ENVIRONMENTAL PROTECTION AGENCY
NORTHWEST REGION, PACIFIC NORTHWEST WATER LABORATORY
THE  BARGED
          OCEAN DISPOSAL
                          OF WASTES
               A Review of Current Practice and
                        Methods of Evaluation
                                    July

-------
              THE BARGED OCEAN DISPOSAL OF WASTES
     A REVIEW OF CURRENT PRACTICE AND METHODS  OF  EVALUATION
B. D. Clark, W. F. Rittall, D.  J. Baumgartner,  and K.  V.  Byram
                Environmental  Protection Agency
              Pacific Northwest Water Laboratory
                 200 S. W. Thirty-fifth Street
                   Con/all is,  Oregon 97330

-------
             EPA Review Notice

This report has been reviewed by the Environ-
mental  Protection  Agency  and  approved for
publication.   Approval does not signify that
the contents  necessarily  reflect  the views
and policies of the Environmental  Protection
Agency,  nor does  mention of trade  names or
commercial  products  constitute  endorsement
or recommendation for use.

-------
                               ABSTRACT

     This report consists of a broad scope examination of barged ocean
disposal  of liquid and solid wastes.  Basic discussions include:  the
physical  characteristics of various selected wastes, economics as a
function of haul distance, reported effects of past discharge operations
and the relative importance of environmental factors such as density
and current profiles.  The major emphasis of the report centers on
physical  fate prediction methods and describes the physical transport
in four separate steps:  convective descent, collapse, long term disper-
sion and bottom transport or resuspension.

     An existing mathematical model developed by Koh and Fan is used and
demonstrates the complex nature of some of the more obvious parameters,
the potential usefulness of the approach to coastline management efforts
while serving as a vehicle for the discussion of current state of the
art limitations and research needs.

-------
                               CONTENTS

Section                                                            Page
I       SUMMARY	"I
II      RECOMMENDATIONS 	  7
III     INTRODUCTION	9
          Purpose and Scope	10
IV      WASTES:  CHARACTERISTICS AND EFFECTS	11
          Dredge Spoils 	 11
          Sewage Sludge 	 16
          Industrial Wastes 	 20
          Radioactive Wastes	21
          Fly Ash and Incinerator Residue	22
          Garbage and Refuse	24
V       TRANSPORT MECHANISMS	27
          Convective Descent	31
          Collapse	34
          Long Term Dispersion	38
          Bottom Transport and  Resuspension  	 43
VI      SOLUTION TECHNIQUES 	 49
          Convective Descent	49
            Waste  Characteristics	55
            Physical Modifications	5?
 VII     BARGE  CHARACTERISTICS  	 75
            Barging Economics  	 81
 VIII    USERS  GUIDE	91
 IX     BIBLIOGRAPHY	95
 X      APPENDICES	103
          Appendix 1	103
          Appendix II	^3

-------
                                FIGURES


No.                                                             Page

 1      Basic Transport Phases 	 30

 2      Convective Descent Terminal Depth
          [after Koh & Fan (2)]	35

 3      Convective Descent Terminal Size
          [after Koh & Fan (2)]	35

 4      Coordinate System for Collapse Phase 	 37

 5      Collapse Size and Time for -y = 0
          [after Koh & Fan (2)]	39

 6      Sediment Concentration Gradient Definition
          Sketch	45

 7      Drag Coefficient for Spheres as Function of
          the Reynolds Number	47

 8      Shields Diagram, as Modified by
          Vanon£ (1964)	47

 9      Penetration Depth and Dilution as Functions
          of the Initial Densimetric Froude Number
          under Linear Density Gradient	51

10      Penetration Depth and Dilution as Functions
          of the Initial Densimetric Froude Number
          under a Strong Pycnocline	52

11      Comparison of Reported Methods of Predicting
          Penetration Depths  	 54

12      Schematic Presentation of  Fate of Material
          when Discharged in  Barge Wake	59

13      Predicted Relative Surface Concentrations for
          Long Term Dispersion Phase	67

14      Maximum Predicted Concentration for Long Term
          Dispersion Stage  	 68
                                   Vll

-------
                                FIGURES
                              (Continued)
No.                                                            Page

