EPA-670/2-74-097
December 1974
CHARACTERIZATION OF
VESSEL WASTES
IN
DULUTH-SUPERIOR HARBOR
By
Garth D. Gumtz
David M. Jordan
Robert Waller
Environmental Quality Systems, Inc.
Rockville, Maryland 20852
Grant No. R-802772
Program Element No. 1BB038
Project Officer
William J. Librizzi, Jr.
Industrial Waste Treatment Research Laboratory
Edison, New Jersey 08817
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 environmentair, 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 serves to quantify some of the pollution problems
associated with vessels in the Great Lakes including oily bilge water,
non-oily ballast water, and garbage/refuse. The data contained herein
provides an initial base for pollution control and abatement efforts.
A. W. Breidenbach, Ph.D.
Director
National Environmental
Research Center, Cincinnati
111
-------
ABSTRACT
Five wastes from United States, Canadian and foreign commercial vessels
were studied at the Duluth-Superior Harbor during late 1973: bilge
water, non-oily ballast water, sewage, garbage/refuse and dunnage.
Vessels generate bilge water at about 6,650 liters/hour with an average
oil content of about 225 milligrams/liter. Waste oil which is apparently
discharged to bilges (about 600 grams/hour) appears more consistent than
either of these two parameters. Bilge water is a substantial pollution
problem: on the average about 40 liters (10 gallons) of oil may be
discharged during each day a vessel spends in the harbor.
Although containing about twice the common water quality contaminants
as the harbor waters, ballast water is not a significant environmental
problem. Large quantities are, however, discharged: about 9,000 metric
tons/visit by lake and bulk carriers.
Sewage is apparently generated onboard vessels consistent with accepted
design rates (100 gallons/man/day). Although largely under control,
sewage may impact wastewater treatment facilities by being significantly
stronger than typical municipal wastewaters. Chemical additives
commonly found in vessel sewage may adversely affect wastewater treatment
facilities and harbor receiving waters.
Foreign vessel garbage/refuse will be more difficult to incinerate than
municipal solid wastes. Water content is higher (about 40 weight per-
cent) and the total wet heat value lower (about 2,400 calories/gram).
The average garbage/refuse generation rate was found to be 3.6 kilograms/
man-day.
Foreign vessel dunnage was not available in large enough quantities during
the field study to permit detailed characterization. Observations
indicated, however, that lumber sizes and metal content vary considerably.
In general, vessels generate wastes with extreme variability. Orders of
magnitude differences in both quantity and quality are common.
This report was submitted in fulfillment of Grant No. R-802772 by
Environmental Quality Systems, Inc., Rockville, Maryland, for the Seaway
Port Authority of Duluth, Duluth, Minnesota, under the sponsorship of
the Environmental Protection Agency. Work was completed as of December
1973.
iv
-------
CONTENTS
Page
Foreword ill-
Abstract iv
List of Tables vi
Acknowledgements vii
Sections
I Conclusions 1
II Recommendations 3
III Introduction 4
IV Project Description 6
V Sampling and Analytical Techniques 8
VI Survey and Analytical Results 14
VII Analysis of Results 40
VIII References 48
IX Appendix 49
v
-------
TABLES
No. Page
1 Basic Vessel and Rate Data for Bilge Water 15
2 Analytical Results for Bilge Water 22
3 Trace Elements in Bilge and Sewage Samples by
Emission Spectroscopy 30
4 Basic Vessel and Discharge Data for Ballast Water 31
5 Analytical Results for Ballast Water 33
6 Basic Vessel and Rate Data for Sewage 34
7 Analytical Results for Sewage 35
8 Shipboard Garbage Generation 36
9 Garbage Sample Analyses 38
10 Dunnage Size Distributuions 39
11 Apparent Bilge Water Generation Rates 41
12 Statistical Summary of Bilge Water Data for
Oil and C.O.D. 43
VI
-------
ACKNOWLEDGEMENTS
The authors are indebted to a very large number of people. First, the
Seaway Port Authority of Duluth (Mr. Thomas Burke, Executive Director,
Mr. Arvid Morkin and the rest of the Port Authority staff) provided not
only a pleasant day-to-day working environment, but also invaluable
liaisons, guidance and assistance during the course of the project.
The shipping agents of Duluth-Superior were of great assistance during
the course of the project. Mr. Richard Bibby, agent for the Hanna
Mining Co., assisted in the scheduling of waste sampling from the
inception of the project. Similar assistance was rendered by the firms
of Guthrie-Hubner, Inc., S. A. McLennan Company, Pickands Mather & Co.,
National Steel, Cleveland Cliffs, Inland Steel, Ford and many others.
The Western Lake Superior Sanitary District (WLSSD) provided continuing
assistance throughout the course of the project. Mr. Gary Baker,
Supervisor of WLSSD1s Duluth Sewage Treatment Plant Laboratory, did a
commendable job in providing analytical services under sometimes trying
conditions. Dr. W. Freiberger, Director, and his staff, of the Michigan
Technological Institute, School of Mining, Research Center at Houghton,
Michigan, comminuted garbage and prepared samples under very tight time
schedules.
Mr. Richard Amatuzio, owner of Bayfront Marine Service, and his employees
provided crucial and extensive assistance to the project, in terms of
both their time and use of their disposal facility. The characterization
of garbage and dunnage which was finally achieved can be credited in
large part to Bayfront Marine Service.
In general, the project staff wishes to thank all the people and organi-
zations of Duluth, Minnesota and Superior, Wisconsin, who assisted in
completion of the study. Their cooperation was outstanding and a credit
to the Duluth-Superior community.
Finally, appreciation must be extended to Mr. William Librizzi, the Project
Officer from the U. S. Environmental Protection Agency, for assisting in
every possible way during the study. He and the analytical staff at the
Industrial Waste Treatment Research Laboratory expeditiously provided oil
analyses of the wastewater samples gathered, often in large numbers.
VI1
-------
SECTION I
CONCLUSIONS
The following conclusions are apparent based on the Seaway Port Authority
of Duluth/Environmental Quality Systems, Inc. study of commercial vessel
waste generation at Duluth-Superior Harbor:
(1) In general, high vessel-to-vessel variability was a characteristic
of all the wastes studied. Variations of 2 or 3 orders of magnitude
are not uncommon for both quantity and quality. Such variability
is due basically to variations in operation and design from ship-to-
ship. The average vessel waste parameters developed from the field
data cannot be reliably applied to a single or small numbers of
vessels. For example, bilge water generation rates and oil concen-
trations varied, as determined in the study, from 5 to 26,000 liters
per hour and 0.4 mg/L to 84 percent by volume respectively. This,
however, does not reduce the value of the averaged data used in
assessing overall impact or in comparing classes of vessels as is
done below.
(2) Bilge water is the most significant (environmentally) vessel waste
being generated at the Harbor. According to the survey, the average
vessel generates 6,650 liters or 1,760 gallons of bilge water per
hour. This wastewater contains, on the average, about 225 milligrams
of oil per liter and 1,200 milligrams per liter of COD.
(3) Recirculation (recycle) sampling of bilge water using a portable
pump is preferred over submersion (grab) sampling; however, when
possible, samples should be taken directly from bilge pump effluent.
(4) Diesel powered vessels generate bilge water with about twice the oil
content as steamers (295 versus 140 milligrams per liter).
(5) Vessels over 20 years in age generate bilge water at about twice
the rate as younger vessels (10,000 versus 5,000 liters per hour).
(6) Canadian, United States and foreign flag vessels were found to have
increasingly poor quality bilge water, in that order; foreign flag,
Canadian and United States vessels were found to generate increasing
amounts of bilge water, again in that order.
(7) Waste oil which is apparently discharged to bilges (on the average
660 grams per hour) appears to be a less variable parameter than
either genration rates or oil content.
(8) Although vessel operators attempt to minimize their bilge water
-------
discharge while in the Duluth-Superior Harbor, these large quantities
of highly contaminated water must eventually be discharged, usually
into the Great Lakes.
(9) The average ore and bulk carrier discharges about 8,800 metric tons
(2,300,000 gallons) of non-oily ballast water per visit to the
Duluth-Superior Harbor; this quantity is 75 percent of total
ballasting capacity.
(10) For the common water quality parameters, non-oily ballast water
apparently has anywhere from 1 to 10 times the concentrations as the
natural harbor waters; this is highly dependent on the particular
vessel being considered.
(11) The quality of ballast waters appears to degrade only slightly in
transit to Duluth-Superior.
(12) Compared to other pollution control and environmental problems, non-
oily ballast water does not appear to be a major concern at the
Duluth-Superior Harbor.
(13) Sewage is apparently generated onboard vessels visiting Duluth-
Superior in line with accepted literature and design values:
380 liters or 100 gallons per man per day for combined wastes (including
galley, laundry, shower, lavatory and sanitary wastewaters) and 95
liters or 25 gallons per man per day for sanitary wastes.
(14) Vessel sewage is highly concentrated: average BOD, suspended, and
settleable solids of 750, 1,800, and 1,600 milligrams per liter
respectively.
(15) Although sewage appears to be reasonably well controlled onboard vessels,
its strength, chemical and oil content could produce severe problems
for small shoreside wastewater treatment facilities which accepted
and attempted to treat same.
(16) Garbage/refuse is generated onboard foreign vessels at fairly high
rates: 3.6 kilograms or 8 pounds per man per day; no dependence was
found on vessel nationality.
(17) Vessel garbage/refuse contains large quantities of food (62 percent)
and glass/ceramic (12%) wastes and relatively small amounts (8.5%)
of paper.
(18) Incineration (or other disinfection) of foreign garbage/refuse may
prove difficult due to high water content (about 40 percent) and low
total heat value (2,400 calories per gram).
(19) Due to the scarcity of dunnage in the harbor during the study period,
little can be concluded on this waste other than that it contains a
large size range of lumber and varying types of metal in relatively
small amounts.
-------
SECTION II
RECOMMENDATIONS
Based on the conclusions and the more detailed analyses in the main body
of the report, the following are recommended:
(1) Vessel bilge water wastes should be controlled more stringently in
the Duluth-Superior Harbor, and in the Great Lakes as a whole. Most
bilge water being discharged is more contaminated than presently
allowed by law and regulations. That which is not is due mainly to
dilution (an unacceptable control technique in most instances).
Monitoring by appropriate governmmental agencies plus advanced bilge
water treatment will probably be necessary for this control.
(2) Until proven a more significant contributor to water pollution problems,
non-oily ballast (at the Duluth-Superior Harbor) should be the subject
of only limited control and regulation. Vessel owners and operators
should be made aware, however, that "non-oily" ballast can be a
noxious waste if not handled properly.
(3) Although vessel owners and operators appear to be controlling sewage
onboard their vessels, municipal wastewater treatment facilities must
be wary of casually accepting such wastes. They are highly contam-
inated and often contain chemical additives. More definitive work
should be done on characterizing vessel sewage and its possible impact
on wastewater treatment plants.
(4) Facilities for foreign vessel garbage/refuse disposal or disinfection
should be designed to handle the lower total heat values and higher
water contents determined in this study.
(5) Foreign vessel dunnage should be more completely characterized before
design and demonstration of advanced disposal and disinfection
facilities is initiated.
(6) Systems for the control or treatment of wastes from a single or small
number of vessels should be designed using data for that particular
vessel(s). High variability in characteristics must be considered
an (or, possibly, the) important property of vessel wastes.
-------
SECTION III
INTRODUCTION
Vessel pollution in harbors is a much discussed but not very well defined
environmental problem.
Bilge waters from both fresh water and ocean going vessels are known to
be oily, and their discharge to receiving waters known to be polluting.
However, the magnitude and variability of the problem, effects of
propulsion system, dependence on vessel age and nationality, etc., are
not well known, if at all. Furthermore, means to estimate the quantity
and quality of bilge water are not readily available. Such means must be
forthcoming if the Nation's fleet, particularly in the Great Lakes, is
to be outfitted with effective bilge water treatment devices.
