscrew, cylinder walls, and breaker
blade in the truck causes a great deal of noise as the material is smashed
and splintered into smaller pieces.  Glass and metal pieces are notably
noisy as they fall by force of gravity from the top of the cylinder.
                          **
Conversations with several Springfield residents living on routes  ser-
viced by the screw compactor trucks showed the people were aware of the
increased noise from this type truck.  One gentlemen remarked that he had
not complained because the truck was faster during loading than others
and so the noise was of a more temporary nature.66  This was a valid
point since the collectors do not have to halt loading operations  for a
compaction cycle.  This interview also verified the tendency of people
noted earlier to discount disturbing noise if it is temporary.  The
majority of collection trucks in Springfield are rear loading hydraulic
compactors of American manufacture, which are relatively quiet while
idling.  We observed that at every third stop a compaction cycle was
necessary, at which time the truck motor is speeded up and the compaction
mechanism is engaged;  As noted above, the two loudest sources are thus
operative simultaneously.

Impact of Wet Systems on Noise from Conventional Collection Vehicles.
Since wet systems collection will require vehicular collection only of
nongrindables, collection frequency will decline.   With collection of
refuse by vehicle decreased from once a week to an average of once every
16 weeks, the frequency of noise from collection vehicles would obviously
decrease.  In neighborhoods which may be near collection or disposal
facilities, corresponding decreases in vehicle traffic could greatly
lower the ambient noise level.  Perhaps the most significant effect
might be felt in dense urban areas where refuse collection contributes
significantly to both noise levels and traffic congestion.

Wet systems will not of course affect noise levels when the trucks do
collect in a neighborhood.  In fact, because the frequency of collection
would be much reduced, the presence of collection trucks might be much
more noticeable.

Impact of the Wet System on Noise Levels in the Home.  The major noise
impact of the wet systems concept is that of increased noise levels in
the home, and the household refuse grinder designed by Foster-Miller in-
corporates several noise reduction features.67  it is a wet process
grinder which reduces noise from the tumbling of loose and flyaway pieces
of refuse in the hopper.  The grinder as designed is supported on  an
elastomer vibration isolation pad.  Additionally,  the hopper and top of
the device are coated with a sound absorbing material to reduce grinding
noise.  In fact, noise levels might be less than from garbage grinders,
                                    66

-------
since the latter has no  lid to mask the grinding noise,  and,  for conven-
ience of installation  in sink, the motor and mountings may not  be shrouded
or mounted on dampers.   Foster-Miller estimates their appliance as de-
signed would be slightly more noisy in operation than a  kitchen dish-
washer,0  with the noise levels estimated to lie in the  range of 65-75
dBA.

Since it is recommended  the grinder be placed in the basement or other
service area, the operator would be probably the only person  receiving
direct effects of the  noise.  The operator's exposure to the  noise would
most likely be infrequent and of brief duration.  This moderates the
relatively high noise  level which might be encountered.   The  direct ef-
fects during operation would include speech interference;  operators
would probably avoid communication at this time, much the same  as the
operator of a vacuum cleaner presently does.  But, if the actual noise
level falls in the upper levels of the estimated range,  hearing damage
risk cannot be dismissed.  Although the average exposure to such levels
would be limited to fractions of hours per week, certain elements of the
population might be subject to prolonged exposure.  For  instance, apart-
ment maintenance men could be exposed to several hours of this  noise
daily.  Although this  exposure would not constitute an imminent hazard
to  hearing, it could nonetheless hasten eventual hearing damage in the
context of cummulative exposure from many sources.  However,  unlike
other noisy appliances,  the operator for the grinder need not remain in
the same room when the machine is grinding.  Sound insulation afforded
by  moving to an adjacent room would reduce noise to acceptable  levels.

Sleep interference would probably be minimal.  In the household,  opera-
tion times of the appliance could be arranged to avoid conflict with
times persons are usually asleep.  Due to service area or basement
placement, noise  levels from the grinder in bedrooms would probably be
negligible.

Annoyance is probably  the most significant of any indirect consequences.
Operator annoyance  is  usually not as severe as annoyance to persons to
whom the noise is  intrusive.  In single family homes, this effect is
deemed  slight since  family members will have control over when  the ap-
pliance is used.  And  as mentioned previously, frustration may  not occur
when the ability  to  stop or reduce intrusive noises is possible.   Neigh-
bors outside the  single family home would not be exposed to significant
noise levels due  to  shielding effects of the structure.

Noise standards  for appliances are currently only a function  of market
place pressures.   A survey of appliance manufacturers by the  EPA showed
that with the exception of air conditioners, there is little  customer
pressure for quieter appliances.69  Indeed, sales of machines such as
vacuum  cleaners  show strong demand for noisier machines; perhaps the
American public  equates power with noise.

In  spite of the  weakness of market pressure, many man*********> ^e
research and development projects aimed at noise reduction.  Most felt
                                    67

-------
further reduction was possible with present technology, but that a 5 -
20 percent cost increase would be the penalty for quieter appliances.70
It thus seems reasonable to conclude that noise from the refuse grinder,
while noticeable, may not be serious.  Provided they are of short dura-
tion, high noise levels are accepted by householders, and are not re-
garded as health hazards.
            Public Health and Visual-Aesthetic Considerations

The impact of refuse collection practices on public health and on the
visual-aesthetic environment are closely related, being both largely de-
pendent on refuse storage methods preceding collection, and hence a si-
multaneous consideration of both elements is indicated.  A wet systems
implementation would largely eliminate the problems of storage of refuse
awaiting collection, and a major beneficial impact on public health and
the visual-aesthetic quality of the environment can be expected.

Storage Facilities.  One study indicates that at least among low-income
families, areas of multi-family dwellings may contribute less waste per
capita than families in less dense, single-family structures,7* and thus
storage space required for each household may be greater in single-family
neighborhoods.  This will be offset to a large degree by the greater
availability of places to store refuse in these areas.  The typical
occupant of a single-family detached home stores his refuse in trash
cans, barrels, or bags.  These are commonly placed outside the dwelling
in a spot out of view such as the back or side of the structure.  In row
houses and apartment buildings, a central deposit area or receptacle is
often provided.  Location may vary from outside a parking lot to indoors
in the basement of the building.

For both single-family and multi-unit dwellings, storage locations can
influence visual-aesthetic and public health impact.  Outside storage
relies on the refuse receptacle itself to hide the refuse, keep it from
pests, and prevent litter and odor.  An inside location, whether inside
the dwelling, in the basement for example, or enclosed in an outside
location, could compensate for inadequate containers.  It should be
noted, however, that insect and rodent vectors may not be impeded by any
but the most secure enclosure.

Observations in the Study Area.  Although Springfield has adequate
legislation to ensure proper refuse disposal, the ordinance is rarely en-
forced.  The vast majority of containers placed on the curbside were
inadequate in one or more of the four categories mentioned above.  Pre-
dictably, these inadequacies were most notable in the poorer sections of
the city, but present to a lesser degree in all sections.

In all but the most well-to-do suburban sections of the city, the major-
ity of rubbish containers were not the standard 30-gallon galvanized
trash cans with lids.  Prevalent in numbers were plastic trash bags
followed by galvanized 30-gallon cans without lids, then industrial metal
                                   68

-------
drums of 50-55 gallon capacity, corrugated cardboard boxes, paper super-
market bags, pressed cardboard drums of 40-60 gallon capacity and ««»ii
standard 30 gallon cans with lids.  No wet-strfngth papS bags made   ll
especially for refuse were observed.

The most popular storage methods in low income and redeveloped areas of
the city observed are plastic bags, large metal and cardboard drums,
paper supermarket bags and corrugated cardboard boxes, in that order.
The plastic bag, often lauded for saving collectors' time, we found to
be frequently misused.  The abuses include stuffing the bag so full it
tears, not tying the bag closed, and placing articles in the bag which
puncture and tear the bag.  These abuses, when they occur, render the
bag useless in preventing litter, odor, pest breeding and feeding.   The
plastic bag properly used should be stored in a rodent and dog-free area
since these creatures can easily make the food inside available by
tearing the plastic.  The plastic bag may retard animals somewhat by re-
taining odors which let the animals know food is near.  However, bags
stored habitually in the same place may not even be effective in this.