15         Predicted X Dimension of Cloud at Level  of
             Maximum Concentration	69

16         Predicted Relative Concentration  Centroid
             Location at Level  of Maximum Concentration .... 70

17         Concentration Distribution with Depth at
             T - 70 hours	73

18         Basic Barge Configurations 	 78

19         Cost per Trip Mile as a Function of Round
             Trip Haul Distance	85

20         Limiting Number of Trips per Year for Preset
             Barge Sizes	85

21         Annual Operating Costs for Round Trip Haul
             Distance of 20 miles	87

22         Annual Operating Costs for Round Trip Haul
             Distance of 50 miles 	 88

23         Annual Operating Costs for Round Trip Haul
             Distance of 100 miles	89

24         Coordinate System and Depth Normalization	116

25         Velocity Profile Normalization	117

26         Vertical Diffusion Coefficient K(y)-
             Normalization Method	 .  .118

27         Depth Normalization	119

-------
                                TABLES

No.                                                             Page
 1       Ocean Dumping:   Types and Amounts,  1968  	   12
 2      Great Lakes Dredging Spoil  Characteristics	15
 3      Typical  Digested Sludge Characteristics  	   17
 4      Typical  Characteristics of Fly  Ash	23
 5      Long Term Diffusion—Example Results	66
 6      Specifications  of Corps of Engineer's
          Hopper Dredges	76
 7      Barge Characteristics 	   80
 8      Reported Costs  of Barging Operations in
          $/Wet Ton	81
 9      Calculated Annual Fixed Costs for Barging
          Operations	83
                                   IX

-------
                              SECTION I


                               SUMMARY

     The ocean dumping policy recommended by the President's Council  on
Environmental  Quality provides that only under conclusive proof of no
damage to the marine environment should dumping be authorized, and even
then regulatory control should be exercised based on standards that
consider:

     1.  The present and future impact on the marine environment,
human health, and amenities.

     2.  Irreversibility of the impact from dumping.

     3.  Volume and concentration of the material involved.

     4.  Location of the disposal site.

     The accomplishment of such a policy clearly shows the need for an
areal model with statistical capabilities.  Most ocean dumping operations
require discharge within specified areas defined by longitude and latitude
and consist of a number of dumps rather than a single site.  Some important
questions regarding the fate of dumped material are:

     1.  What is the ultimate or equilibrium buildup within the water
column or on the bottom?

     2.  What percent will be retained in the dump area?

     3.  What is the distribution of the waste within the dump area?

     4.  Where will the portion not retained go?

     5.  What is the effect on the marine resource?

-------
      It  is not only desirable but mandatory that the answers to these
 similar  questions be known within some limit of statistical confidence
 if such  a policy is to be implemented.

     This report shows that the barged ocean disposal of wastes is an
 established practice which has been justified primarily through economic
 models that exclude consideration of possible damage to the marine re-
 source.  The deleterious effects of past operations are outlined in the
 Introduction, Section III, and serve to reinforce this showing.  The
 objective of this report is not to condemn but to present a discussion
 of the available methods for determining the physical fate of wastes
 discharged through barging operations.

     The fate models presented here are generally good hydrodynamic models
 but are quite deficient in allowing for chemical, physical and biological
 interactions that may occur between the waste, the sea water and its con-
 stituents.  The use of these models assumes a knowledge of a number of
 important environmental  parameters or characteristics which include:

     1.  The magnitude of the diffusion coefficients, their variability
with scale and depth and density.

     2.  Ocean shear, drag and added mass effects.

     3. - Ocean current and density structures.

     A computerized analytical technique developed by Koh and Fan (2) was
adapted to the barge disposal case and is emphasized as one of the better
models available.  The program is available for general use and provides
the greatest degree of input variability of any model considered.

     The physical transport of the waste discharged from a barge was
sequentially described by the following four separate transport phases:

-------
     1 .   Convective  descent

     2.   Collapse

     3.   Long term diffusion

     4.   Bottom transport and resuspension

     The effects of the presence of solids in the wastes was investigated
and it was shown that interactions between settling velocity, concentration,
and diffusion rates do exist but that the analytical  technique is effective
only when these effects can be ignored and the solids assumed to act like
fluid elements.

     The convective descent analysis provides results for the short term
distribution of the waste discharge.  The results include the effects of
shear flows, cloud drag and added mass, entrainment and non-linear density
gradients.  The results of a series of example cases were presented to
identify the effects and trends caused by varying the environmental and
discharge parameters.  The relationship between penetration and dilution
was emphasized and methods of control were discussed.

     The collapse phase was the most hypothetical and was recommended only
for determining the relative effects on the  long term diffusion of the
waste.  The basic assumption made was that at a position of buoyant equi-
librium within the water column, the internal density structure of the
cloud seeking  a position of hydrostatic equilibrium was characterized by
a  dynamic horizontal collapse.  This collaose was assumed to occur without
further dilution and with  the effects of  particle settling  completely
ignored.

      The  example problems  presented  show  that the effects of the  collapse
mechanism appear to  cause  variances  in the long  term diffusion  values of
several  orders of magnitude  over  those when  no collapse  is  allowed.