A similar state of affairs exists for ballast water discharges. Although
several studies have been conducted to characterize oily ballast water
from tankers, no information is available on the nature of ballast water
discharges from other than oil tankers. "Non-oily ballast" may also be
contaminated and present a threat to the environment. Although contamina-
tion with hydrocarbons is bound to be less, other pollutants (suspended
solids, biological oxygen demand, coliform bacteria, etc.) may be present
in high concentrations, especially in cases where ballasting may occur
in polluted waters (e.g., certain ports on the Great Lakes). Due to the
large quantities of water involved, the pollution potential of non-oily
ballast water should be assessed.
Although quantity and quality data are available from the literature on
vessel sewage, little of this relates directly to the high variability
of conditions actually encountered onboard vessels. Existing sources
are primarily data from marine sanitary system manufacturers and studies
of various treatment concepts under controlled conditions. Little
information can be found which addresses the problems of overloaded
systems, chemical additives, high waste concentrations, wastes from recycle
systems, holding tanks which have gone septic, etc. Such information
could be very valuable in determining the suitability of shoreside waste-
water treatment for existing vessels.
Garbage/refuse from vessels is often assumed to be similar to that gen-
erated in municipalities. Data collected by the U. S. Navy, however,
indicates that this assumption may not be valid. Addition of household
and industrial garbage/refuse generation rates does not exactly model
conditions onboard vessels: (1) dietary habits are perturbed by ship-
board environments, (2) disposable materials are often used due to
limitations on laundry, dishwashing and other cleaning services, (3) ex-
tensive use of readily available water can lead to very wet solid wastes,
(4) solid wastes from foreign vessels may be infested and (5) used materials,
4
-------
which would not be discarded onshore, are often added to the solid waste
stream. These differences make more careful characterization of commercial
vessel garbage/refuse generation important, especially for-larger ports.
Dunnage (waste "wood" previously used for supporting, protecting and
holding cargo) can be an especially large problem at ports. Disposal of
this material is in itself troublesome. Foreign dunnage may be infested
with agricultural pests adding to the basic disposal problem by requiring
disinfestation. Size variations of the lumber comprising dunnage plus
substantial amounts of contained metal make characterization important.
For instance, in a particular load of dunnage anywhere from 12 foot lengths
of 10 x 10's to short lengths of lathing may be found as well as upwards
of one percent by weight metal in the form of nails, staples, bolts,
strap, wire and cable.
The remaining pages of this report describe a study of the vessel wastes
mentioned above at the Duluth-Superior Harbor on Lake Superior. This
study was conducted for the Seaway Port Authority of Duluth under a grant
from the U. S. Environmental Protection Agency. In Section IV the aim
of this particular study is described briefly. Section V presents a
summary of the sampling and analytical techniques which were used in the
study. Section VI presents the data which was gathered during late 1973
at the Duluth-Superior Harbor. Section VII is a brief analysis of the
study accompanied by a brief discussion of same.
-------
SECTION IV
PROJECT DESCRIPTION
The primary thrust of the vessel waste study at the Duluth-Superior Harbor
was in collecting and analyzing waste samples to characterize bilge water,
non-oily ballast water, sewage, garbage/refuse and dunnage. Data was also
collected on the magnitude of waste generation. The latter tends to be a
very complex and expensive proposition, especially for liquid wastes.
Therefore, most generation data, as well as information on operations,
came from surveys and interviews of vessel crew members and service com-
panies. After an initial two week set-up and shakedown period under the
direction of the project supervisor, the project's field engineer and
technician gathered all vessel and operations data, took selected samples
of the various wastes, and prepared the latter for analysis; in some
cases (i.e., for garbage and dunnage) all or part of the analyses were
performed by this field team. Extensive cooperation from vessel and
service owners and operators was required for the successful completion of
these tasks.
Bilge water was the waste studied most thoroughly during the project. In-
itially, 25 vessels were to be sampled; this was increased to 35 and, in
the end, 37 were actually sampled. Every effort was made to consider a
wide spectrum of vessels: type (bulk, lake or general cargo carrier),
nationality, propulsion system, and age. Several different techniques for
sampling bilge water were used; comparison of the results for these
different techniques provides a necessary foundation for any future sampling
work as well as means for estimating sampling reliability.
Bilge (as well as ballast and sewage) water samples were collected with the
assistance of vessel agents, owners and operators. Usually, the field en-
gineer made an appointment to board a particular vessel, interview the
chief engineer and take samples. The interview provided basic information
on the vessel and data necessary for estimating waste generation rates.
When possible, the engineer's log book was used as a basic data source.
After the field engineer and technician set up for sampling onboard a vessel,
matters became quite routine and interviews could be conducted concurrently.
After sampling, about an hour was usually required to pack up equipment
and return to the Port Authority offices; here, samples were acidified
(except those used for biochemical oxygen demand and coliform determinations)
and either refrigerated or transported to the Duluth Sewage Treatment Plant
laboratory. Samples requiring total organic carbon, oil and metals
analyses were shipped by air freight to the Edison Water Quality Laboratory
in New Jersey. Easy access to the bilges onboard vessels made bilge water
sampling relatively easy. Dry bilges, lack of adequate power for the
sampling pump, and lack of access prevented sampling at times as did equip-
ment failures.
-------
When required, non-oily ballast water samples were taken using a Kemmerer
sampler; either the engineer or technician did this while the other took
bilge water samples. These samples were also preserved and routed to the
appropriate analytical facility. Since manhole access to ballast tanks
is typical, grab sampling ballast water was usually easy. Originally,
5 ballast water samples were to be taken; this number was eventually raised
to 10 and, finally, 11 (plus 2 harbor samples) were collected and analyzed.
Non-oily ballast water was not studied in great detail. The point of the
study was basically to assess the possibility that a pollution problem
exists.
Sewage samples were generally difficult to obtain. With the assistance of
ship's engineers the field engineer generally had to open holding tanks
or bleed samples from pumps. There was no routine way to do such sampling
as is indicated in the report section which follows. When required,
sewage samples were taken concurrently with bilge water samples. Six
samples were required and eventually taken.
Both foreign vessel garbage/refuse and dunnage were studied relatively
independently of actual vessel operations. This was possible only through
the cooperation of the disposal company at the harbor, Bayfront Marine
Services. The project required weighing approximately 25 percent of the
foreign garbage offloaded at the harbor; this was done primarily by dis-
posal company employees using a truck scale. Garbage analyses were per-
formed by the field crew (hand sorting into basic categories) with assis-
tance from others. Combustible wastes had to be transported to Michigan
for grinding and mixing; this required a special permit from the U. S.
Department of Agriculture as well as careful handling and packaging due
to possible contamination. Samples prepared in Michigan were delivered
to a laboratory in Minneapolis, Minnesota for water content and total heat
value determinations. The field engineer was in charge of this operation.
Four samples from 2 vessels were subjected to such detailed preparation
and analysis. In a similar fashion, dunnage (when available) was weighed
using a truck scale; size analyses were determined by measuring and
counting, and metal content by separating and weighing. Although 2 loads
of dunnage were scheduled for thorough analysis, the second load never
materialized.
-------
SECTION V
SAMPLING AND ANALYTICAL TECHNIQUES
Although the samples of vessel wastes gathered at Duluth-Superior Harbor
could be analyzed using fairly standard techniques, the sampling itself
was by no means "standard". That is, there were no widely accepted
procedures for sampling the vessel wastes of interest. These should
eventually be developed if widespread agreement upon and resolution of the
vessel waste problem is to be achieved.
The following paragraphs describe the sampling and analytical procedures
used for the five waste categories. No attempt is made to present stan-
dard techniques in detail.
BILGE WATER SAMPLING
Bilge water was the most difficult waste for which to collect samples.
Since "small" volumes of water spread over a relatively large area are
usually involved, some care must be taken to simulate bilge pumping during
sampling; dipping (grabbing) samples is probably not acceptable. In
general, obtaining representative samples must be a primary aim. Analysis
of bilge water samples, on the other hand, can, to a large extent, be
according to "standard methods'1 or procedures acceptable to EPA.
Several techniques were used for sampling bilge water from vessels. The
primary method was to recirculate the water located in the bilge area of
the engine room. A portable 40 gpm pump (Montgomery Ward, Model SEN 25415)
was used for recirculation. The suction side of the pump was placed near
the bilge water intake for a ship's bilge pump. This intake is almost
always located in the aft section of the engine room below the propeller
shaft. The water withdrawn at the intake was discharged in the fore
section of the engine room. Discharge was to a dry area or to relatively
stagnant water while submerged. This technique was used to simulate the
ship's bilge pump which causes circulation mixing of floatables and
settleables as well as actual discharge of such materials. In the pumping
process, a portion of floating oil and suspended material is usually with-
drawn with the water. The recirculation method can be used since the
engine room bilge is a common area; that is, it is usually not divided by
bulkheads. Water can, therefore, flow freely from fore to aft in the
engine room.
Note that the recirculation technique can be used only under specific
conditions. The following can prevent using this method:
(1) Shallow bilge water (10 or so centimeters^ pump surging.
(2) A "dry" bilge.
-------
(3) An inaccessible bilge water sump.
(4) Absence of 110 volt, 60 cycle, AC power in the engine room.
(5) Too great a vertical lift for the portable pump used.
(6) Impossibility of obtaining a representative sample (e.g. due to
extreme bulkheading of the engine compartment), and
(7) The availability of an alternative sampling technique which will
provide a more representative sample.
The recirculation method was tested by taking samples over a period of
pumping time to ascertain the effects of recirculation. Samples could
usually be confidently taken after 15 or 30 minutes of recirculation.
The second bilge water sampling technique was submersion of a sample bottle
into the bilge water; this is a grab sampling procedure. A stoppered
bottle was submerged, the stopper was removed and replaced once the bottle
was filled while remaining submerged. Sampling depth varied from several
centimeters to about a meter depending upon the amount of water in a
bilge and the apparent thickness of the layer of oil floating on the water.
The latter was, in general, difficult to circumvent other than by taking
repetitive samples.
The submersion technique has limitations similar to those above for re-
circulation. Limiting conditions are as follows:
(1) Bilge water is not deep enough for submersion,
(2) Inaccessible bilge water sump (no water elsewhere),
(3) Too deep an oil/water layer to allow a representative sample, and
(4) Another sampling technique would be more representative.
A combination of sampling techniques was sometimes used on a given vessel.
Using more than one provides a basis for comparison of the various methods.
Furthermore, the "best" can possibly be determined by using and comparing
different methods of sampling.
The third sampling technique was sampling directly from the ships bilge
pump. Many bilge pumps were equipped with a valve or a pressure gauge
port where a sample could be taken. The samples were taken either directly
from the valve or port or by shunting discharge water to the bilge area.
Removal of a pipe coupling, pressure gauge line, or other breaking of a
discharge line were also used. Variations in sampling from the pump are
dependent upon the type of pump(s), accessibility, and the general con-
figuration of the bilge area. This is, obviously, the preferred technique
for obtaining bilge water samples.
-------
BALLAST WATER SAMPLING
Ballast water can also be analyzed using mainly standard methods. Since
ballast water is deep and a relatively stagnant volume, sampling can be
done using state-of-the-art procedures (e.g., bottle type samplers).
Note, however, that this is for non-oily ballast; oily ballast must be
sampled much more carefully since oil/water stratification occurs
readily in ballast tanks.
Ballast samples were taken using a 3.8 liter (one gallon) Kemmerer sampler,
Ballast tanks were accessible through manholes on either the weather deck
or the tunnels adjacent to the cargo holds, depending on the type vessel
and its general configuration. Composite samples were made from samples
taken at various depths in the tank. Given sufficient water in a tank,
the depths sampled were about one foot from the bottom, at the middle of
the tank, and about one foot from the surface. Ballast tanks were
typically 26 feet deep; therefore, a typical composite sample consisted
of equivolumetric samples from levels of 1, 13 and 25 feet.
SEWAGE SAMPLING
Sewage is also a very difficult waste to sample under field conditions.
Sample ports were not generally observed on either sewage holding tanks or
treatment systems. Without cutting into piping, samples must generally
be taken through manholes, drain pipes, or petcocks on wastewater pumps.
Again, sample analyses require only standard wastewater procedures.