The  large  drums  were found to have several disadvantages.  They are not
manufactured  for rubbish and so have no lids, making their contents
available  to  rodents, flies, and dogs.  Litter is generated from these
cans  in  another  way.  The cans are so large and often so heavy that they
are  tipped on their side and rolled by the collectors from curbside to
the  packer.   This tipping often results in refuse being spilled out of
the  top  of the  drum since the drums are packed full to overflowing.  The
 collectors seem  to resent these drums and may not bother to pickup this
 spillage.  As an alternate to tipping, the drums are often dragged to
 the  truck.  The  cardboard containers develop holes in their bottoms from
 this,  and are 'also susceptible to wetness from rain and wet garbage.
 The  other two containers common in the low income areas, paper supermar-
 ket  bags and  cardboard boxes, are obviously deficient in all areas.  Ad-
 ditionally, when wet from rain or garbage, these bags and boxes often
 rip  when picked up, depositing their contents on the roadway or alley.

 It is interesting that our survey teams noticed that the care taken in
 packaging the refuse at curbside was reflected in the care taken by the
 collectors in loading the refuse and in cleaning up spills and litter.
 Consequently,  in areas where inadequate containers cause spills, where
 rubbish and garbage is dripped through wet paper sacks, torn bags, and
 tipped drums, collection crews are  likely to leave the mess right where
 it falls.   On spills which were their own fault, the collectors were
 usually diligent in cleaning up their mistakes.

 The above phenomena may have a correlation with individual pride.  In
 the single family neighborhood, more space is available for Placement of
 refuse for pickup.  A messy, untidy pile of rubbish is ^^^"f*
 the home owner.   In multi-dwelling  units where less space  is P«»sent. un
 sightly communal storage  areas do not reflect as directly  on an  Divid-
 ual occupant, hence the  individual's sense of responsibility is  absent.
                                un-
69

-------
Cross Connections.  Perhaps the most serious potential public health im-
pact, and the one that would certainly attract close attention from
public health authorities, is the possibility of a faulty cross connec-
tion in the grinder.73  This possibility exists because of the need to
provide fresh water for the flushing cycle that must follow grinding.74
Cross connections are a recognized public health hazard, and major epi-
demics of water borne diseases in the developed countries almost invari-
ably follow the incursion of wastewater into the water distribution
system via some form of cross connection.

However, it could be argued that the stringent laws and regulations con-
cerning cross connections were developed of necessity in a time when
water and wastewater systems were introduced to urban areas haphazardly,
and without adequate supervision.  But if the grinder is designed in such
a way as to inherently avoid possible hazards, the anticipated fears of
the public health authorities might well be allayed.  The engineering
principles of cross connection control are well understood,75 and no
insurmountable difficulties are anticipated in adhering to these in the
design of refuse grinders.

Impact on Rodent Control.  As all putrescible components of refuse would
be quickly disposed of through the grinder, no burden is placed on the
householder to safeguard the refuse or garbage against vector infestation.
Refuse would, in the storage and collection phases, virtually be elimi-
nated as a source of food for rodents and insects, with the possible
exception of rodent populations in the sewers.  It is conceivable that
the rodent problem in sewers might be increased if more food in the form
of ground garbage and food waste were delivered to the sewers.  Evidence
has been offered to support the fact that a high density of garbage
grinders in a community may be associated with increased rodent popula-
tions. 76  This relationship is uncertain, however, because rodents com-
monly reside in sewers, even when'food sources are on the outside.  In
the study area, serious rat problems occur in areas of combined storm
and sanitary sewers, rather than in the areas where garbage grinders
are prevalent.  Indeed, there is evidence that domestic sewage without
ground refuse may be capable of supporting rat populations.  The Building
Research Advisory Board of the National Academy of Sciences feels that
grinding garbage for disposal to sewers offers a net benefit to control
of a community's rat population.77  And since it is known that shredded
raw refuse in landfills is not suitable for rodent consumption, one can
only surmize that mixed refuse suspended in water would be even less
suitable.

The elimination of curbside refuse storage has some further unexpected
and beneficial consequences.  For example, a recent study linked urban
stray dogs to the urban solid waste problem78 and established the free-
running urban dog as a part of the cycle of urban waste, litter, and
disease.  It was found that the dogs may exist in a symbiotic relationship
with urban rats.  The dogs often assist the rats unknowingly by tearing
bags and spilling trash cans in their own  search  for food, and were ob-
served feeding on refuse  side by side with rats.  To the extent that this
                                   70

-------
food source would be  eliminated, a wet systems implementation might play
a substantial role  in the control of the stray dog population, a problem
that is becoming to be recognized as very serious.


                               Water Quality

The impact of a wet systems implementation on water quality in the Spring-
field area must be  put in the context of existing shortcomings in water
quality management  in the region.  Three areas are currently of concern;
the landfill at Bondi Island, the discharge of stormwater overflows into
the Connecticut, and  the inadequacies of the existing primary wastewater
treatment facility.   Only in the last area are current plans sufficient,
and the new planned secondary treatment facility is about to enter con-
struction.

The refuse disposal site currently used in the study area is regarded by
many as a threat to surface and groundwater supplies.  Located on Bondi
Island in the town  of Agawam, Massachusetts, adjacent to the Westfield
and Connecticut Rivers, most of the site is a flood plain that includes
a post-glacial oxbow  pond.  But recommendations to close this facility
must be put in the  light of the realities of the fragmented New England
governmental structure, and the exceptional difficulty of finding
politically acceptable sites.

In Springfield, combined sewers still serve the downtown and surrounding
older sections of the city while newer suburbs have separate systems.
The impact of stormwater overflows on the water quality of the Connecti-
cut River below Springfield is substantial.  Presently, a two month fre-
quency storm of 30  minutes duration results in the carrying of 10 tons
of BOD into the river.79  Significant reduction of dissolved oxygen
 levels in  the river results, although the depletion is not sufficient to
violate the Massachusetts C water classification of the Connecticut.
Present plans to alleviate this situation were outlined in Section VII,
above.

 Impact of Wet Systems on Water Quality.  Although recognizing the limita-
 tions of the computer model and its results as discussed in Section VII,
 we would not anticipate any difficulty in meeting the same effluent
 quality with ground refuse as will be the case for the new planned facili-
 ty treating just wastewater, and the water quality of the Connecticut
 River, as measured  by conventional parameters, would not be compromised.
 Indeed, to the extent that the Bondi Island landfill site, which might be
 unsatisfactory from a water quality viewpoint, would be used for land-
 filling of a stabilized digested sludge containing no free moisture rather
 than for filling of raw refuse, the net impact on water quality would be
 positive.

 With regard to the  combined sewer overflows, we have already noted that
 the implementation  of a wet system in Springfield would be contingent on
 a prio? or simultaneous separation of combined sewers.  But since we an-
 ticipate that grinders would first be installed in the relatively affluent
                                     71

-------
        OQ
suburbs,   all that would need replacement initially are any combined
interceptors that carry wastewater from outlying areas through the cen-
tral areas where combined sewers are prevalent.  This would avoid any
additional stormwater loads on the Connecticut River resulting from
ground refuse additions.

Need for Revised Effluent Standards.  In view of the rather limited in-
formation on the possible quality of effluents from wastewater treatment
facilities handling ground refuse, the need for revised effluent stand-
ards is difficult to ascertain.  Metcalf § Eddy's experiments do show,
however, that the COD of a conventional secondary effluent would be sub-
stantially higher than present levels, due mainly to fine cellulose
fibers in suspension.  A standard for COD would therefore be indicated
for such facilities, although establishment of the permissible levels
would need further bench and pilot scale studies.  As indicated by our
computer model results, there appears to be little difficulty injecting
existing effluent quality limitations for BOD, suspended solids, dis-
solved nutrients or other parameters commonly in use for sewage treatment
effluents.
                          Socioeconomic Impacts

The prediction of socioeconomic impacts, and in particular predictions in
quantitative terms, is a difficult and complex task.   Two chapters in the
case study report address these impacts at some length,* to which the
reader is referred for further detail, and we shall here summarize only
the more important points.  A detailed econometric analysis of refuse
generation in the Springfield SMSA was also conducted as part of this
study with a view to identifying socioeconomic factors affecting spatial
and seasonal variations in generation rates.  The results of this analysis
are reported separately in a technical paper.  (See Section XIII0

Impact on Private Capital.  A shift from conventional handling to wet
systems would reduce the scope of both private industry and municipal
activity in refuse collection and disposal, and increase municipal ac-
tivity in sewerage.  If the shift were to be widespread, it would replace
some frequent private house-to-house collection in sewered areas, leaving
only occasional collection of nongrindables.  Thus, some investment op-
portunities will be lost to private enterprise.  But as any shift seems
bound to be very gradual, and as the total private capital is small rela-
tive to the national economy or to any particular metropolitan area, the
effect will hardly be noticeable in the aggregate in any one year.  Our
conclusion here is more optimistic than in the case of labor (see later
discussion), because the owners of capital are undoubtedly more capable,
on the average, of adjusting to the decline.  They can turn their talents
and money elsewhere.  A very important factor in smoothing the adjustment
is the market for used equipment.  There is national advertising of used
      Meier, 'Runner and Bolton, (Ref.  22),  Chapters  VII  and  VIII.
                                   72

-------
equipment, and  it  can be transferred to other geographical areas at a
cost which is small  relative to its value (a ten-year old collection ve-
hicle might still  sell for $3,000 to $4,000 in the second hand market).
Used equipment  is  bought and sold by private operators, municipalities,
new equipment dealers, and firms specializing in reconditioning and re-
fitting older vehicles.81  There will still be need for landfills, and
the general scarcity should prevent serious decline in value.  Naturally,
there may be some  difficult adjustments and permanently adverse conse-
quences on a few entrepreneurs, but neither the waste of capital,  nor
the losses to individuals, should be serious enough to count for much in
the aggregate picture.

Unemployment.   As  the primary cost-reducing effect of wet systems  is in
the elimination of labor in conventional collection,  the unemployment
problem naturally  comes to mind.  Indeed, pressure from workers and
unions to avoid reductions in municipal forces, and public sympathy for
unemployed workers,  will be an important barrier to adoption of wet
systems.  It might be especially strong in a city like Springfield,  which
is in a labor market area which has had unemployment well above the na-
tional average  in  many years.  In April and May 1973, for example,  the
unemployment rate  in the. Springfield-Chicopee-Holyoke labor market  area
was 6.8 percent (in June the flood of students looking for jobs raised
it to 8.2 percent, but that was a seasonal increase).  The average
monthly rates were 8.5 percent in 1971 and 7.9 percent in 1972.8*

The concern with unemployment is entirely valid in principle.  However,
we feel it is not  a very important consideration in the present context.
Any adoption of wet systems should come gradually, and some of the  workers
 laid off  in  refuse operations will find jobs in the sewerage operation of
 the same  city,  or  in other public works departments.   The small number of
 workers laid off in any given year would easily be absorbed elsewhere as
 normal attrition opened up new vacancies.  It should be stressed that the
 case will not be one of massive layoffs, as have occurred or been  threat-
 ened in cities  where refuse collection has been drastically altered over-
 night, such  as  by eliminating backyard collection, or by reducing  the
 frequency of collection, and resulted in drastic reduction in crews.

 There  are about 225,000 workers in the labor force in Springfield's
 labor market area, and the few workers who would be released outright,
 not having been transferred to other jobs in the city, would have  an
 almost negligible effect on the total picture.  Even a fairly rapid
 changeover to wet systems by Springfield would take 5 to 10 years,  and
 the number of workers released outright could be as low as 20 per  year.
 Furthermore, refuse work requires no special skill or experience,  so the
 workers will not suddenly find themselves with obsolete human capital
 which  they have built up at cost to themselves.  Obviously, the city
 would find  it  desirable to smooth the workers' transition to new jobs,
 perhaps with special unemployment payments and certainly by careful
 phasing  in  of new methods.  The point is that such extra costs would be
 very small  and  should not be a significant barrier to the new method.
                                     73

-------
Impact on Municipal Finance.  On the municipal side, the net financial
impact is to decrease expenditures out of current income and to increase
expenditures financed by borrowing.  The net effect of this depends on
the monetary cost comparisons between the two alternative methods -
counting only the monetary cost to the local government.  This depends on
the total monetary cost comparisons in the economy as a whole and on the
shares of the costs which local governments must pay, and also on whether
waste reclamation brings net profits to municipalities.

A very important factor will be state and federal policy on grants for
sewerage facilities construction, which, unfortunately is very difficult
to predict.  The question is whether such grants will be as liberal when
facilities handle solid waste.  On the one hand, there is a long tradi-
tion of sewerage grants, and the federal stance was reaffirmed in the
1972 legislation which mandated higher standards to be achieved.  On the
other hand, this tradition has been essentially lacking in solid waste.
This is not an arbitrary distinction; the principle has been that sewerage
aid was needed to prevent pollution of environmental resources, which
benefitted numbers of people far greater than the population of a single
town or city.  This principle is much less applicable to solid waste,
although not totally irrelevant there.  Especially the federal government
would find it hard to justify aid for facilities to handle solid waste,
although it has a clear role in research, demonstration, regulation of
packaging, incentives for reclamation, etc.

Whatever the federal and state governments decide to do will be crucial
for whether wet systems are introduced extensively.   For many areas, the
total costs of the two alternatives are bound to be closely matched, and
even a small change in the percentage of cost defrayed by grant will tip
the scales.  This is not to deny that wet systems will sooner or later
make some headway in some areas even without any grants to cover the
added sewerage costs, or that even extensive grants will not be suffi-
cient in some other areas.

There is another aspect of grants policy which bears mention.   All federal
sewerage grants are now being made conditional on the municipality levying
user charges, which distribute costs to users in proportion to their loads
on the system, and sufficient to cover all operation and maintenance (in-
cluding replacement of components) costs.  Some municipalities will have
to increase user charges over what they would want to charge in the absence
of needing the grant, others will have to institute sewer charges for the
first time.  This will accentuate the difference between conventional
handling and sewage treatment in the relative reliance on user charges,
to which we now turn.

It is not inevitable that conventional methods be financed more by taxes
than user charges.  Nor would wet systems automatically result in more
reliance on user charges.  The technical nature of handling does not dic-
tate financing; we note for example the great variation from city to city
in the use of charges in both solid waste and sewerage.  However, it
seems appropriate to compare the two systems taking as given the present
differences in financing of government costs, as shown by the national
averages.

                                    74

-------
The picture is clouded,  however, by the heavy subsidy from state and
local governments  in  sewerage.   Do we measure reliance on charges by
their proportion of private costs plus municipal costs, or their propor-
tion of total national  costs?  The second proportion is much lower.  The
present subsidy is composed of two parts.  The first is the federal tax
treatment of municipal  interest, which applies to all local capital ex-
penditures financed by  borrowing, sewerage among others.  The second
is the construction grant,  which is tied to sewerage p_ejr se_.   The first
part might be justified on  very general grounds--it is essentially a
form of revenue sharing.  The second part might be justified by con-
sidering rivers and lakes as national resources, and assuming localities
deserve to be offered heavy subsidies for avoiding polluting those re-
sources.  The subsidy does  not now seem to be offered as an incentive
to avoid pollution, because recent legislation has moved toward pro-
hibiting that in any  event.

The first part of  the subsidy would continue to operate if sewage treat-
ment plants were enlarged to handle refuse—the extra capital cost would
be covered by borrowing at  the same interest rates as on other municipal
borrowing.  We shall  assume here that the second part would not be ex-
tended to cover such  extra  capital costs, that is, we assume  that locali-
ties will have to  finance the extra costs of handling solid waste on their
own.  This is a reasonable  assumption given the histories of  federal poli-
cies on sewerage and  solid  waste, and the already high grants projected
for sanitary sewerage alone.  We also assume that grinding costs will not
be subsidized, certainly not by state and federal governments.   All these
assumptions imply  that  there definitely will be more reliance on user
charges in wet systems  than there is at present, resulting in an improve-
ment in efficiency.

One important characteristic of wet systems is that the transactions
 costs of using the price system are much lower in wet systems than under
 conventional methods, if the load on the sewerage system can  continue to
be fairly well measured by  water flow.  Presently, water flow is commonly
 assumed to be an equitable  proxy for sanitary sewerage costs.   It is
 widely used to determine sewerage charges, and is expressly permitted in
 the new federal regulations on acceptable charge schemes for  facilities
 aided by construction grants.83  If some households have grinders and
 others do not, water  flow-based charges can be used for all households,
 but the level of rates  would be higher for those with grinders (their
 charges for conventional collection would be lower, of course).

 Water use is already  measured for many households and the additional
 clerical and administrative costs (and complaints) would be very low.  If
 water  is not already  metered, the advantages of simultaneously measuring
 water usage,  liquid waste loads, and solid waste loads would  justify the
 expense of  introducing  metering, for the transactions costs would contrib-
 ute to efficiency  and equity for several public services and  a much larger
 block  of scarce resources.

 Water  use  is  certainly not  a perfect measure of waste loads under wet
 systems.   But a  charge  based on water use would be preferable to no
                                     75

-------
charge at all or a charge completely unrelated to waste flows.  The goal
in cost assessment cannot be perfection, for the transactions costs are
too high for that; it is merely to strike a balance which improves ef-
ficiency and equity without using too many resources in measuring and
billing.  The scope for this seems considerably greater for the wet sys-
tem approach than for conventional management.  (More research is needed
to determine the accuracy of water flow as a measure of sewerage costs
under wet systems.  Some of the research suggested in Section XI would
also accomplish this result.)