Sewage samples were taken from holding tanks on ocean-going vessels and
U. S. and foreign lakers. No single technique could be established due
to ship-to-ship variations in sewage handling systems and holding tanks.
If possible, a grab sample was taken after internal recirculation of
sewage in a holding tank. However, accessibility often necessitated
alternative techniques. For instance, samples were taken from a valve
on a discharge line or directly off a wastewater pump. Several systems
observed were completely inaccessible while others would not allow
sampling since the wastewater would have had to been illegally discharged.
GARBAGE SAMPLING
Sampling garbage was also a difficult problem. Since only two containers
of garbage were thoroughly analyzed, the results cannot be applied gen-
erally. Selection of garbage for weighing was done by the field crew
on essentially a random basis with assistance of the disposal service
operators. Analyses were basically standard with the one exception being
in sample preparation: completely ideal garbage comminution equipment
was not available and a make-shift arrangement was, therefore, used.
The garbage was weighed from about one of every 4 foreign flag vessels.
The disposal service personnel in charge of garbage collection did this
weighing when possible. Due to dock inaccessibility, some vessel garbage
is transported via boat, unloaded at the garbage facility and incinerated;
10
-------
this garbage could not be weighed. Also, some loads were not very large
so garbage from more than one ship was loaded onto a single truck. Other
constraints such as time, availability, high levels of activity, weather
and operating personnel prevented maintaining a consistent schedule for
weighing garbage. However, approximately 25% of all loads collected during
the 2.5 month field program by the local contractor were weighed.
The two containers of garbage that were sorted, comminuted and analyzed
were selected (in consultation with one of the disposal operators who
has had several years experience in handling garbage from foreign flag
vessels) as being "typical". Two 55 gallon containers were sorted at
the site of the garbage incinerator, separated into gross solid waste
categories, bagged, weighed, comminuted to a homogeneous mixture and
analyzed at a local laboratory.
DUNNAGE SAMPLING
A typical batch of dunnage consisted of more than one truckload. An
attempt was made to weigh every truckload; however, due to the time and
personnel involved (longshoremen, ship's crew and several employees of
the local contractor), this was not possible. However, 50% of the truck-
loads were weighed. Size and metal content analyses were made for two
truckloads of dunnage. Due to sample limitations (i.e., dunnage from
only one vessel during the 2 month study) weights and analytical results
cannot be generalized.
WASTEWATER (BILGE, BALLAST, AND SEWAGE) ANALYSES
Analyses for wastewater samples (bilge, ballast and sewage) were, in large
measure, carried out according to "Standard Methods for the Examination
of Water and Wastewater, 13th Edition". Most analyses were performed
at the Duluth Sewage Treatment Plant Laboratory, which is now owned and
operated by the Western Lake Superior Sanitary District (WLSSD). The
tests performed were as follows:
1) Biochemical Oxygen Demand, 5-day (BOD5)
2) Chemical Oxygen Demand (COD)
3) Suspended Solids
4) Volatile Suspended Solids
5) Settleable Solids
6) Total Phosphorus
7) pH
8) Total Coliform
11
-------
9) Oil
10) Total Organic Carbon
11) Atomic Absorption for Metals
Brief discussions of each of these analyses are presented in the Appendix
to this report.
GARBAGE ANALYSES
Samples of garbage consisted of two 55-gallon drums, each weighing 220 to
300 pounds. It is at best impractical to separate, measure, and weigh all
the garbage that is unloaded from a single ship, let alone an entire
shipping season's garbage. In a recent study ("Analysis of Solid Waste
Composition" by Carruth and Klee of the Public Health Service) 200 pound
samples were shown to yield statistically valid results for the composition
of garbage. A major possibility is that samples from more than two loads
of garbage should be analyzed, however, to characterize garbage generation
for the entire harbor. This same study listed nine standard garbage/
refuse component classifications; these are:
1) Food waste
2) Paper products
3) Textiles
4) Wood
5) Plastic, rubber, and leather
6) Metals
7) Glass and ceramics
8) Ash rocks and dirt (inerts)
9) Garden waste
Note that garbage/refuse from foreign vessels is subject to the quarantine
restrictions of the U. S. Department of Agriculture. That is, transport of
such material must be under very controlled conditions. This is adequately
illustrated in the following paragraphs.
Upon receipt, the contents of a sample drum were dumped onto a 20 foot
plastic tarp. The waste was then hand sorted into the nine categories (if
applicable). Once the segregation of the waste was completed, waste from
each category was placed in a separate plastic garbage bag which was
appropriately labeled. This procedure was then repeated for the second
sample drum. The bags of waste and containers in which they were placed
12
-------
were then sealed by a representative of the U. S. Department of Agriculture,
On the following day, the sealed samples were transported to the Michigan
Technological Institute Mining Research Center at Houghton, Michigan.
Each component bag was weighed and the necessary seal broken by a represen-
tative of the Michigan Department of Agriculture. The combustible portion
of the waste (i.e., food waste, paper, textiles, plastic, and wood) was
then ground in a hammermill with a 1/4 inch screen. However, large
quantities of textiles and plastics could not be comminuted into a homo-
genous sample; they were, therefore, left out of the sample and later
accounted for by estimating their effects. Therefore, the homogenous
sample that was eventually analyzed consisted primarily of food waste,
paper products, and wood, with only minor amounts of textiles and plastics.
The overall sample was mixed and split (cut) by hand and two one quart
samples were withdrawn for each 55 gallon drum of garbage/refuse processed.
The ground garbage was returned to the bags and containers, resealed by
the Michigan Department of Agriculture, and returned to the disposal
facility at Duluth.
The one quart garbage samples were frozen at Duluth until they could be
delivered to the Twin Cities Testing Laboratories, Inc. at Minneapolis,
Minnesota. This laboratory determined moisture contents and heat values
for a sample from each of the 55 gallon drums. These analyses were
conducted substantially in accord with the standard procedures as pre-
sented in "Methods of Solid Waste Testing" by the U. S. Environmental
Protection Agency, 1973. Such test results can be used to estimate
ultimate disposal requirements; for instance, the amount of water which
must be evaporated and heat of combustion supplied by the garbage itself
can be estimated.
DUNNAGE ANALYSES
Dunnage was weighed immediately using a truck scale and isolated at the
disposal site for further analysis prior to burning. Dunnage was
measured using an 8 foot long by 12 inch wide by 12 inch deep jig. In-
dividual pieces of lumber were placed on the jig and their dimensions
recorded. Prior to measurement, all metal (e.g., straps, cables and
nails) was removed and weighed to the nearest pound. After the size
distributions had been determined, the dunnage was burned by the disposal
service operators in line with their standard procedures. Dunnage from
foreign vessels is also subject to quarantine regulations. Size analyses
and composition estimates should be useful in designing systems for in-
cineration or shredding and disinfestation of dunnage.
13
-------
SECTION VI
SURVEY AND ANALYTICAL RESULTS
Previous sections have described the study in general as well as the
sampling and analytical techniques used. This report section is in
essence a tabular presentation of the results of the vessel surveys,
sampling and analyses. Only brief descriptions which serve to explain the
various tables are provided. The data presented is largely self-
explanatory; obvious anomalies are, however, discussed briefly.
The study results are presented below in the same order used in previous
discussions: bilge water, ballast water, sewage, garbage and dunnage,
respectively.
BILGE WATER
Table 1 summarizes the vessel survey data for bilge water. This data
was solicited from vessel operators to (1) characterize particular ships
and (2) develop estimates of bilge water generation. The first row
identifies the bilge water samples taken from the particular vessel in
question. The second row lists the vessel type. General cargo vessels
and tankers were not present in large numbers during the late 1973 season
at Duluth-Superior. Lake and bulk carriers are similar although the
latter are used for both bulk and general cargo and are often ocean
going. Although only the broad categories of United States, Canadian
and foreign vessels are of much practical use, the third line in Table 1
presents a detailed identification of vessel nationality; note that the
term "nationality" refers to the predominant home origin of the crew.
Row 4 describes the general nature of the vessel's propulsion systems.
Although steam turbine and other steam driven vessels are differentiated
on the table, for purposes of bilge water characterization they should
probably be considered together. Vessel age in row 5 is self-explanatory
although it is somewhat surprising that 3 crews were unable to supply a
vessel age.
Row 6 of Table 1 lists probable bilge water sources; these were determined
through interviews with vessel engineers and inspection. Under the
bilge water sources is a listing of the number of bilge water pumps on-
board the vessel. The frequency with which bilges are pumped is indicated
in row 8. These frequencies, as well as the information in the following
2 rows, were obtained from ship's logs or by interviewing operating engi-
neers. At best pumping frequencies are as scheduled while at worst they
are educated guesses; no entry in row 8 indicates that a vessel's crew
was unwilling to hazard an estimate of pumping frequency. Row 9 indicates
the volume of bilge water discharged during each pumping activity; these
numbers were either supplied by the vessel engineers or calculated on the
basis of pumping duration (row 10) and bilge pump capacity. In some cases
14
-------
Table 1. BASIC VESSEL AND RATE DATA FOR BILGE WATER
Bilge Sample Number
Vessel Type
Vessel Nationality
Propulsion System
Vessel Age (yrs. )
Bilge Water Sources
Bilge Pumps, Number
Pumping Frequency
Bilge Water Pumped,
m3 (gal.)
Duration of Pumping
Apparent Generation
Rate: liters/hour
(gallons/hour)
1-3
Lake
Carrier
U.S.
St.Turb.
19
pumps ,
condensers,
gauges
2
-
-
-
4
Bulk
Carrier
Armenian
Diesel
4
engine,
pumps
3
automatic
-
-
-
5
Bulk
Carrier
German
Diesel
-
engine,
pumps,
sanitary
1
I/week
18(4800)
2hrs.
108
(29)
6-8
Bulk
Carrier
Canadian
St.Turb.
14
shaft, pumps,
cooling water,
sanitary
3
automatic
«
^^^^ ^"^^«^
9-11
Lake
Carrier
U.S.
St.Turb.
24
shaft,
bearings
1
l/40min.
0.3(80)
2min.
450
(120)
^
12
Lake
Carrier
U.S.
Steam
50
shaft,
pumps
1
4-5/4hrs.
3-4min.
13,47
Bulk
Carrier
Canadian
Diesel
5
condensate
2
l/4hrs.
-
5-10min.
^^^
-------
Table 1 (Continued). BASIC VESSEL AND RATE DATA FOR BILGE WATER
Bilge Sample Number
Vessel Type
Vessel Nationality
Propulsion System
Vessel Age (yrs. )
Bilge Water Sources
Bilge Pumps, Number
Pumping Frequency
Bilge Water Pumped,
m3 (gal.)
Duration of Pumping
Apparent Generation
Rate: liters/hour
(gallons/hour)
14
Lake
Carrier
U.S.
St.Turb.
20
pumps,
shaft seal ,
condensers
1
1/20 min.
7.6
(2000)
4-5min.
23
(6100)
15,16
Bulk
Carrier
British
Diesel
4
engine,
ballast
pumps
2
l/2days
-
-
-
17
Bulk
Carrier
Canadian
St.Turb.
23
shaft, condensors,
pumps, shaft seal ,
cooling water
2
automatic
-
-
-
18,19
Lake
Carrier
U.S.
St.Turb.
28
pumps,
shaft
seal
2
l/4hrs.
20
(5200)
lOmin.
4900
(1300)
20-26
Bulk
Carrier
British
Diesel
4
shaft seal ,
bal last pumps,
condensation
1
I/day
3.4
(900)
lOmin.
140
(38)
27
Lake
Carrier
U.S.
Tr.Expan. Steam
67
shaft seal ,
cooling water,
condensation
1
continuous
-
-
-
-------
Table 1 (Continued). BASIC VESSEL AND RATE DATA FOR BILGE WATER
Bilge Sample Number
Vessel Type
Vessel Nationality
Propulsion System
Vessel Age (yrs. )
Bilge Water Sources
Bilge Pumps, Number
Pumping Frequency
Bilge Water Pumped,
ITH (gal.)
Duration of Pumping
Apparent Generation
Rate: liters/hour
(gallons/hour)
28
Lake
Carrier
U.S.