A by-rproduct of wet systems and user charges would seem to be improvement
in efficiency of use for water in other purposes than grinding, for the
amount of money spent on water by a typical household would be more no-
ticeable than at present and make it worthwhile to economize on its use.

Income Distribution Effects.  The change toward user charges identified
above will have an extremely small effect on income distribution.  User
charges will replace taxes, and the net effect will depend on the rela-
tionship between income and collection costs on one hand, and between in-
come and tax burdens on the other.  Neither relationship can be specified
precisely.  Under the most plausible assumptions, the change would prob-
ably cause some extremely small change toward more inequality if all
families had grinders and user charges replaced taxes in financing munici-
pal costs.  However, if adoption came about voluntarily, lower income
families would not necessarily be absolutely worse off.  If only upper
income families adopted grinders, and paid higher user charges, the effect
would be less inequality, because the higher user charges would replace
some taxes paid by lower income groups.  But whatever the effect may in
fact be, its magnitude is likely to be quite insignificant.

Collective Bargaining.  Various aspects of collective bargaining offset
the choice between conventional systems and alternative systems.  Labor
relations .are relatively very important in conventional collection, be-
cause the function is inherently labor intensive and is also important
enough for public health and general amenity that disruptions due to labor
disputes are very costly.  Strikes in some functions have little more ef-
fect than to delay production of useful output for a time.  The only cost
is from the change in time of availability of a product, and that cost
may be minor.  In refuse collection, however, as in some other public
services, even relatively short delays can have serious consequences.
Disruptions in sewerage could also be costly, but the system is more
mechanized and depends much  less on constant, massive labor inputs; this
would still be true even if  sewerage were expanded to carry and treat
solid waste.  A work stoppage in sewerage, although possibly costly if
treatment must be partially  cut back, is less likely to cause a public
emergency than a refuse collection stoppage.

The nature of the occupation and the vulnerability of the city to work
stoppages contribute to union activity and use of strikes or threats of
strikes as means of winning  higher wages in refuse collection.  Refuse
collection is no different in this from many  other activities in the
                                    76

-------
economy.  However, for  a  long time,  legal prohibitions and public opinion
inhibited unionization  and  strikes among public employees.  More recently,
some of the legal barriers  have been falling.

In 1940, Ziskind   could  already note that strikes by refuse collectors
were especially common  relative to other government employees:  "The
removers of refuse, the pariahs of the municipal personnel, have mani-
fested a group consciousness  and a maturity of strike techniques that have
surpassed by far the protective efforts of the higher paid municipal em-
ployees."  While he listed  refuse collection strikes in 26 cities,  Ziskind
devoted most attention  to their history in New York and Chicago.  In 1907
and 1911 in New York, strikes of several days duration were broken by the
importation of strike breakers, which occasioned violence.  In the 1920's
and 1930's, there were  seven  strikes in Chicago, some several days  long,
over a variety of issues, such as the attempt of the depression-plagued
city to reduce forces or  hours and its failure to pay full wages,  dis-
putes over rights to jobs in  return for political activity (for instance,
in 1933, "a spectacular struggle between politicians and trade-union
bosses," p. 89), and control  of employment at private disposal sites where
the city dumped its refuse.

Ziskind also commented  on strikes of sewerage workers, contrasting  their
near-negligible effect  on the public with the effect of refuse workers:
"The strikes have...never seriously interfered with the normal functioning
of existing sewer systems," although they may have retarded the extension
or improvement of a system  (p. 95).

Over the 1960's, the number of strikes in refuse collection,  and man-days
idle in strikes, increased  greatly,  along with other strikes  by other
government employees.   Refuse collection has continued to be  one of the
government functions most plagued by strikes,  although it comes nowhere
close to the schools in that  respect.85  Both the costs of strikes  and
the time, energy and resources expended in collective bargaining must be
considered costs of conventional methods of solid waste handling, and
costs which will grow as  unionization increases employee awareness  of
wage parity; employees  in one function make demands with a view to  what
others  are demanding or have  already won at the bargaining table.86  The
unionization of a function  for the first time thus has a true marginal
 effect  greater than indicated by the resources devoted to bargaining in
 that function alone.  Another aspect is the predominance of black  labor
 in  some municipal collection  forces.  Some labor disputes have been aggra-
 vated by more general civil rights grievances.  And it is not only  wage
negotiations which absorb time and energy; negotiations over working con-  gg
 ditions are also costly,  in time and possibly in inefficient  practices.   »

 All of  these factors may  make alternative methods appear attractive to
 many.   But one must not forget that alternatives which reduce the  scope
 for labor disputes may  have the inevitable consequences of altering the
 balance of political power  in a city, consequences which are impossible
 to  evaluate on economic criteria alone.  The true significance of public
 employee unionism may be  as much political as economic.  It offers some
                                     77

-------
group  (public employees or blacks) the chance to increase their political
power.  One therefore cannot unequivocally say that an alternative like
wet systems would be superior.

Wellington and Winter note that the costs of collective bargaining are
less severe when public functions, including refuse collection, are per-
formed by private firms, even if the firms' employees are unionized.°9
There may be times when economic and political forces encourage near-
general strikes by all public employees, although such have been rare in
fact.  Public refuse collection may be stopped even when the collective
bargaining issues are less relevant in refuse collection than in some
other functions.  Also, private employers may resist union demands more
than public employers, because they are better organized and have more
experience in bargaining, and because the latter are more open to politi-
cal pressure to yield.  Governments are organized in ways (division of
powers, checks.and balances) which make it difficult to present a united
front.  Even if a city government is forced to intervene in a strike of
a private firm, the government is a neutral intervenor and not a direct
adversary of the union.  The authors also argue that a private employer
can offer a more credible threat of unemployment as a result of higher
wages - he is better able to substitute capital for labor and is more
likely to raise prices to cover higher costs and thus permit a slower
growth in demand.  And governments seem more able to pass on higher costs
to other levels of government, or to hide the increased cost in a complex
and little understood budget.  For these reasons, a government is more
likely to meet a strike with a higher settlement, and thus to be struck
in the first place.90  The authors thus conclude that whether a particu-
lar function is appropriate for government now depends in part on the
growth of public employee unionism, and they note refuse collection is one
example where the choice between public performance and private firms de-
pends partly on unionization.  Similarly, the relevance of unionization
for conventional handling depends somewhat on whether it is performed by
public agencies or not.

All this may be true, but it remains ambiguous how to value these factors.
While administrative costs and wages may be kept lower when private firms
perform the function, employees are deprived of an avenue of economic
power.  This consideration may be important in assessing conventional
methods versus alternatives.  The wet systems method certainly inherently
offers municipal workers somewhat less scope for exercising political and
economic power.   We can say, however, that the total numbers of workers
affected by this would be small relative to the total of municipal em-
ployees, even if large scale changes were affected.  Because any adoption
of wet systems is likely to be gradual, the quantitative significance of
loss of power will be even less.
                                    78

-------
                                 SECTION X

                      TOTAL  SYSTEM COST COMPARISONS

                          Economic Feasibility


The estimated total system costs,  on an annual per household basis,  for
both the wet system and  a continuation of existing conventional  practice
are shown on Table 11.   The  1988 costs, which are shown in terms of  1973
dollars, have been derived on  the basis discussed in Chapter VIII, uti-
lizing a two percent annual  increase in labor costs.  A comparison of
the system totals is even more unfavorable to the wet system than the
generalized comparisons  of Phase I (where the 1973 figures were  esti-
mated at $125-$182 for the wet system as against $43-$101  for conventional
practice).  Evidently, conventional waste management practices in Spring-
field are very efficient, with per household costs at the  very bottom of
the national spectrum.

But it should be reiterated  that the $80-$105 annual cost  of grinding is
that for the Foster-Miller or  Nebiker-Meier grinding concepts, both of
which were rejected on the basis of excessive cost.   Moreover, this is
an annualized cost, which, again as noted previously,  is not the basis
upon which individuals would base a decision on grinder purchase.  Con-
sequently, the totals inclusive of grinding do not appear  very useful.

If the cost of grinding  is excluded, on the assumption that  grinding costs
would be borne by the householder, then the remaining costs  are  those to
just the public sector,  and  these are considerably more favorable to the
wet sys.tem, possibly 10  to 15  percent lower than conventional  by 1988.
Because this margin is rather  small, the conclusion would  be that imple-
mentation of the wet systems concept is still very distant.   Yet  on the
other hand, if our prediction  of potential consumer acceptance is correct,
implying heavy involvement of  the private sector appliance industry in
developing and marketing refuse grinders, then it behoves  the  public sec-
tor to examine the consequences on its waste management systems,  as it
might otherwise be faced with  the necessity of banning refuse  grinders.