St.Turb.
20
pumos ,
shaft
seal
1
l/30-45min.
3.0
(790)
4-5min.
4.7
(1250)
29-34
Lake
Carrier
U.S.
St.Turb.
32
shaft seal ,
cooling water,
pumps, leaks
2
continuous
-
-
16,000
(4200)
mm^mtmmmmmtmmmmm
35,36
Lake
Carrier
Yugoslavian
Diesel
5
misc.
1
continuous
18,000
(4800)
^HMBBMHBiMH
37
Bulk
Carrier
Polish
Diesel
4
misc.
1
continuous
25,000
(6700)
38
Lake
Carrier
U.S.
Tr.Expan. Steam
47
shaft seal ,
cooling
water
1
1/10-1 5m1n.
-
4-5min.
^« B««IMi^^-«
23,000
(6000)
39
Lake
Carrier
U.S.
Steam
14
shaft,
cooling
water
1
1/hr.
lOmin.
-
-------
Table 1 (Continued). BASIC VESSEL AND RATE DATA FOR BILGE WATER
H '
00
Bilge Sample Number
Vessel Type
Vessel Nationality
Propulsion System
Vessel Age (yrs. )
Bilge Water Sources
Bilge Pumps, Number
Pumping Frequency
Bilge Water Pumped,
m3 (gal.)
Duration of Pumping
Apparent Generation
Rate: liters/hour
(gallons/hour)
40
General
Cargo
Danish
Diesel
-
shaft
seal
2
-
-
-
15,000
(4000)
41
Bulk
Carrier
Greek
Diesel
2
general
leakage
2
1/mo.
6
(1600)
20 min.
8
(2.2)
42
Bulk
Carrier
U.S.
St.Turb.
62
cooling
water
1
continuous
-
_
19
(5)
43
Bulk
Carrier
British
Diesel
1
general
leakage
1
1/wk.
5
(1300)
15 min.
30
(8)
44
Bulk
Carrier
Norwegian
Diesel
4
wash
water
2
1/mo.
22
(5900)
10 min.
30
(8)
45
Bulk
Carrier
French
Diesel
2
general
leakage
1
1/mo.
17
(4400)
10 min.
23
(6)
46
Bulk
Carrier
Yugoslavian
Diesel
_
general
leakage
1
as necessary
-
.
-
-------
Table 1 (Continued). BASIC VESSEL AND RATE DATA FOR BILGE WATER
H
VD
Bilge Sample Number
Vessel Type
Vessel Nationality
Propulsion System
Vessel Age (yrs. )
Bilge Water Sources
Bilge Pumps, Number
Pumping Frequency
Bilge Water Pumped,
m3 (gal.)
Duration of Pumping
Apparent Generation
Rate: liters/hour
(gallons/hour)
48
Bulk
Carrier
Canadian
Diesel
7
shaft
seal
1
_
-
_
8
(2)
49
Tanker
Norweigan
Diesel
18
shaft seal , wash
water, condensate,
sink wastes
1
I/day
11
(3000)
30 min.
470
(125)
50
General
Cargo
Yugoslavian
Diesel
7
shaft seal ,
engine
1
1/2-3 days
3.5
(940)
-
60
(16)
51
Bulk
Cargo
German
Diesel
7
misc.
1
I/day
45
(12,000)
25 min.
1900
(500)
52
Bulk
Carrier
Canadian
Diesel
8
shaft seal ,
engines,
ballast pumps
1
I/day
20
(5250)
15-20 min.
830
(220)
i
53
Bulk
Carrier
Canadian
Triple Expansion
Steam
60
cooling water,
shaft seal
1
-
-
-
^VBMM
-------
Table 1 (Continued). BASIC VESSEL AND RATE DATA FOR BILGE WATER
tsJ
o
Bilge Sample Number
Vessel Type
Vessel Nationality
Propulsion System
Vessel Age (yrs.)
Bilge Water Sources
Bilge Pumps, Number
Pumping Frequency
Bilne Water Pumped,
tin (gal.)
Duration of Pumping
Apparent Generation
Rate: liters/hour
(gallons/hour)
54
Bulk
Carrier
Canadian
Diesel
5
shaft seal , wash
water, laundry,
drinking water
2
1/2 weeks
1.1
(300)
-
(1.8)
55
Bulk
Carrier
Canadian
St.Turb.
10
shaft seal ,
cooling water,
pumps
2
I/day
120
(32,000)
90 min.
5100
(1350)
56
Bulk
Carrier
Canadian
St.Turb.
16
shaft seal ,
cooling water,
pumps
1
1/2 hr.
3.4
(900)
2-4 min.
1700
(450)
57
Bulk
St.Turb.
12
stern
gland,
pumps
1
1/4 hr.
26
(6800)
20 min.
6400
(1700)
58
Bulk
St.Turb.
9
shaft
seal ,
pumps
1
1/2 to 1 hr.
20
(5250)
15-20 min.
26,000
(7000)
-------
engineers were willing to provide estimates of bilge water pumped, in
others only of pumping duration. The last row in Table 1 indicates the
apparent bilge water generation rate as calculated from the more basic
data in the 3 rows above; these numbers are essentially the bilge water
pumped as averaged over the inverse of the pumping frequency. For
instance, for sample 58 the data indicates that 20 cubic meters are
pumped every 0.75 hour (the mid-range value); the apparent rate is, there-
fore, 26 cubic meters per hour.
Table 2 gives the analytical results for the bilge water samples. These
sample analyses may be related to the vessel characteristics of Table 1
by referring to the bilge sample numbers. Sampling location refers
generally to the area in which samples were taken up (e.g., by a sample
bottle or the suction of the sampling pump). Most samples were obtained
from propeller shaft alleys, a common bilge area which most always con-
tains standing water and bilge pump. In two cases pump room and engine
room samples were taken for the purpose of comparison to shaft alley
samples. "General bilge" refers to samples off bilge pumps or oil/water
separators. Row 3 of Table 2 refers to sample type or sampling procedure
as described in Section V. "Grab" and "recycle" are shorthand terminology
for submersion and recirculation sampling respectively; sampling times
as measured from starting of the sampling pump are included with the
latter. The bilge pump and separator discharge samples are by far the
most representative samples possible. Rows 4 to 13 contain the analytical
results for the bilge water samples, as described in Section V. Also,
some samples were analyzed only for oil. Sample loss (e.g., breakage
of bottles) and undue analytical interference (e.g., by large quantities
of oil) also are responsible for the lack of certain data points. Note
that grab sampling predominates towards the end of the two month study
period. An attempt was made to gather more grab samples to balance out
the program; however, this attempt almost became permanent when the
sample pump broke down and took over 3 weeks to be repaired.
Table 3 presents the results of emission spectroscopy analyses for trace
elements on 8 bilge water samples and one sewage sample. This table is
self-explanatory. Vessel, generation rate and analytical data for
sewage sample number 1 will be presented later.
BALLAST WATER
Table 4 presents basic vessel and discharge data for ballast water samples.
The first 3 rows are similar to those for bilge water. Cargo type is
given in row 4 because of possible cross-contamination between cargo and
ballast water. Row 5 presents total ballast water capacity (in metric
tons) as provided by the ships' engineers from design specifications.
Trip ballast (row 6) was obtained from the engineers' logs and represents
the quantity of ballast water actually loaded at the last port of call;
the source water body for this original ballast is listed in row 7. Row 8
gives the total ballast water discharged at the Duluth-Superior Harbor.
The ballast tank sampled is identified in row 9 by size and location. Row
10 indicates the amount which the sampled tank's water was diluted enroute
to Duluth-Superior, if any. The source of the dilution ballast water is
21
-------
Table 2. ANALYTICAL RESULTS FOR BILGE WATER
Bilge Sample Number
Sampling Location
Sample Type
Biological Oxygen Demand
(5 day)(mg/l)
Chemical Oxygen Demand
(mg/1)
Suspended Solids (mg/1)
Volatile Suspended Solids
(mg/1)
Settleable Solids (mg/1)
Phosphorus (mg/1)
PH
Total Coliform (MPN/lOOml)
Total Organic Carbon (mg/1)
Oil (mg/1)
1
shaft
alley
30 mm.
recycle
64
321
33
23
.02
.51
8.25
3.6xl05
-
137
2
shaft
alley
60 min.
recycle
40
275
27
18
.02
.48
8.15
1.7xl05
-
115
3
pump
room
30 min.
recycle
4
54
20
9
.02
.14
8.12
4.8xl04
-
16.6
4
shaft
alley
30 min.
recycle
80
312
288
236
156
1.49
5.58
10
-
-
5
shaft
alley
grab
168
6464
212
181
157
1.54
7.45
2. 88x1 O5
-
2370
6
shaft
alley
sump
20 min.
recycle
14.5
128
47
31
25
.07
7.35
1.43xl05
-
-
7
shaft
alley
sump
40 min.
recycle
18.0
116
49
33
26
,06
7.45
1.89xl05
-
-
8
shaft
alley
$ump
60 min.
recycle
19.0
172
68
45
39
.08
7.45
1.02xl05
_
45.8
-------
Table 2 (Continued). ANALYTICAL RESULTS FOR BILGE WATER
LO
Bilge Sample Number
Sampling Location
Sample Type
Biological Oxygen Demand
(5 day)(mg/l)
Chemical Oxygen Demand
(mg/1)
Suspended Solids (mg/1)
Volatile Suspended Solids
(mg/1)
Settleable Solids (mg/1)
Phosphorus (mg/1)
PH
Total Col i form (MPN/ 100ml)
Total Organic Carbon (mg/1)
Oil (mg/1)
9
shaft
si ley
30 min.
recycle
18
752
196
93
118
2.48
7.10
52x1 O3
-
10
shaft
alley
60 min.
recycle
130
1216
392
387
288
3.47
7.10
1.4x1 0s
-
-
I^Mi^HBM
11
shaft
alley
90 min.
recycle
52
1416
-
-
-
4.58
-
-
160
462
MBMBPMMi
«
12
general
bilge
bilge
pump
14
632
30
21
15
.74
6.75
20.2xl05
20,000*
99
148
I^^«^»«B«
^^MHBMIi
13
general
bilge
separator
discharge
25
424
45
44
23
.05
7.10
5x1 04
78
88.7
14
general
bilge
bilge
pump
18
356
37
9
27
.06
7.35
4600
21
0.4
15
general
bilge
separator
discharge
48
84
24
16
12
.18
7.00
84x1 O3
33
4.3
16
shaft
alley
grab
49
148
17
16
12
.19
9.33
lOxlO2
61
35.8
*Fecal Coliform
-------
Table 2 (Continued). ANALYTICAL RESULTS FOR BILGE WATER
Bilge Sample Number
Sampling Location
Sample Type
Biological Oxygen Demand
(5 day)(mg/l)
Chemical Oxygen Demand
(mg/1)
Suspended Solids (mg/1)
Volatile Suspended Solids
(mg/1)
Settleable Solids (mg/1)
Phosphorus (mg/1)
PH
Total Col i form (MPN/ 100ml)
Total Organic Carbon (mg/1)
Oil (mg/1)
17
shaft
alley
grab
10
84
32
21
9
.13
7.50
7.2x105
-
2.8
18
shaft
alley
grab
55
68
42
20
30
.21
7.38
35x103
-
16.8
19
engine
room
grab
7
8
8
7
5
1.46
6.92
1.62x103
-
3.3
20
shaft
alley
initial
recycle
85
3800
70
50
13
.48
6.93
13.2x105
-
214
21
shaft
alley
30 min.
recycle
-
-
-
-
-
-
-
-
-
259
22
shaft
alley
60 min.
recycle
-
-
-
-
-
-
-
-
-
215
23
shaft
alley
90 min.
recycle
-
-
-
-
-
_
-
-
-
408
24
shaft
alley
120 min.