Other than the grinding  cost component, most of the uncertainty  in the
wet system cost is attributable to the cost of sewer maintenance.  As
explained in Section VI, it  is simply not yet possible to  determine the
necessary flushing frequency;  yet it is the flushing frequency that pri-
marily determines sewer  maintenance costs.   If flushing is necessary
four times rather than only  once per year,  then the $6-$9  range  assumed
in Table 11 for 1973 would rise to $24-$36.

A problem perhaps inadequately addressed in the economic analysis is that
of transition costs, i.e., the costs during that period when both conven-
tional systems and wet systems coexist.  The cost items for sewage treat-
ment, sewer maintenance  and  grinding might be assumed to rise linearly
with increasing number of installed grinders; however, it  is unclear
                                    79

-------
                 Table  11:  Total Annual System Cost per Household Comparisons  for Springfield
00
o
1973

Refuse Collection
Collection of Nongrindables
Bulky Refuse Collection
Disposal
Sewer Maintenance
Grinding
Treatment
Sludge Handling
Sludge Disposal
Total
Total to Public Sector
(Excluding Grinder)
Conventional
$17.50
•*
1.00
3.50
2.00
-
9.00
5.50
0.75
$39.25
$39.25
Wet
System
$2.50
2.00
1.00
6.00-9.00
80-105
12.50
8.50
2.00
$114.50-142.50
$ 34.50-37.50
1988
Conventional
$21.50
1.50
4.50
3.75
-
10.00
6.00
1.00
$48.25
$48.25
Wet
System
$4.00
3.00
1.50
7.50-11.00
80-105
14.00
10.00
2.50
$122.50-151.00
$ 42.50-46.00

-------
whether the cost items  for collection and disposal would also decrease
linearly, and thus  the  possibility of a higher than expected transitional
cost cannot be discounted.   It may therefore be only at the point of com-
plete implementation  of the wet system that the expected wet system costs
are attained.  But  until such time as the necessary detailed data on
refuse characteristics  becomes available, data that was unavailable to
us, any attempt at  quantifying the transition would be excessively specu-
lative.  This is not, however, to deny the importance of the problem.


                               Conclusion

Having shown the economics of the wet system to be somewhat unfavorable
over existing and alternative concepts, having raised serious questions
as to its technical feasibility,  but also having shown some significant
(even if unquantifiable)  environmental and socioeconomic benefits,  what,
then, is our final  judgment at the conclusion of our investigation?

Unfortunately, the  complexity of the concept and its place  within  the
many interlinked efforts at rational environmental management make  any
definitive pronouncement on the concept at the present time a very  dan-
gerous undertaking.   The major reason for our reluctance to express
absolute conclusions  is the limited information on the technical aspects
of the concept that is  to date available.  Taken at face value, the re-
sults of the economic analysis presented in these pages is  indeed un-
favorable to the wet  systems concept, showing approximately a doubling
of total per household  cost for all the elements of domestic waste manage-
ment.  Yet at the same  time, we have shown that if the cost of the  grinder
is borne by the householder, there may be situations where  the cost of the
wet systems concept to  the public sector may be less than any conceivable
improvement of today's  concepts.   Thus, given the premise of technical
feasibility, we arrive  at the conclusion that its economic  feasibility,
by which we mean that cost is no greater than that of its competitive
alternatives, rests upon the willingness of householders  to purchase a
grinder for benefits  of convenience and aesthetic or environmental im-
provement of their  homes.  Such a premise is not to be dismissed lightly,
given the indications of continuing increases in disposable income.  And
skeptics might be reminded of the realities of the current  market for trash
compactors, an appliance that surely offers fewer convenience benefits
than refuse grinders.

Certainly we may conclude that demonstration scale research investments
are as yet unwarranted.  At the same time,  however,  we believe that a
limited program of  further studies directed specifically  at  the resolution
of a few key technological uncertainties would allow judgments that are
based on more substantial evidence than the largely circumstantial  infer-
ences that are available to us at the present time.   If these further
studies, as elaborated  in the following section,  have  favorable results,
then we believe that  the wet systems concept would gain consumer acceptance
to the point where  implementation became feasible.   This  is contingent,
however, not only on  a  confirmation of technical  feasibility,  but on the
availability of a grinder whose cost is compatible with consumer acceptance.
                                     81

-------
                              SECTION XI

                 RECOMMENDATIONS FOR FURTHER RESEARCH


Our investigation has pinpointed several areas of technical uncertainty,
and consequently definitive judgment on the technical feasibility of the
ideal are made only with great difficulty at this time.  However, it has
been possible to analyse these uncertainties to the point where a program
of a few selected, but pertinent, investigations could resolve these un-
answered questions, and to this end we have set forth a program of re-
search tasks.  The sequence of tasks is ordered in such a way that the
most critical questions be addressed first, and so that further invest-
ments may be terminated if the results of the most critical investiga-
tions are negative.

Because these unresolved technical questions are quite substantial, we
feel that investment of research monies as the next step, at the level
required for demonstration scale project would be premature, and a deci-
sion on such investments should be preceded by the program of laboratory
scale studies herein described.  A detailed analysis of the quality para-
meters and conventional treatibility of ground refuse-sewage mixtures may
be regarded as a first priority, and ,is already under consideration as an
in-house study at the EPA National Environmental Research Center in
Cincinnati.  The information yielded by this study will considerably ex-
tend the existing knowledge on the subject developed by the Metcalf and
Eddy investigations, and will allow much more reliable estimates to be
made for the likely cost of treatment.  There remain to be addressed,
however, the following series of tasks, presented in the recommended
sequence.  Note that Task 5 is probably the most costly, and should
therefore be undertaken after the other, relatively cheaper, tasks.

Task 1:  Hydraulic Studies Under Unsteady Flow Conditions.  The most im-
portant known limitation to the technical feasibility of the wet systems
concept concerns the ability of an existing gravity sewer system to trans-
port ground refuse.  The results of the Springfield case study show very
clearly that in many sewer laterals, the scour velocities required to
prevent excessive deposition of the glass and ceramic component are
rarely present, if indeed at all.  Moreover, it remains questionable
whether the generally optimistic conclusions of the Foster-Miller studies,
which would appear to show few problems to occur under steady-state con-
ditions, are at all valid for the unsteady flow conditions actually en-
countered in most parts of a gravity sewer system, and in particular, in
the laterals at the periphery of the system.

In order to resolve these problems, a study of hydraulic transport of
ground refuse in a simulated system of sewer laterals is proposed, with
the objective of extending the findings of the existing Foster-Miller
studies to unsteady flow and solids load conditions.  The experiments
should be conducted in a test loop of eight^inch glass sewer several
hundred feet long, and with a representative selection of appurtenances.
                                  82

-------
The injection of ground  refuse should be through simulated building con-
nections spaced at  appropriate intervals along the test loop, and dis-
charges timed to simulate  as closely as possible real conditions.  The ground
refuse injected should include the glass fraction, and the experiments run
over at least weekly cycles  in order that quantitative judgments on accu-
mulation problems can be substantiated.  Although it is recognized that
such a simulation program  may involve substantial effort and considerable
sophistication in laboratory procedures, anything less would not yield
information that will provide conclusive evidence on the feasibility of
hydraulic transport in real  systems.

Task 2:  Field Survey of Refuse Generation and Water Consumption.  The
case study was unable to fully answer the questions of the feasibility of
refuse grinding and sewer  transport because of certain data inadequacies.
In particular, no data is  available on' the refuse generation, water con-
sumption and sewage production of single households.   The use of average
values for groups of households may be dangerous if the degree of cor-
relation between these three variables is unknown.  If in fact households
with high refuse generation  also generate high sewage quantities, the
problem of adequate grinding water will be less acute than is suggested
by use of average values.  Unfortunately, knowledge of this correlation
is crucial to a judgment of  feasibility, and, therefore a field investi-
gation to rectify this deficiency is proposed.  This  field survey would
involve the selection of a series of representative refuse collection
truck routes in the study  area, and sample a series of households on this
route for water consumption  and refuse generation.  The program might
include:

     Weekly readings of  water meters on the day of refuse collection.
     Water meters can generally be read without requiring access to
     the interior of a dwelling unit, thus involving  no inconvenience
     to householders.