recycle
-
-
-
-
-
_
-
-
-
545
-------
Table 2 (Continued). ANALYTICAL RESULTS FOR BILGE WATER
Bilge Sample Number
Sampling Location
Sample Type
Biological Oxygen Demand
(5day)(mg/l)
Chemical Oxygen Demand
(mg/1)
Suspended Solids (mg/1)
Volatile Suspended Solids
(mg/1)
Settleable Solids (mg/1)
Phosphorus (mg/1)
pH
Total Coliform (MPN/lOOml)
Total Organic Carbon (mg/1)
Oil (mg/1)
25
shaft
alley
150 min.
recycle
80
5000
47
35
6
.40
7.29
5.9xl05
-
215
26
general
bilge
separator
discharge
50
760
75
67
12
.76
7.15
l.SxlO5
-
155
27
shaft
alley
grab
24
56
18
15
14
1.00
7.58
1.1x10
-
-
28
general
bilge
bilge
pump
75
1 1,040
476
266
56
12.8
4.24
0
0
-
29
general
bilge
bilge
pump
<1.0
40
32
12
14
.008
8.59
300
-
-
30
shaft
alley
grab
5.0
324
108
91
96
.013
8.32
900
-
-
31
general
bilge
bilge
pump
3.0
108
40
28
20
.010
8.24
400
-
21.5
32
general
bilge
bilge
pump
not tested
at Duluth
-
-
-
-
~
50.9
K)
Ln
-------
Table 2 (Continued). ANALYTICAL RESULTS FOR BILGE WATER
ho
o>
Bilge Sample Number
Sampling Location
Sample Type
Biological Oxygen Demand
(5 day)(mg/l)
Chemical Oxygen Demand
(mg/D
Suspended Solids (mg/1)
Volatile Suspended Solids
(mg/1)
Settleable Solids (mg/1)
Phosphorus (mg/1 )
PH
Total Col i form (MPN/ 100ml)
Total Organic Carbon (mg/1)
Oil (mg/1)
33
general
bilge
bilge
pump
-
-
-
-
-
-
-
-
-
39.9
34
general
bilge
bilge
pump
4.8
180
46
34
18
.042
8.03
200
-
29.9
35
shaft
a 1 1 ey
grab
380
3568
3600
3460
3358
0.37
7.73
5.7xl04
-
17%
36
general
bilge
bilge
pump
410
3624
1464
1310
1188
.16
7.27
5.84x10
-
35%
37
shaft
alley
grab
-
-
-
-
-
-
-
)
-
16%
38
general
bilge
bilge
pump
96
104
61
53
40
.42
7.11
300
-
57.6
39
general
bilge
bilge
pump
88
120
90
66
61
2.72
7.19
200
-
52.7
40
shaft
alley
grab
350
1856
1022
1016
989
.96
5.13
8xl04
-
2273
-------
Table 2 (Continued). ANALYTICAL RESULTS FOR BILGE WATER
Bilge Sample Number
Sampling Location
Sample Type
Biological Oxygen Demand
(5 day)(mg/l)
Chemical Oxygen Demand
(mg/D
Suspended Solids (mg/1)
Volatile Suspended Solids
(mg/1)
Settleable Solids (mg/1)
Phosphorus (mg/1)
PH
Total Col i form (MPN/ 100ml)
Total Organic Carbon (mg/1)
Oil (mg/1)
41
shaft
alley
grab
>350
2208
1754
1682
1192
3.73
7.30
2.1xl07
-
3%
42
eneral
bilge
bilge
pump
>350
1424
159
123
97
8.53
7.75
9.9xl05
-
1840
43
shaft
alley
grab
>350
49£80
-
-
-
1.14
-
2.5xl05
-
81%
44
shaft
alley
grab
107S
4496
375
326
206
2.80
7.32
4x1 06
-
-
45
shaft
alley
grab
-
-
-
-
-
-
-
-
-
202
46
shaft
alley
grab
>3700
-
-
-
-
.27
-
9.3xl05
-
72%
47
general
bilge
separator
discharge
7.5
88
426
406
219
0.075
9.44
-
-
6.8
48
general
bilge
separator
discharge
>740
3168
32
20
10
1.09
10.24
<100
-
847
K3
-------
Table 2 (Continued). ANALYTICAL RESULTS FOR BILGE WATER
Bilge Sample Number
Sampling Location
Sample Type
Biological Oxygen Demand
(5 day)(mg/l)
Chemical Oxygen Demand
(nig/1)
Suspended Solids (mg/1)
Volatile Suspended Solids
(mg/1)
Settleable Solids (mg/1)
Phosphorus (mg/1 )
pH
Total Col i form (MPN/ 100ml)
Total Organic Carbon (mg/1)
Oil (mg/1)
49
shaft
alley
grab
XI850
16,800
292
262
146
.93
8.68
3.52xl06
-
25%
50
shaft
alley
grab
1688
51,680
5216
5008
4912
.82
8.36
2.29xl06
-
21%
51
shaft
alley
grab
198
1152
210
196
110
.86
7.69
1.9xl05
-
339
52
shaft
alley
grab
1750
40,640
2280
2010
2240
8.80
9.43
4x1 06
-
63%
53
general
bilge
bilge
pump
1
64
14
9
5
.13
7.89
3.52xl04
-
11.4
54
shaft
alley
grab
1100
22,320
330
286
251
.064
9.96
4.3xl06
-
4%
55
general
bilge
separator
discharge
1
56
16
6
14
.040
10.34
9x1 02
-
19.2
56
shaft
alley
grab
-
192
97
71
74
.21
7.69
4.3xl06
-
76.2
10
00
-------
Table 2 (Continued). ANALYTICAL RESULTS FOR BILGE WATER
Bilge Sample Number
Sampling Location
Sample Type
Biological Oxygen Demand
(5 day) (mg/1)
Chemical Oxygen Demand
(mg/1)
Suspended Solids (mg/1)
Volatile Suspended Solids
(mg/D
Settleable Solids (mg/1)
Phosphorus (mg/1 )
PH
Total Col i form (MPN/lOOml)
Total Organic Carbon (mg/1)
Oil (mg/1)
57
general
bilge
bilge
pump
-
76
14
10
9
.033
9.44
1.36xlOJ
-
18.6
58
shaft
al ley
30 min.
recycle
10
496
2299
149
188
3.33
8.42
1.2xl04
-
143
-------
CO
O
Table 3. TRACE ELEMENTS IN BILGE AND SEWAGE SAMPLES BY EMISSION SPECTROSCOPY
(mg/1)
Sample Type :
Sample Number:
Element
Aq
Al
As
n ..I....
B
Ba
. - p: - '- - - ..!-.!
Be
Cd
Co
Cr
Cu
Fe
Mn
Mo
Ni
Pb
Sb .
Sn
Sr
Ti
y
Zn
Zr
Ca '
Mg 1
Sewage Bi I ge
1 1
<.08 <.03
1.3 1.2
<. 8 <. 3
<.3 O
<. 3 <. 1
<.08 <.03
<.3 <. 1
<.3 <.03
<.08 <.03
1.8 .2
3.3 2.4
.08 .06
<.08 <.03
<.08 <.03
<3. <3.
<.08 <.03
<.8 <. 3
<.08 <.03
<. 3 <. 1
<.08 <.03
T08 TT4
<.8 <. 3
<.08 <.03
3.9 16.4
0.4 9.6
Na 46.3 21.7
Si 1
2.7 17.6
Bi Ige
2
<.03
.8
<. 3
<.08
<.08
<.03
<.08
<.03
<.03
.2
2.4
.06
<.03
<.03
<3.
<.03
<.3
<.03
<.08
<.03
.15
< .3
<.03
18.0
9.0
16.1
5.5
Bilge
3
<.03
2.1
<.3
<1 .9
<.09
<.03
<.09
<.03
<.03
.4
3.3
.09
<.03
<.03
<3.
<.03
<.3
<.03
<.09
<.03
<.03
<. 3
<.03
15.9
8.5
8.7
22.1
Bi Ige
11
<0.04
6.8
<0.4
<0.1
<0.1
<0.04
<0.1
-------
Table 4. BASIC VESSEL AND DISCHARGE DATA FOR BALLAST WATER
Rail a st. Sample Number
Vessel Type
Vessel Nationality
Caran Type
Total Ballast. M.T.
Trio Ballast, M.T.
Oriainal Ballast Sourc*
Total Ballast Dis-
rfigrgprl. M.T.
Sampled Ballast Tank,
si7e. M.T.. location
Sampled Tank Dilution
inrou ue , 1*1 . i .
Source of Dilution
Ballast
1-3
Lake
Carrier
U.S.
15.3QQ
15,300
Detroit
15.300
1,280
#4
starboard
yen
Lakes
Huron &
Superior
4
Bulk
Carrier
Liberian
general
ft Grain
Detroit
248
248
)ort-side
top wing
n
n/a
5 & 6
Lake
Carrier
Canadian
grain
is.?nn
8,200
Buffalo
8 200
3,050
#2
port
o
n/a
7
Lake
Carrier
Canadian
grain
11,400
11,400
Lake
Erie
..1.&90Q ...
120
aft
._B£flk
0
n/a
8
Bulk
Carrier
British
steel
A Grain
7.300
7,300
Detroit
4T160
301
#3 starboard,
....top, wing
0
n/a
9-12
Lake
Carrier
U.S.
ore
13.700
12,300
Ashtabula
12,300
2,080
#6
starboard
0
Lake
Superior
13
Lake
Carrier
Canadian
^irain
11.400 .
11,400
Cleveland
10.550
910
#2
oort
0
n/a
-------
given in row 11; note that for the sixth vessel, ballast water dilution
occured although not for the sampled tank.
The analytical results for ballast water are presented in Table 5. These
may be related to the vessel data of Table 4 by the sample numbers. All
the samples were of the "grab" type with a Kemmerer Sampler supplied by
the Western Lake Superior Sanitary District. Sample type (row 2) is,
therefore, differentiated only on the basis of whether a single or
composite grab sample was obtained. Single grab samples were taken
initially a different depths to determine if water quality varied dras-
tically with depth. Composite samples were taken thereafter to reduce
the analytical burden. Row 3 gives the sampling and total water depth
for the ballast tank of interest; for sample 1, for instance, the
sample was taken at 1.5 meters (5 feet) from the tank bottom with a total
water depth of 3.0 meters (10 feet). Row 4 through 13 present the waste-
water analytical data as described in Section V. Again, the Edison
Water Quality Laboratory was requested to run oil and TOG analyses on
only a selected few samples. Samples 9 through 12 were from a trip from
Ashtabula, Ohio to Duluth. Samples 9 and 12 are for harbor waters as is
noted on the table; the ballast water of samples 10 and 11 was obtained
from the Ashtabula Harbor (sample 9) and discharged to the Duluth
Harbor (sample 12).
SEWAGE
Table 6 presents the basic vessel and rate data collected during the
study for sewage. This table is self-explanatory. Row 6, sewage system
size, refers basically to holding capacity and, therefore, does not apply
to the flow through system of the fifth vessel. Combined wastes (row 7)
refer to toilet, urinal, lavatory, shower, galley, shop and laundry
wastewaters; sanitary wastes are only the first two of these. Deodorizers
and disinfectants used are indicated in row 8. Recirculation or mixing
capability (row 9) refers to whether sewage could be stirred, manually
or otherwise, in a tank. Pumpout frequency (row 10) and volume (row 11)
were given to the field crew by either the vessel captain or engineer.
Apparent generations rates (row 12) were calculated using median values
for crew size, pumpout frequency and pumpout volume. Since the flow
through system was not instrumented, a generation rate estimate was not
possible.
Table 7 gives the analytical results for the sewage samples. The second
row indicates how the samples were taken: from wastewater effluent line
or pump or directly from the holding tank. The remainder of the table
is laboratory data; oil was analyzed for only a select 3 samples.
GARBAGE/REFUSE
Garbage generation by foreign vessels at Duluth-Superior Harbor during
the Fall of 1973 is summarized in Table 8; the 20 vessels listed repre-
sent approximately 25% of those which were serviced by Bayfront Marine Ser-
vices, Inc. during the 2-1/2 month field study period. Column 1 identifies
32
-------
Table 5. ANALYTICAL RESULTS FOR BALLAST WATER
Sample Number
Sample Type
Sample and Water Depth,
n./m. , (ft. /ft.)
Biological Oxygen Demand
(5 day), mg/1
Chemical Oxygen Demand,
mg/1
Suspended Solids,
mg/1
Volatile Suspended
Solids, mg/1
Settleable Solids, mg/1
Phosphorous, mg/1
pH
Oil, mg/1
Total Coliform, MPN/100 ml
Total Organic Carbon, mg/1
grab
1.5/3.0
(viol
0.4
12
2.0
1.6
Tr.
0.03
7.56
-
S.lxlO3
-
grab
3.0/3.0
(10/10)
0.6
8
2.4
1.6
Tr.
0.013
7.65
-
2.7xl03
-
3
grab
0.3/3.0
(1/10)
2.0
16
2.8
2.0
Tr.
0.017
7.70
-
1.5xl03
-
4
grab
1.2/2.4
(4/8)
2.5
58
7.6
4.4
2.J8
0.03
6.40
2.7
2.1x10*
-
5
grab
0/3.7
(0/12)
0.8
36
9
5
4
0.01
7.05
0.0
4.3xl03
5
6
grab
1.8/3.7
(6/12)
0.6
32
10
6
4
0.01
7.15
0.0
5.5xl03
5
7
composite
0,0.9,1.8/1.8
(0,3.6/6)
8.4
12
11
7
10
0.05
7.38
2.4
3x1 02
22
8
composite
P.I. 8. 3. 7/3. 7
(0,6,12/12)
5.5
140
126
57
109
0.39
6.50
43
l.SxlO5
27
sn f
composite
0.3.1.8,3.7/n.a
(1.6,12/n.a)
2.0
4
17
6
15
0.022
7.32
-
8x1 02
-
10
composite
0.3,3.7.7.3/7.6
(1,12,24/25)
2.0
12
15
4
10
0.025
7.71
-
5.6xl02
-
11
composite
0.3,3.7.7.3/7.6
(1.12.24/25)
2.8
40
7
7
3
0.17
8.01
-
6.6xl02
-
12
7}
composite
0.3.1.8.3.7/n.4
(1.6.12/na)
1.0
4
4
4
4
0.013
8.09
-
5.2X102
-
13,
gr»!>
1.5/7.6
(5/26) .
1.0
16
13
11
9
0.087
8.40
0.7
l.Oxlo5
-
Emission spectroscopy data presented separately. (1) Ashtabula Harbor water. (2) Duluth Harbor water
-------
Table 6. BASIC VESSEL AND RATE DATA FOR SEWAGE
Sample Number
Vessel Type
Vessel Nationality
Crew Size
Sewage System Type
Sewage System Size, liters (gallons)
Waste Tvoe
Chemicals Used
Recirculation Capability
Pumoout Frequency Reported (days)
Pumoout Volume Reported, liters (gallons'
Apparent Generation Rate, liters/day/man
(gal ./day/man)
i
Cutter
U. S.
45
holding
18,900
(5000)
combined
beozene &
D05725
(1)
none
1/2-1/3
3800-11,40(
(1000-
3000)
415
(no)
?
Bulk
Carrier
Liberian
25-27
septic
tank
5,700
(1500)
combined
none
none
-
)
-
(1) Brulin Co. (3) Hus
(2) Monogram Industries hold-
1
Lake
Carrier
U. S.
38
holding
80,200
(21,200)
combined
NaOCl
hand
stirrer
5-7
75,700
(20,000)
340
(90)
d
Lake
Carrier
U. S.
32
holding
(3)
6,800
(1,800)
sanitary
MC 1000
(2)
hand
stirrer
10
5100-5500
(1350-
1450
17
(4.5)
R
Lake
Carrier
U. S.
32
Flow
through
(Biological
-
combined
Ca(OCl)2
-
-
-
n/a
6
Lake
Carrier
Canadian
32
holding
15,100
(4000)
sanitary
Ca(OCl)2
none
4
15,100
(4000)
115
(30)
water reused unti seni to
'ng tank (every 4 or 5 days).
-------
Table 7. ANALYTICAL RESULTS FOR SEWAGE
Sample Number
Biological Oxygen
Demand (5 day) (mg/1)
Chemical Oxygen
Demand (mg/1)
Suspended Solids (mg/1)
Volatile Suspended
Solids (mg/1)
Settleable Solids (mg/1)
Phosphorus, mg/1
PH
Oil (mg/1)
1*
I-roni 1"
effluent
250
708
260
226
106
13.6
7.07
220
2
bled
from
900
2000
644
476
208
20.8
7.00
3.1
3
grab
from tank
1700
6856
8520
8272
8168
18.6
6.70
-
4
bled from
tank w/
820
44,400
692
524
484
67.3
8.52
-
5
from 1"
effluent
68
344
185
162
163
6.4
8.39
-
fi
from 4"
effluent
,1.1 HQ
1112
628
520
452
9.9
8.12
57.3
Ul
* Emission spectroscopy data presented separately
-------
Table 8. SHIPBOARD GARBAGE GENERATION
Vessel
Nationality
Cypriot
Yugoslavian
Spanish
Greek
Greek
Greek
British
Panamanian
Belgian
British
Norwegian
Panamanian
British
Greek
Norwegian
Norwegian
Yugoslavian
Yugoslavian
French
Greek
Total
Garbage Weight,
Net (kg)
635
435
270
435
710
890
425
1,080
1,805
1,480
845
735
590
660
770
955
425
915
170
180
14,400
Days for
Generation
7
10
3
8
10
6
3
6
11
8
4
5
5
9
5
6
7
7
2
2
Crew
Size (no.)
38
36
27
29
32
39
38
27
30
37
30
31
32
26
28
38
34
34
32
33
Generation
Man -days
266
360
81
232
320
234
114
162
330
296
120
155
160
234
140
228
238
238
64
64
4,036
Generation
Rate
(kg/man -day)
2.4
1.2
3.8
1.9
2.2
3.8
3.8
6.7
5.4
5.0
7.0
4.8
3.8
2.8
5.5
4.2
1.8
3.8
2.7
2.8
Average Generation Rate: 3.6 kg/man-day (7.9 Ibs./man-day)
36
-------
vessels by nationality. Note that only foreign garbage/refuse is con-
sidered because of agricultural regulations relating to possible infesta-
tion and requiring disposal within 24 hours. Only foreign vessel garbage/
refuse is collected at a central location in the harbor and amenable to
sampling; therefore, U. S. and Canadian vessels could not be considered
in this study. The weight of garbage from each vessel (column 2) was
determined by weighing the collection truck on a truck scale at a local
scrap yard. The days for garbage/refuse generation and crew size
(columns 3 and 4, respectively) were determined through interview of vessel
captains or other knowledgeable officers. Days for generation multiplied
by crew size gives generation man-days (column 5); this latter quantity
divided into the net garbage weight gives the generation rate (column 6).
The totals (for all the vessels) were used similarly to come up with
the overall average generation rate listed at the bottom of Table 8.
Table 9 gives garbage sample analyses. Samples 1 and 2 were from the
Belgian and 3 and 4 from the third Yugoslavian vessel listed on Table 8.
The top portion of Table 9 contains the waste category compositions for
the samples as determined by hand sorting (see Section V). Note that
samples 1 and 4 contained exceptionally large amounts of textiles and
metals respectively; the former appeared to be waste shop rags, the latter
waste iron scale. The weight percent water and the dry and wet heat
values at the bottom of the table were determined by the Twin Cities
Testing Laboratory. Since these basic values apply only to the total
of food, paper and wood wastes, the total heat value for the garbage/
refuse had to be estimated. Data from Principles and Practices of
Incineration (p. 7) by Richard C. Corey was used to adjust the laboratory
data; the estimated total heat values so obtained are for wet garbage/
refuse.
DUNNAGE
Only two vessel loads of dunnage were available during the last 2 months
of the field study; due to the attention required of the other waste
categories, dunnage was not considered during the first two weeks. A
bulk carrier (Greek) carrying steel unloaded 7.7 metric tons (16,920
pounds) of dunnage in early September 1973. In mid-September another bulk
carrier unloaded approximately 9.1 metric tons (20,000 pounds) although
the field crew, due to complaints by longshoremen, could only weigh half (5)
the truckloads. This, unexpectedly, proved to be the last dunnage available
during the course of the project.
Table 10 presents a size distribution analysis for the first load of
dunnage weighed. The distribution was determined as described in Section V.
This load of dunnage contained about a pound or 450 grams of scrap metal,
mainly in the form of nails, and was comprised entirely of pine lumber.
Table 10 shows uncoupled distributions for length, width, and thickness.
37
-------
Table 9. GARBAGE SAMPLE ANALYSES
Sample Number
1
2
3
Component % by weight
Food Wastes
Paper Products
Plastic, Rubber, Leather
Textiles
Wood
Metals
Glass, Ceramics
Total
53.4
5.5
2.0
23.2
0.4
2.1
13.4
100.0
71.3
8.6
6.0
1.7
0.0
3.9
8.5
100.0
62.8
11.4
1.1
4.5
0.0
1.7
18.5
100.0
4
59.4
8.2
0.5
1.5
0.0
21.7*
8.7
100.0
*Includes 20 percentage points due to waste iron scale.
Weight % H20**
Dry Heat Value, cal/g,
(BTU/lb)**
Wet Heat Value, cal/g,
(BTU/lb)**
Estimated Total Heat Value,
cal/g, (BTU/lb)***
32.2
5120
(9210)
3470
(6240)
3080
(5550)
31.8
5300
(9540)
3620
(6510)
3350
(6030)
51.4
4010
(7220)
1950
(3510)
1690
(3050)
40.0
3130
(5640)
1880
(3380)
1380
(2480)
** Applies to total of food, paper and wood wastes only.
*** Lab value corrected to account for the other four waste components
by using published data.
38
-------
Table 10. DUNNAGE SIZE DISTRIBUTIONS
Length
Width
VO
Range(m)
0-0.3
0.3-0.61
0.61-0.91
0.91-1.22
1.22-1.52
1.52-1.83
1.83-2.13
2.13-2.44
2.44-2.74
2.74-3.05
3.05-3.35
3.35-3.66
3.66-3.96
>3.96
No. Percent
0.0
0.0
1.6
5.4
6.0
16.3
14.7
9.8
26.7
5.4
11.4
2.2
0.5
0.0
100.0
Average Length: 2.2 m.
Range (cm)
0- 5.1
5.1-lo;i
10.1-15.2
15.2-20.3
20.3-25.4
25.4-30.4
30.4-35.5
35.5-40.6
40.6-45.6
45.6-50.7
50.7-55.8
55.8-60.9
>60.9
No. Percent
0.0
15.9
35.4
20.7
12.5
6.0
4.8
2.1
1.0
0.0
1.1
0.5
0.0
TOO"
Average Width: 16.6 cm.
Size (cm)
<2.5
2.5
3.8
5.1
6.3
7.6
>7.6
No. Percent
0.0
88.0
6.0
0.5
0.5
4.9
0.0
TooTo
Average Thickness: 2.9 cm
-------
SECTION VII
ANALYSIS OF RESULTS
Except for bilge water, the results of this study can be analyzed and
summarized directly from the data of Section VI. The analyses which follow
serve to highlight the major results and conclusions of the study. More
extensive field work could, of course, lead to much more data and statis-
tically exact analyses of same, including estimates of pure error. For
the purposes of this study, analysis alms primarily at quantifying
environmental and pollution problems for the five waste categories.
BILGE WATER
Only 26 of the 37 vessels sampled for bilge water were able to supply use-
ful data for estimating generation rates. Table 11 summarizes the bilge
water generation rate data. The number of vessels in a particular
category is given along with apparent generation rate and standard
deviation for same. The standard deviations are not estimates of pure
error, although the latter contributes to them; rather, the deviations
reflect real differences in bilge water generation from vessel to vessel
These variations are large and very significant. Of course, more exact
insight into vessel-to-vessel variations can be gained by detailed study
of the data in the previous section.