     Curbside weighing of  refuse containers on the day of collection
     prior to pick-up by the collection crews.  This  sampling might
     be conducted half an  hour ahead of the collection,  involving no
     impediment to  the collection crews.  Portable scales mounted on
     a suitable push-cart  would be adequate for this  purpose.   By
     sampling only  a portion of the households on each route,  the
     sampling crew  could operate at the same speed as the collection
     crews, thus preserving  a constant difference between the two
     operations.  A certain  number of samples could be collected
     in   plastic bags  for subsequent composition analysis and deter-
     mination of which  components could not be ground by virtue of
     their bulk.

     Determination  of the  maximum flow depth (and hence velocity)  of
     the sewage flow at  manholes along the sample routes.

     Continuous monitoring of sewage flows in a selection of resi-
     dences.
                                      83

-------
     Statistical analysis and interpretation of all data collected.

Because of strong seasonal variations, this study should be conducted
over a 52 week period.  It is, however, probably unnecessary to sample
all households every week; once every four weeks would be sufficient for
all but a few sample households.

Only data collected at this level of detail will allow meaningful con-
clusions on the feasibility of refuse grinding and the development of
adequate grinder specifications.  Ideally, this task should run concur-
rently and be closely coordinated with Task 1, and be conducted in the
same Springfield case study area as used by this investigation.

Task 3:  Energy Balance Studies.  In the 18 months that have elapsed since
the start of this wet systems evaluation, national energy shortages have
reached major proportions.  An evaluation of any radical innovation, such
as the wet systems concept, must therefore include examination of the im-
pact of its implementation on energy consumption, but unfortunately could
not be included as part of this study for reasons of time and resources.
A focus on energy consumption is especially important in view of the
energy market distortions resulting from government controls.   Energy
costs, and total systems costs, may thus not necessarily be an adequate
indicator of the true social costs of the various alternatives.

(At first glance, the trade-off appears to be savings in fuel for collec-
tion vehicles as opposed to increased electricity use for grinding and
treatment plant operations.  However, to the extent that anaerobic di-
gestion and its methane by-product could satisfy a substantial portion of
treatment facility power requirements, the net impact of a wet system on
energy consumption would probably be positive.)

Task 4:  Evaluation of Potentially Hazardous Components in Refuse-Sewage
Mixtures.  Phase I of this investigation established the possibility of
potential problems arising from the addition of potentially hazardous
materials to the treatment plant.  But because there is no actual evi-
dence, only indirect estimates that are inconclusive, it is recommended
that a series of laboratory investigations provide some facts on this
question.  Advanced instrumentation analyses of simulated treatment plant
inflows and secondary effluents should provide the necessary evidence of
higher (or lower) concentrations of specific materials (pesticides, herbi-
cides, toxic heavy metals) resulting from the addition of ground refuse.
Such evidence would further allow appropriate additional advanced waste
treatment processes (or changes needed to already existing processes) to
be identified with some degree of confidence and thus make estimates of
treatment cost in the presence of ground refuse more credible than those
that can presently be made.

Task 5:  Development of Alternative Grinding Concepts.  The results -of
our investigations show clearly that existing grinding concepts envisaged
for wet systems are unlikely to be implemented for economic reasons.  Re-
plumbing of existing housing as would be necessary under the Foster-Miller
                                   84

-------
concept is prohibitively  expensive and must be regarded as infeasible.
On the other hand, grinding  at high consistencies, as called for by the
Nebiker-Meier concept, requires a pressurized building connection,  the
cost of which is also very high.   Moreover, the question of how much
grinding water is available,  and what minimum solids concentrations are
necessary to prevent deleterious effects in the sewer system under  the
unsteady flow conditions  encountered in practice are obviously crucial
to the development of grinder specifications.  Consequently,  additional
effort in this area rests on the results of Tasks 1 and 2.   Since the
major cost of existing grinding concepts lies not in the grinder itself,
but in ancillary arrangements, a successful grinding concept that would
avoid these would have some  chance of success in removing a major barrier
to a wet systems implementation.   One such alternative concept might in-
volve grinding at high consistency, but discharging to the building con-
nections only at such times  when the flow from other domestic appurte-
nances, such as following toilet flushing or washing machine discharge,
would provide adequate dilution water.

Task 5 should be divided  into 3 phases.  In Phase I, alternative con-
cepts should be evaluated with the help of laboratory simulations that
would examine the behavior of refuse slurries and their dilution and
mixing with other domestic wastewaters.  Phase II would develop an
appropriate design which  should be constructed and tested in  a third
phase.  Any demonstration project would obviously depend on the avail-
ability of an already constructed and tested grinder.

Task 6:  Comprehensive Reevaluation of the Concept.   Completion of  the
previously specified Tasks 1 through 5, and the study shortly to be  con-
ducted by the EPA National Environmental Research Center in Cincinnati
that will obtain further  data on the quality parameters and conventional
treatability of ground refuse-sewage mixtures, should allow an evaluation
of the concept that  is founded on much more concrete evidence than has
been available to this study.  This proposed reevaluation would be  in a
position to make a definitive judgment on the economic and  technical
feasibility of the concept,  and, if found to be feasible, would have as
a major goal the planning of a subsequent demonstration project.  This
reevaluation is felt to be very necessary prior to commencement of  a
demonstration program that would involve investments of a very much higher
order of magnitude than the  studies previously conducted.
                                    85

-------
                                SECTION XII

                                REFERENCES

 1.   Clark,  R.,  "Urban Solid Wastes Management:   An Economic Case Study,"
     EPA Environmental Protection Series EPA-R2-73-012,  August 1972.

 2.   Two unpublished memoranda report pn the results of these experi-
     ments;  one  dated September 13, 1963,  entitled, "Report  on a Study
     of the  Feasibility of Disposing of Rubbish  into a Sewage Disposal
     System," the other dated May 26, 1965, by R.  Whitley,  is entitled,
     "Evaluation of Settling Characteristics of  Ground Refuse in Sewage,"

 3.   Waller, D.  H., "An Examination of the Benefits and Disadvantages
     With Respect to the Disposal of Solid Wastes," ASCE Combined Sewer
     Separation  Project, Technical Memorandum No.  10,  February 1968,

 4.   Nebiker, J. and P. Meier, "Hydraulic Disposal of Solid  Wastes,"
     Proc. 1st Annual Northeastern Regional Anti-Pollution  Conference,
     University  of Rhode Island, Kingston,, ,Rb,pde Island, July 1968,

 5.   Etzel,  R.,  "One System Treats Sewage and Solid Waste,"  Chemical  and
     Engineering News, March 23, 1970, pp.  44-45.

 6.   Contract CPE-70-115, originally let by the  Bureau of Solid Waste
     Management, Environmental Control Administration, Department of
     Health, Education and Welfare, in June 1970.

 7.   Guzdar, A,  and S. Rhee, "Feasibility of Hydraulic Transport and
     Treatment of Ground Household Refuse Through Sewers,"  report by
     Foster-Miller Associates to EPA Solid Waste Research Laboratory,
     Cincinnati, Ohio, January 1973.

 8.   Fisk, A. T. and A. R. Guzdar, "Preliminary  Design of a Household
     Refuse Grinder," report by Foster-Miller Associates to EPA
     Solid Waste Research Laboratory, Cincinnati,  Ohio, January 1973.
 9.   McKinney,  R.  and J.  Mahlock,  "An Economic Evaluation of Garbage
     Grinding Versus Surface Collection and Disposal," Water and Sewage
     Works,  Reference Number, November 1971.

10.   The authors were apparently unaware of the Foster-Miller research
     and of recent developments in the pulp and paper industry.

11.   McGraw-Hill's 1972 report on Business and the Environment, pp. 6-7.

12.   Personal communication with Mr.  Maxwell Small of Brookhaven National
     Laboratory, November 1972.
                                    86

-------
13.   Monaghan, D. A. and A. Guzdar,  "Feasibility of Hydraulic Transport
     of Ground Refuse Through  Sewer  Appurtenances," report  by Foster-
     Miller Associates to EPA  Solid  Waste Research Laboratory, Cincinnati,
     Ohio, March 1973.

14.   Courtney, R. and D. Sexton,  "Refuse Collection from Houses and  Flats
     by Pipeline," Proc., 2nd  International  Conf.  on Hydraulic Transport
     of Solids in Pipes, Coventry, England,  1972.

15.   "Pipeline Transportation  of  Solid Waste,"  report by the  Stanford
     Research Institute to  Bureau of Solid Waste Management,  Environmental
     Control Administration, Department of Health,  Education,  and Welfare,
     (Contract PHS-1-DP-1-U1-00208-01)  February 1971.

16.   Zandi, I., and J. Hayden,  "A Pneumo-Slurry System of Collecting  and
     Removing Solid Wastes," in I. Zandi,  Editor,  "Advances in Solid-
     Liquid Flow in Pipes and  its Applications," Pergamon Press, New
     York, 1971.

17.   See, "Pipeline Transport  of  Shredded Solid Waste,"  report by the
     Western Company to the EPA Solid Waste  Management Program,  1971,
     open-file Report SW-36C,  and Overfield, J.  L.,  et al., "Increasing
     Wastewater Flow Velocity  Using  Chemical Additives," J. WPCF, 41, 9,
     pp.  1570-1585, September  1969.

18.   See  for example, Bishop,  F.  W.  and A. E, Drew,  "Disposal  of Hydrous
     Sludges from a Paper Mill,"  Journal of  the Technical Association of
     the  Pulp and Paper Industry, Volume 54, No.  11,  (November 1971), pp.
     1830-1832.

19.  See  for example, Herbert,  W., "Recycling Municipal  Waste,"  Chemical
     Engineering, January 10,  1972,  pp. 66-67.

20.  Meier, P., J. Kuhner and  C.  J.  Martel,  "A  Preliminary Assessment of
     Wet  Systems for Residential  Refuse Collection," report by Curran
     Associates, Inc. to EPA Solid Waste Research Laboratory,  Cincinnati,
     Ohio, March 1974.

21.  Kos, P., M. Joyce  and  P.  Meier, "Economic  Analysis  of Refuse Sludge
     Processing  and  Disposal," report by Curran Associates, Inc to EPA
     Solid Waste Research  Laboratory, Cincinnati, Ohio,  March 1974.

22   Meier  P   J. Kuhner and  R.  E.  Bolton,  "Wet Systems for  Residential
     SE; Collection; A Case Study for Springfield, Mass."  Report by
     Curran Associates, Inc. to EPA  Solid Waste Research Laboratory,
     Cincinnati, Ohio, March 1974.

23   Meier  P    and  G   Fisette, "Modifications  to the Executive Computer
23'  £ ger,m^Steady-State  Simulation of  ^stewater Treatment Facili-
     ties " report by  Curran Associates, Inc.,  to EPA Solid Waste Re
     search laboratory, Cincinnati,  Ohio, March 1974.
                                    87

-------
24.   Fisk, A. T. and A. R. Guzdar, op. cit.

25.   Buying Guide Issues, Consumer Reports 1953 + 1968, Consumers Union
     of the United States, Inc., Mt. Vermon, New York.

26.   Meier, Kuhner and Martel, op. cit., p. 112.

27.   Meier, Kuhner and Bolton, op. cit., p. 42.

28.   See e.g., "Data Acquisition and Analysis System, Quarterly Reports by
     ACT Systems, Inc., Winter Park, Florida to the EPA Solid Waste Re-
     search Laboratory, Cincinnati, Ohio, 1971, 1972 and 1973.  See also
     Meier, Kuhner and Martel, op. cit., pp. 27-29-

29.   Meier, Kuhner and Martel, op. cit., pp. 110-118.

30.   See Meier, Kuhner and Bolton, op. cit., p. 193.

31.   Guzdar, A. and S. Rhee, op. cit.

32.   Monaghan, D. A. and A. Guzdar, op.  cit.

33.   Meier, Kuhner and Martel, op. cit., Chapter III.

34.   "Design and Construction of Sanitary and Storm Sewers," WPCF Manual
     of Practice No. 9, Water Pollution Control Federation, 1969, p. 128.

35.   Guzdar and Rhee, op. cit., p. 90.

36.   Meier, Kuhner and Martel, op. cit., pp. 51-52.

37.   Ibid., p. 52-53.

38.   Metcalf § Eddy, Inc., "Feasibility of Treating Sewage with Household
     Refuse," report to Foster-Miller Associates, Inc., 1971, included
     as part of Guzdar and Rhee, op. cit., Chapter 5.

39.   For further details of this model,  see Meier and Fisette, op. cit.,
     Chapter IV.

40.   "City of Springfield, Massachusetts - Report on Sewage and Industrial
     Waste Disposal," report by Camp, Dresser and McKee, Inc., 1971.

41.   "Supplemental Study of the Connecticut River in Connection with the
     Proposed Bondi Island STP," report by Camp, Dresser and McKee, Inc.,
     1971.

42.   This model of the planned treatment facility is also discussed in
     detail in Meier and Fisette, op. cit.
                                   88

-------
43.  Metcalf §  Eddy,  op.  cit.

44.  Kos,  et al.,  op.  cit.,  p.  13.

45.  Ibid.,  p.  14.

46.  Ibid.,  p.  20.

47.  Ibid.,  p.  69.

48.  Ibid.,  Chapter IV.

49.  Meier,  Kuhner and Martel,  op. cit.t Chapter IV.

50.  The calculations  assume the following average fractions  of refuse
     categories  to be  ground:   Paper 1.0; Garbage 1.0;  Glass  and Ceramics
     1.0;.  Garden Waste 1.0;  Wood 1.0;  Plastics 0.9;  Rubber, Leather  and
     Textiles 0.8;  Metal  0.3, Other 0.1.

51.  "Contract  and Specifications for Bondi Island Treatment  Plant Con-
     struction," report by Camp,  Dresser and McKee,  Inc.,  1972.

52.  Metcalf &  Eddy, op.  cit.

53.  Metcalf §  Eddy, op.  cit.,  p.  4-2.

54.  "Sludge Conditioning -  Can Waste Paper be Used?" Water and  Wastes
     Engineering,  10,  5,  p.  c-10,  May 1973.

55.  Konicek, Z.,  P. Kos  and C.  Pardus,  "Gravity Thickening of Industrial
     Sludges,"  Proceedings,  Conf.  of the Czechoslovak Civil Engineering
     Society, Brno, 1970.

56.  Pfeffer, J. T., "Reclamation  of Energy from Organic Refuse," EPA
     Solid Waste Program  Grant  EP  00364,  Progress  Report, September 1971,
     p. 31.

57.  Wise,  D. L., "Fuel Gas  Production from  Solid Waste," Proc., NSF Bio-
     conversion  Energy Research Conference,  University of Massachusetts,
     Amherst, Massachusetts, June  1973.

58.  Patterson,  W.  L., R.  F. Banker,  "Estimating Costs and Manpower
     Requirements  for  Conventional Wastewater Treatment Facilities,"
     Report  by  Black § Veatch to  Office  of Research and Monitoring, EPA,
     Project #17090DAN, October 1971.

59.  Meier,  P.,  "Management Aspects of Refuse  Bioconversion,"  Proc.,  NSF
     Bioconversion Energy  Research Conference, University of Massachusetts,
     Amherst, Massachusetts, June  1973, pp.  79-87.

60.  Based  on 7.1 miles/capita-day and adjusted  for 1970 population of
     163,905; see Wilbur Smith and Associates, "Land Use and Transporta-
     tion Inventories, Volume I,"  May 1969, p. 100.


                                   89

-------
61.  It is estimated that over 100 truck-hours each day may be spent
     idling.  Applying an emissions factor of 6 between idling and
     cruising and an average speed of travel in a metropolitan area of
     15 miles/hour, this idling time may produce emissions equivalent to
     an additional 10,000 miles of travel.

62.  See Meier, Kuhner and Martel, op. cit., Chapter VI  for further dis-
     cussion on this point.

63.  New York Times, October 17, 1973.

64.  U.S. EPA, "Noise Pollution," August 1972, p. 1.

65.  See e.g., "Noise Facts Digest," report by Informatics, Inc. to
     Office of Noise Abatement and Control, EPA, Washington, June 1972,
     p. 3.

66.  The conversations occurred in June-July 1973.

67.  Fisk and Guzdar, pp. cit.

68.  Ibid., p. 94.

69.  The Administrator U.S. EPA, "Report to the President and Congress on
     Noise," February 1972, pp. 3-55.

70.  Ibid.
71.  Davidson, George R., Jr., "Residential Solid Waste Generated in Low
     Income Areas," U.S. EPA, 1972.

72.  For further discussion on this  point see, Nebiker, J. H. and P. M.
     Meier, op. cit., pp. 133-138.

73.  For an explanation of the hazards involved see, "Water Supply and
     Plumbing Cross Connections," U,STPVH,S, Publication 957, 3rd printing,
     1969.

74.  See also Meier, Kuhner and Martel, op. cit., Chapter II.

75.  See e.g., Salvato, J., "Environmental Engineering and Sanitation,"
     Wiley Interscience, New York, N.Y., 2nd Edition, 1972.