The average vessel studied generated 6,650 liters or 1,760 gallons of
bilge water per hour. This is generally in line with the values used
(not estimated through extensive survey) in other studies. For instance
Environmental Quality Systems, Inc., in a study for the Upper Great '
Lakes Regional Commission used a typical generation rate of 12,000 liters per
hour; Frederic R. Harris, Inc. used a value of 600 liters per hour for
pre-1953 steam driven Lake carriers. Steam driven vessels apparently
generate bilge water at about 2-1/2 times the rate of diesels. This can
be explained by the formers' relatively greater ages and sizes plus leakier
propeller shaft seals. This result, of course, applies to vessels at
Duluth-Superior, or, perhaps more generally, to vessels plying the Great
Lakes. The ages, vessel sizes and predominant steam propulsion of the
United States fleet on the Lakes is reflected in the generation rates
for U. S., Canadian and other foreign vessels: about 10,000, 5,700 and
5,100 liters per hour respectively. Similarly, the data indicates that
vessels 20 or more years old generate bilge water at about twice the rate
of younger vessels.
The last 3 rows of Table 11 give rates for 3 categories of bilge water
generators: low, medium and high, defined by requiring equal numbers of
vessels in each group. A 3 orders of magnitude spread in bilge water
generation is apparent. These categories were developed mainly to provide
40
-------
Table 11. APPARENT BILGE WATER GENERATION RATES
Vessel Category
All
Steamers
Diesels
United States
Canadian
Foreign
New (0 to 5 years)
Middle Aged (6 to 19 years
Old (20 years or more)
Low Bilge Water Generator
(0 to 30 gph)
Medium Bilge Water Generatt
(30 to 1500 gph)
High Bilge Water Generator
(greater than 1500 gph)
umber
26
11
15
7
7
12
8
9
7
9
Dr
9
8
pparent Generation Rate!
liters/hr. (qal./hr.)
6,650
10,600
4,110
10,300
5,720
5,060
5,400
4,720
10,300
32.
2,240
19,100
(1,760)
(2,790)
(1,090)
(2,720)
(1,510)
(1,340)
(1,430)
(1,250)
(2,720)
6 (8.6)
(550)
(5,000)
Standard Deviation
liters/hr.
9,090
9,920
8,130
10,200
9,290
8,890
10,100
8,290
10,200
32.8
2,080
6.570
41
-------
a basis for comparing waste oil charged to the bilges of vessels gen-
erating different amounts of bilge water.
Table 12 summarizes the bilge water data on oil and COD. The same comments
about standard deviations hold for this table as for Table 11. Similar
tabulations could be made for other important wastewater parameters (e.g.,
BOD5 and suspended solids) but the general conclusions would be about the'
same. The average BODc for the recycle and bilge pump samples collected
was 70 mg/1 and suspended solids 250 mg/1.
Rows 1 and 2 of Table 12 indicate a large discrepancy between recirculation
and grab sampling while rows 3 and 4 show differences between the samples
taken by the field engineer and field technician. Even grab samples which
the field engineer thought were acceptable (row 5) had markedly higher oil
contents than the recycle samples. The conclusion is that grab sampling
is unacceptable for bilge water. The general bilge samples split from
bilge pump effluent streams were the most representative taken during the
study; the oil and COD values for these samples compare very well with
those for the shaft alley recycle samples on an average basis. Re-
circulation, then, in the absence of samples taken from an operating
bilge pump seems to be a good technique for obtaining bilge water samples.
It, of course, requires taking more than sample bottles onboard vessels
(e.g., pump and hose) and a suitable power source. Values from previous
studies are 200 mg/1 oil used by Environmental Quality Systems, Inc. and
17 to 154 mg/1 oil measured by Frederic R. Harris, Inc. When compared
to the bilge pump samples, the oil and COD analyses for samples taken from
oil/water separator effluents indicate little in the way of beneficial
performance.
Although the overall results by vessel nationality (rows 14, 15 and 16)
are biased by the inclusion of grab sample analyses, the general trend
fits the impresssions formed by the field crew. That is, U. S. vessels
"appeared" cleaner than Canadian vessels which, in turn, appeared cleaner
than other foreign vessels; the visual differences, however, were not so
dramatic as implied by the analyses. Cleaner operations, more dilution
water (e.g., from seal leaks) and use of less diesel power for U. S. and
Canadian vessels all serve to explain the analytical differences. The
overall results for diesel vessels in the second from the bottom row are
obviously biased by the inclusion of 14 grab sample analyses while the
similar result for steamers is biased only slightly (i.e., without the
5 grab samples oil would be about 170 milligrams/liter); the results for
recycle and discharge sampling of diesels indicate substantially lower oil
and COD values. The large relative values for oil and COD resulting from
grab (submersion) sampling of diesels are illustrated by row 2 on the
second page of Table 12. On the other hand, grab sampling of steamers
produced relatively low oil and COD concentrations; this occured for
apparently the same reasons as discussed for U. S. vessels above (e.g.,
dilution effects). Note that grab sampling appears in general to be a
very biased technique.
Lines 22, 23 and 24 of Table 12 give results for new, middle-aged and
42
-------
Table 12. STATISTICAL SUMMARY OF BILGE WATER DATA FOR OIL AND C.O.D.
Shaft Alley/Recycle Samples
Shaft Alley/Grab Samples
Shaft Alley/Grab Samples by
Field Engineer
Shaft Alley/Grab Samples by
Field Technician
Shaft Alley/Grab Samples
Preferred by Field Engineer
General Bilge Samples
Off Bilge Pumps
General Bilge Samples
Off Separators
Recycle & Bilge Pump Samples:
United States
Canadian
Foreign Flag
Grab Samples:
United States
Canadian
Foreign Flag
United States Vessels
Canadian Vessels
Foreign Vessels
Diesel-Driven Vessels
Steam-Driven Vessels
No.
11
17
12
5
7
11
6
12
4
6
2
4
12
15
13
20
24
24
(mg/1 )
251
178,000
122,000
312,000
41,500
206
187
246
55
309
10
168,000
196,000
199
51,600
135,400
140,800
140
D Oil
Gng/D
177
273,000
186,000
414,000
93,100
543
329
517
61
138
10
309,000
282,000
468
174,000
240,000
245,000
375
Mo.
12
17
13
4
9
11
6
14
6
3
4
4
10
19
15
15
19
30
(mq/1)
1,167
11,900
7,460
14,600
4,570
1,285
1,430
1,285
175
3,040
114
15,800
13,800
973
4,540
10,400
11,700
670
SD COD
(mg/1)
1,588
18,160
17,660
23,500
8,620
3,250
2,030
2,850
162
2,435
142
19,600
20,000
2,480
11,500
16,900
16,900
2,000
43
-------
Table 12 (continued)
Category
Diesels by Recycle or
Discharge Sampling
Diesels by Grab Sampling
Steamers by Grab Sampling
*New Vessels
*Middle-Aged Vessels
*01d Vessels
**Low Bilge Water Generator
**Medium Bilge Water Generator
**High Bilge Water Generator
For Recycle or Discharge Samp
**Low Bilge Water Generator
(44 gms/hr. oil waste)
**Medium Bilge Water Generator
(620 gms/hr oil waste)
**High Bilge Water Generator
(860 gms/hr oil waste)
No.
10
14
5
18
14
13
8
15
12
les:
2
9
8
Mean Oil
(mg/1)
295
241,000
27
86,800
78,000
206
137,000
58,900
56,900
1,343
277
45
SD Oil
(mg/1)
246
283,000
31
202,000
179,000
506
281,000
229,000
112,000
702
165
44
Nn
7
12
7
14
16
17
8
14
11
2
8
8
Mean COD
(mq/1)
2,510
17,100
126
7,190
7,220
1,050
17,700
6,210
976
2,300
3,500
170
SD COD
(mq/1)
2,180
19,500
106
13,500
15,900
2,620
21,500
11,040
1,390
1,230
3,580
158
* Mew Vessels: 0 to 5 years
Middle-Aged Vessels: 6 to 19 years
Old Vessels: 20 years and over
** Low Bilge Water Generator: 0 to 30 gph
Medium Bilge Water Generator: 30 to 1500 gph
High Bilge Water Generator: greater than 1500 gph
44
-------
old vessels; although biased by grab samples, this data shows that older
vessels discharge bilge water with lower oil concentrations due primarily
to dilution. The remaining 6 rows of Table 12 generally illustrate the
same fact, even when grab samples are included for low, medium and high
bilge water generators (rows 25, 26 and 27). When only recycle and
discharge sampling is considered for these 3 degrees of bilge water
generation, dilution effects are even more apparent. In fact, the
relative constancy of the apparent waste oil discharged with the bilge
water is quite surprising (see the parenthetical expressions in the last
3 lines of Table 12); the average oil discharge rate is about 660
grams/hour for the 19 samples considered.
In conclusion, bilge water appears to be a serious pollution problem at
the Duluth-Superior Harbor and on the Great Lakes. Improvement in oil/
water separator technology would seem to be in order. Bilge water
generation rates are significantly larger, even for new and diesel powered
vessels, than often reported in the literature (e.g., "Port Collection and
Separation Facilities for Oily Wastes, Volume I" by Frederic R. Harris,
Inc.); however, the rates are lower than estimated by the authors of this
report in "Duluth-Superior Harbor Pollution Control Program" prepared
for the Upper Great Lakes Regional Commission (29 versus 53 gallons/
minute). The study has shown that bilge water contains 200 to 250 milli-
grams/liter of oil and substantial quantities of other common wastewater
contaminants. Relative to bilge water generation rates and oil concen-
trations, the waste oil discharge with bilge water is relatively consistent
at 660 grams/hour on the average.
BALLAST WATER
Conclusions regarding non-oily ballast water can be made directly by
inspection of Tables 4 and 5 in Section VI. First, deballasting accounts
for a considerable discharge of water to the Harbor: 13,800, 6,800 and
8,800 metric tons per visit for ore, bulk grain and all vessels, res-
pectively. These quantities represent 95, 65 and 75 percent of total
ballasting capacity, respectively; the overall value is somewhat higher
than previously estimated in the report "Duluth-Superior Harbor Pollution
Control Program" (67 percent). At present, ballast dilution is, for all
practical purposes, negligible at one percent. The analytical results
for sample number 12 in Table 5 give an idea as to the existing quality
of the water in the Duluth-Superior Harbor. With one exception, the
ballast water analyses indicate only from 1/2 to 10 times the magnitudes
of the various wastewater parameters for the harbor water. The exception
is the single British ship (sample 8) which was carrying quite highly
contaminated ballast water; 43 milligrams per liter of apparent oil
and grease in a supposedly non-oily wastewater is somewhat disconcerting.
Given the analyses of Table 5, a good rule of thumb appears to be that
non-oily ballast water has 2 to 5 times the contaminants as harbor water
(not including the results for sample 8). Sample series 9 to 12 in-
dicates that Ashtabula Harbor water is slightly more contaminated than
that at Duluth; also, ballast water changed enroute to Duluth: solids
45
-------
decreased and COD and phosphorous increased due, apparently, to the
addition of chemicals and contaminants. Of course, much more severely
contaminated water at ballasting points (e.g., upriver at Cleveland)
would present more of a problem (e.g., the British ship); however, the
project did not allow for extensive harbor water sampling.
In conclusion, as compared to other pollution control and environmental
problems at Duluth-Superior, non-oily ballast water does not appear to
be a very important wastewater.
SEWAGE
Tables 6 and 7 in Section V gave the basic vessel, rate and analytical
data for sewage; again, conclusions are apparent directly from the data.
There is considerable variation in type and capacity for sewage systems
onboard vessels. Chemical addition is usual practice and, especially
for holding systems, could present a real hazard to shoreside wastewater
treatment facilities which accept vessel sewage. This is reflected in
the number of very high COD analyses in Table 7. Sewage generation rates
are in line with accepted values. For this study, combined wastewaters
are apparently generated at about 380 liters or 100 gallons per man per
day, and sanitary wastewaters at 95 liters or 25 gallons per man per day
(accounting for the recycle onboard vessel 4). Municipal wastewaters
typically have about 200 milligrams per liter of both BOD5 and suspended
solids. The average values for these two parameters in this study are
750 and 1,600, respectively. This indicates something close to an order
of magnitude greater contamination. Vessel sewage is, however, probably
overly dosed with bactericides which could also severely impact waste-
water treatment facilities.
In conclusion, sevage generation onboard vessels occurs at rates similar
to those presented in the literature and used in design. These wastes
are, however, highly contaminated compared to municipal wastewaters and
include large amounts of chemicals and even oil.
GARBAGE/REFUSE
Data on garbage generation and analyses was presented in Section VI,
Tables 8 and 9. Significant variation was observed in generation rates
from vessel-to-vessel (about a 2-to-l peak-to-average ratio) with an
average of 3.6 kilograms or 8 pounds of garbage/refuse per man-day.
This is high compared to the typical municipal rate of 5.5 pounds per
man-day used by the Bureau of Solid Waste Management. No particular
dependence on vessel nationality can be discerned from the data. Using
Principles and Practices of Incineration as a reference, food wastes com-
prise a much higher percentage of the total for vessels than for municipal
garbage/refuse (62 versus 12 percent); paper wastes are, on the other hand,
much less prevalent (8.4 versus 42 percent) while twice as much glass is
found (12 versus 6 percent). A problem peculiar to vessels is the
inclusion of highly variable quantities of industrial-type wastes; sample
analyses 1 and 4 reflect large amounts of shop rags and iron scale from
46
-------
hull cleaning respectively.
In terms of incineration of vessel garbage/refuse, two problems arise:
(1) high moisture content and (2) low heat value. The samples collected
at Duluth-Superior Harbor contained an average of 39 percent water which
is about twice the value for municipal garbage/refuse. This much water
will not only decrease combustibility, but will also cause problems
with garbage/refuse transport and handling. The average total heat
value for the four samples was 4,300 Btu per pound which compares un-
favorably to 6,200 Btu per pound for typical municipal solid wastes.
In conclusion, garbage/refuse is generated at higher rates onboard foreign
vessels than municipally and contains large amounts of water which
account, primarily, for low total heat values.
DUNNAGE
Not much more than has already been said can be said about dunnage due
to lack of the basic waste material during the study period. Data is
much too sketchy to allow any sort of an estimate of generation rates.
The single size distribution analysis which was made indicates that pre-
vious observations of wide size variations hold true. Metal content
measurements were also too limited to permit any generalizations al-
though, once again, observations appear correct that the form and
amounts of this waste are highly variable.
GENERAL
The data gathered during the study indicates that high variability is
the only uniform characteristic of vessel wastes. Although average
results can be very useful for describing waste problems on a harbor-
wide basis, they cannot be so used for individual vessels. From vessel-
to-vessel, both waste quality and quantity can vary by orders of magnitude.
The study has shown this most dramatically for bilge water, but it also
holds for the other four waste categories. Control or treatment systems
for a particular vessel or small number of vessels must be designed
using data highly specific to same. Such data is not provided by a
study such as that which has just been discussed; rather, special studies
will be required.
47
-------
SECTION VIII
REFERENCES
1. American Public Health Association, American Water Works Association,
and Water Pollution Control Federation, Standard Methods for the
Examination of Water and Wastewater. 13th Edition, Washington, D C
(1971). ~~
2. Carruth, Dennis E., and Klee, Albert J., "Analysis of Solid Waste
Composition Statistical Technique to Determine Sample Size", U. S.
Department of Health, Education, and Welfare, Bureau of Solid Waste
Management, (1969).
3. Corey, Richard C., Principles and Practices of Incineration, Wiley-
Interscience, New York, (1969).
4. Environmental Protection Agency, "Methods for Chemical Analysis of
Water and Wastes 1971", 16020-7/71, Cincinnati, Ohio, (1971).
5. Environmental Protection Agency, Office of Research and Monitoring,
"Physical Chemical and Microbiological Methods of Solid Waste
Testing", EPA-6700-73-01, Cincinnati, Ohio, (1973).
6. Environmental Quality Systems, Inc., "Duluth-Superior Harbor Pollution
Control Program", Prepared for Upper Great Lakes Regional Commission,
Washington, D. C., (1973).
7. Forster, Richard L., Moyer, J. E., Firstman, S. I., "Port Collection
and Separation Facilities for Oily Wastes", Volumes I-IV, Frederick R.
Harris, Inc., Report to U.S. Department Commerce, Maritime Adminis-
tration, Washington, D. C., (1973).
8. Moreau, J. 0., "Oil/Water Interface Detector", Final Report by Esso
Research and Engineering Company, For U. S. Department of Commerce,
Maritime Administration, Washington, D. C., (1971).
9. Weinberger, Leon W., Waller, Robert, and Gumtz, Garth D., "Vessel
Waste Control at Duluth-Superior Harbor", 1973 Pollution Control
in the Marine Industries, International Pollution Control Association,
Washington, D. C., pp. 221-226, (1973).
48
-------
SECTION IX
APPENDIX
WASTEWATER ANALYSES DESCRIPTIONS
The following pages describe briefly the particular wastewater analyses
which were used during the vessel waste study at Duluth-Superior.
BODc, COD, suspended solids, volatile suspended solids, settleable solids,
phosphorous, pH and total coliform bacteria were determined at the WLSSD s
Duluth Sewage Treatment Plant Laboratory. Sample transport from the port
to the laboratory took only about 10 minutes, and sample refrigeration
was possible at both the Port Authority Terminal and the laboratory.
Therefore, little sample degradation could take place. Descriptions of
the 8 tests mentioned above are as follows:
(1) BOD5
Five day BOD tests were performed according to the procedures set
forth in "Standard Methods", Section 219. BOD is an empirical test
used for determining the relative oxygen requirements of waste-
water. Oxygen demand in this particular test is usually attributed
to carbonaceous matter. BOD measurements somewhat attempt to
simulate natural oxidation. Typical precision is + 5%.
(2) COD
COD is a measure of the amount of organic matter oxidized by a strong
chemical oxidant. It is especially valuable in industrial waste
studies. COD was determined according to "Standard Methods",
Section 220. The precision of COD tests is about 6.5% at 200 mg/1.
Usually COD reflects 95 to 100% of the theoretical value for
oxidizable matter. Organic compounds such as benzene and acetic
acid are not oxidized by this dichromate method, however.
(3) SUSPENDED SOLIDS
Suspended solids were determined according to "Standard Methods",
Section 224C. Solids are isolated for drying and weighing by
filtration with a Gelman filter. Precision depends upon the amount
of suspended matter in the sample; at 242 mg/1, + 10% variation
is typical. Total suspended matter is a gross indicator of
pollution in wastewaters.
(4) VOLATILE SUSPENDED SOLIDS
Volatile suspended solids are determined in a continuation of the
49
-------
procedure for suspended solids ("Standard Methods", Section 224D).
Filtered residue is ignited for 15 minutes at 550°C. At 170 mg/1,"
coefficients of variation are about 6.5%. Volatile suspended
solids are usually used as an indicator of the amount of organic
solids contained in a wastewater sample.
(5) SETTLEABLE SOLIDS
For the majority of samples, concentrations were determined on a
weight basis (mg/1). The procedure used is stated in "Standard
Methods", Section 224F. The portion of the wastewater that either
settles or floats is determined by difference (suspended less
nonsettleable matter). Settleable solids represent, more than
anything else, an operational parameter relating to possible
separation in settling basins, thickeners, etc.
(6) PHOSPHORUS
The phosphorus level in wastewater was determined by the stannous
chloride technique as described in "Standard Methods", Section 223E.
This colorimetric method is most suited to concentration of
0.01-6 mg/1. Relative error is usually between 4 and 5%. Phosphorus
is, of course, a key element (often limiting) for biological growth
in water bodies.
(7) PH
A Corning pH meter was used for determining the pH of wastewater
samples. The limit of accuracy for this instrument was given as
0.1 pH unit with possible interference from high levels of dissolved
salts, oil, and fine particles. Of course, pH gives an indication
of the acidity or basicity of a water sample; neutral samples are
generally considered in the pH range of 5.5 to 8.5.
(8) TOTAL COLIFORM BACTERIA
Total coliform bacteria is done by the membrane filter techniques
as set forth in "Standard Methods", Section 408A. Disposable pre-
sterilized pipettes and petri dishes were used as well as Erds broth
for growing the cultures. Enrichment (as used for drinking quality
water) was not necessary due to the high levels of coliform bacteria
in the wastewater samples. Total coliform bacteria serve as an
indicator of wastewater contamination by microorganisms.
Oil and grease; total organic carbon and metals (by emission spectroscopy)
in wastewater samples were determined by the Edison Industrial Waste
Treatment Research Laboratory of the U. S. Environmental Protection Agency.
Since these analyses necessarily involved sample preparation and shipment
from Duluth, Minnesota, to Edison, New Jersey, care was taken in selecting
particular samples for analysis. Analyses were generally in conformance
to "Methods for Chemical Analysis of Water and Wastes, 1971" by the
50
-------
Environmental Protection Agency. Descriptions of the tests are as
follows:
(1) TOTAL ORGANIC CARBON (TOC)
TOG is determined by catalytic combustion of a micro sample of
wastewater followed by infrared analysis for C02- Such an analysis
provides a much better indicator of contained organics than does
volatile suspended solids. Comparison with analyses for oil and
grease then permits assessment of "biological" contamination.
Typical precision is + 8% at 100 mg/1.
(2) OIL AND GREASE
Oil analyses were run during the study by hexane Soxhlet extraction
at low pH followed by evaporation and weighing of the residue. The
procedure essentially defines greases and oils. EPA's method can be
used for a concentration range of 5 to 1000 mg/1; for higher con-
centrations (as were often encountered during the study) volumetric
techniques are used. Precision and accuracy data are not available.
(3) METALS BY EMISSION SPECTROSCOPY
Emission spectroscopy permits the simultaneous determination of a
large number of metals (27 for the 8 samples analyzed in this study)
in a micro sample. Briefly, the constituents of a sample are
ionized in a flame; upon returning to their ground state, each
metal emits characteristic radiation. A spectral scan of the
emissions combined with standard emission data permits quantification
of the metals in wastewater samples. The resulting concentrations
can then be used to determine whether or not metals are present
in toxic or unusual amounts.
51
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-670/2-74-097
3. RECIPIENT'S ACCESSION>NO.
4. TITLE AND SUBTITLE
CHARACTERIZATION OF VESSEL WASTES IN
DULUTH-SUPERIOR HARBOR
5. REPORT DATE
December 1974; Issuing Date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Garth D. Gumtz, David M. Jordan, and Robert Waller
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORG MMIZATION NAME AND ADDRESS
Environmental Quality Systems, Inc.
6110 Executive Boulevard, Suite 750
Rockville, Maryland 20852
10. PROGRAM ELEMENT NO.
1BB038; ROAP 21-BBU
11. dOWMR***/GRANT NO.
R-802772
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
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
Five wastes from United States, Canadian, and foreign commercial vessels were studied
at the Duluth-Superior Harbor during late 1973: bilge water, non-oily ballast water,
sewage, garbage/refuse, and dunnage. Vessels generate bilge water at about 6,650
liters/hour with an average oil content of about 225 milligrams/liter. Waste oil
which is apparently discharged to bilges (about 600 grams/hour) appears more con-
sistent than either of these two parameters. Bilge water is a substantial pollution
problem: on the average about 40 liters (10 gallons) of oil may be discharged during
each day a vessel spends in the harbor. Although containing about twice the common
water quality contaminants as the harbor waters, ballast water is not a significant
environmental problem. Large quantities are, however, discharged: about 9,000
metric tons/visit by lake and bulk carriers. Sewage is apparently generated onboard
vessels consistent with accepted design rates (100 gallons/man/day). Although
largely under control, sewage may impact wastewater treatment facilities by being
significantly stronger than typical municipal wastewaters. Chemical additives
commonly found in vessel sewage may adversely affect wastewater treatment facilities
and harbor receiving waters.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Ships
Harbors
Water pollution
Refuse
Minnesota
Duluth-Superior Harbor
Lake Superior
Vessels
Waste characterization
13B
3. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
21. NO. OF PAGES
60
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
52
U. S. GOVERNMENT PRINTING OFFICE: 1975-657-591/53A6 Reg i on No. 5-
------- |