76.  U.S. Department of Health, Education and Welfare, Public Health
     Service, "Control of Domestic Rats and Mice," 1970, p. 16.

77.  Davis, P. L., "Solid Waste Management in Buildings," Building Re-
     search Institute, November 15-16, 1972.

78.  Beck, A. M., "The Ecology of Stray Dogs; A study of free-ranging
     urban animals," York Press, Baltimore, 1973.
                                   90

-------
79.  Camp, Dresser and McKee,  Inc.,  "Report  on Overflows  for the  City  of
     Springfield, Mass,," October  1971.

80.  See above, Section  IV.

81.  The monthly issues  of Solid Waste Management,  a national trade publi-
     cation, are a good  source of  information on the used equipment market,

82.  Monthly data are published in U.S.  Department  of Labor,  Manpower
     Administration, Area Trends in  Employment and  Unemployment,  about
     two months after the month to which the data apply.

83.  Federal Register, Vol.  38, No.  161, August 21, 1973, p.  22527.

84.  Ziskind, David, One Thousand  Strikes of Government Employees,
     Columbia U.P.,  1940, pp.  94-95.

85.  For extensive data  on strikes in public employment,  see Harry H.
     Wellington and  Ralph K. Winter,  Jr., The Unions and  the Cities,
     Brookings, Washington,  1971,  Appendix B; U.S.  Bureau of Labor Statis-
     tics, Work Stoppages in Government, 1958-68, Report  348,  1970; see
     also the Bureau of  Labor  Statistics annual bulletin, Analysis of
     Work Stoppages, the latest of which is  for work stoppages in 1971
      (Bulletin 1777, published in  1973).  Work stoppages  by  sewerage
     workers are not reported  separately by  the BLS, but  are combined
     with other occupations  in a wider category.

86.  Wellington and  Winter,  op. cit., p. 61; David  T.  Stanley,  Managing
     Local Government Under  Union  Pressure,  Brookings, Washington, 1972,
     pp. 68, 79.  Stanley notes a  case where an unusual increase  for sani-
     tation men resulted in  a  larger-than-planned increase for police and
     firemen as well.

87.  Stanley, op. cit.,  pp.  3, 68; also  W. Donald Heisel, "Anatomy of a
     Strike," Public Personnel Review, Vol.  XXX, No. 4, October 1969, pp.
     226-232, where  details  are given of a strike of public  employees
      (refuse collectors  and  others)  in which racial feelings were a
     factor.

88.  Stanley, op; cit.,  cites  many examples, including some  in refuse col-
      lection.

89.  Wellington and  Winter,  op. cit., pp. 62-64.

90.   Ibid., pp. 195-196.
                                    91

-------
                               SECTION XIII

                               PUBLICATIONS


 The following papers are in preparation for publication in technical
 journals.   Some further papers are in the planning stages.  At least one
 of these will be submitted to German and Swiss journals.

, Bolton, R., "An Econometric Analysis of Household Refuse  Generation," in
      preparation for submission to J.  of Environmental Economics and
      Management.

 Meier, P.  and J.  Kuhner, "Simulation Analysis of Wet Systems for Residen-
      tial Refuse Collection," to be submitted for publication in the
      J. Environmental Engineering Division,  ASCE.

 Meier, P.,  p. Kos and J. M.  Joyce,  "The Treatment  of Ground Refuse at
      Modified Sewage Treatment Facilities,"  to be submitted for publica-
      tion in the J.  Water Pollution Control  Federation.

 The following papers were presented or will  be presented  at technical
 conferences:

 Meier, P.,  "Management Aspects of Refuse Bioconversion,"  presented to the
      National Science Foundation Bioconversion Energy Research Conference,
      University of Massachusetts, Amherst, Massachusetts, June 1973.
      (Published in the conference proceedings, p.  79-87;  available from
      The Institute for Man and His Environment, University of Massachusetts,
      Amherst, Massachusetts  01002.)

 Meier, P.,  "Wet Systems for Handling Solid Waste," to be  presented at the
      annual meeting of A.I.Ch.E., Pittsburg, June 4, 1974.

 The following technical reports were prepared in the course of the re-
 search project and are available from the National Technical Information
 Service, Springfield, Virginia  22151.

 Meier, P.,  J. Kuhner and C.  J. Martel,  "A Preliminary Assessment of Wet
      Systems for Residential Refuse Collection."

 Kos, P., J. M. Joyce and P.  Meier,  "Economic Analysis of  Refuse Sludge
      Processing and Disposal."

 Meier, P.,  and G. Fisette, "Modifications to the Executive Computer
      Program for Steady-State Simulation of  Wastewater Treatment
      Facilities."

 Meier, P.,  J. Kuhner and R.  E. Bolton,  "Wet  Systems for Residential
      Refuse Collection; A Case Study for Springfield, Mass."
                                    92

-------
                       fflease read Instructions on the reverse before completing)
 REPORT NO.

   EPA-670/2-74-068
                 3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE

AN ASSESSMENT OF WET SYSTEMS  FOR RESIDENTIAL
REFUSE  COLLECTION:  Summary Report
                 5. REPORT DATE
                 August  1974;  Issuing Date
                 6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
                                                 8. PERFORMING ORGANIZATION REPORT NO
P. M.' Meier,  J. Kuhner,  and  R.  E.  Bolton
9. PERFORMING ORGANIZATION NAME AND ADDRESS

Curran  Associates, Inc.
182  Main  Street
Northampton,  Massachusetts   01060
                 10. PROGRAM ELEMENT NO.
                 lDB064/ROAP-09ADA/Task 04
                 11. CONTRACTXQfOMflXNO.
                 68-03-0183
12. SPONSORING AGENCY NAME AND ADDRESS
National  Environmental Research Center
Office  of Research and Development
U.S.  Environmental Protection Agency
Cincinnati,  Ohio  45268
                 13. TYPE OF REPORT AND PERIOD COVERED
                 14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES  See also  E P A - 6 7 0 / 2 - 7 4 - 0 3 7 (PB-234 498/AS) ;  EPA-670/2-
74-038  CPB-234 499/AS);  EPA-670/2-74-065 (PB-234 496/AS);  and EPA-670/2-
74-066  CPB-254 497-/AS)	
16. ABSTRACT
The  most  promising wet  system alternative is identified  as  a  system
using  individual household  grinders in low density areas, vacuum collec-
tion and  neighborhood grinders  in high density areas,  dilute  slurry
transport of ground refuse  in the existing sanitary sewer system,  and
joint  treatment of refuse and sewage at an expanded treatment facility
that includes anaerobic  digestion for methane generation.   However, the
economic  feasibility of  even  the most promising alternative is  doubtful
because of the high cost  of grinding, and hydraulic transport of ground
refuse in existing sewer  systems may be feasible only'if both metals and
  flass  are excluded.  A  conventional collection of the  nongrindable and
  ulky  constituents of residential refuse will still be needed,  albeit of
a diminished frequency.   The  total 1973 cost of such a wet  system  is
estimated at $113-$196  per  household per year, of which  $80-$105 is for
grinding.  This compares  to a total cost for refuse and  sewage  collec-
tion and  disposal under  existing concepts of up to $115  per household
per  year.  Projections  for  1988 indicate that the wet  system  cost  might
rise to $130-$228, against  a  range of $61-$143 under conventional  con-
cepts.  However, if the  cost  of the grinder is excluded  (on the basis
that individuals would  purchase grinders for their convenience  benefits),
the  1973  wet system costs  (to the public sector) would be only  $33-$91,
thus possibly less than  existing public sector costs of  about $43-$115
per  household per year.	
17.
                           KEY WORDS AND DOCUMENT ANALYSIS
               DESCRIPTORS
                                      b.IDENTIFIERS/OPEN ENDED TERMS
                            c.  COSATI Field/Group
 *Refuse, Collection, Sanitary
 sewers, Garbage  disposal,
 Economic analysis,  *Pipeline
 transportation,  Mathematical
 models, Sewage  treatment,
 *Grinders
*Solid waste collection,
Ground refuse transport,
*Home refuse grinders,
Economic feasibility  study,
*Hydraulic transport,
*Combined refuse/sewage
treatment
     13B
'18. DISTRIBUTION STATEMENT
            RELEASE TO  PUBLIC
       19. SECURITY CLASS (This Report)
          UNCLASSIFIED
21. NO. OF PAGES
     103
      20. SECURITY CLASS (This page)

          UNCLASSIFIED
                                                             22. PRICE
EPA Form 2220-1 (9-73)
                                     93
                                           U.S. GOVERNMENT PRINTING OFFICE: I974-657-58II/5305 Region No. 5-11

-------