-------
      These  variations are felt to be too large to be ignored and a clear
 need for research of this phenomenon exists.  Future studies should attempt
 to determine the driving mechanism, the resulting dilution and entrainment
 while developing criteria for the description of interfacial instability
 and the  effects of particle settling on the internal density structure
 of the cloud.

      In  its present form the long term diffusion model is a normalized,
 diraensionless model  that uses the input from the collapse phase to deter-
 mine  the  initial conditions.  The present uncertainty in collapse phase
 theory consequently results in low confidence limits for the long term
 diffusion analysis.   However, if the initial conditions are known then
 this model should give reliable and consistent results.  The ouptut
 tabulates and plots  the  concentration distribution as a function of both
 space  and time and locates the X, Z positions of the concentration centroid
 and the cloud center.   The X, Z size of the cloud can also be determined.
 The predictive information is of the form that, within given limits, will
 be  invaluable in establishing effective monitoring and discharge procedures.
 Its present value is,  however, limited and only provides a method whereby
 the two extremes, either complete collapse or no collapse, can be determined.

     A section was included which if complete would require a separate
 report.  This section  deals with resuspension and subsequent transport
 of materials which settle to the bottom.  Current mathematical models are
 not designed to handle this area of concern and circumvent its existence
 through proper choice  of assumptions.

     Methods are presented which for a set of assumed or measured conditions
will allow settling  and  shear velocities to be determined and compared to
 calculated critical  shear stresses for particles of prescribed size and
 density.   This section is at most cursory in nature and completely ignores

-------
the problems associated with multi-phase wastes  which settle at different
rates and are distributed in graded form over a  varying topography.

     Section VII  of this report investigates some existing operations,
attempting to show the variability in barge size, design and use.   The
barging costs, excluding loading and shore facilities, were examined
through an analysis technique that provides a guide to the decision maker
as to the size and number of barges required for a combination of  haul
distance and annual load.

     Practical use methods are discussed in the  final section and  one
example is given  which traces the steps required to meet an arbitrarily
selected criterion.  The data presented were cited as not being general
in nature and furthermore not descriptive of any particular operation
or geographic area and used only to explore a use method.  It was  shown
that this format  could be used to determine the  dilution, spread,  and
drift of the discharged material and was subsequently applied to the
determination of  the where, when and how of a discharge to preclude vi-
olating established water quality standards.  The sample problem also
evaluated the economics of increasing haul distances to insure such com-
pliance.

-------
                              SECTION II

                           RECOMMENDATIONS

     This report has established the fact that current ocean dumping
operations embody many separate risks to the total integrity of the
marine environment.  The report attempts to isolate only those problems
associated with physical fate prediction from which the following re-
search needs and recommendations are derived.

     Techniques must be developed which will allow the material to be
discharged to be characterized as a multi-phase rather than a single
phase material.  The segregation, distribution and deposition of materials
must be handled in such a way that physical fate can be predicted for
three separate areas (a) the floatable portions that rise to the surface
either permanently or for some finite time after which they resettle
(b) the portion which remains in suspension due to temperature, density
or turbulence levels of the receiving fluid and (c) the portions which
settle and are distributed over the bottom.

     Phenomena such as the collapse phase have been shown to exist but
remain in the realm of the unknown when it comes to the description of
the driving mechanism and the resulting dilution and entrainment rates.
Therefore, this area is considered one deserving of a great deal of ex-
perimental and theoretical research due to its effect on the results of
a long term analyses of the physcial distribution of suspended and settle-
able materials.

     The inclusion of particle settling is a necessity and should be
coupled with an approach that allows for a variable bottom topography
for an adequate description of the subsequent distribution.

     One other important area which was not specifically identified in
the text exists and this is the description of the initial conditions

-------
of the discharge. The question(s) to be answered here is whether
the actual barge discharge dimensions are adequate or whether these
need to be described in terms of an effective size, flow, etc. to
account for possible initial  accelerations,  decelerations or chemical
changes during the free descent of the materials.

     The Environmental  Protection Agency's National Coastal Pollution
Research Program is currently sponsoring research designed to answer
several of these needs, however this current effort is in no way
sufficient to provide the  total information  necessary to satisfactorily
predict the physical  distribution of wastes  and only touches on possible
chemical  and biological effects resulting from discharges to the marine
environment.
                                    8

-------
                             SECTION III

                             INTRODUCTION

     The by-product of life on this planet is waste; nearly every
creature produces a waste product in direct proportion to its popula-
tion growth rate.  Man, the primary exception to this generality,
produces a variety of waste by-products at rates that now exceed
population growth by several orders of magnitude.  Waste disposal
has been primarily terrestrial but present population and urbaniza-
tion trends, fostering competition for available land, have neces-
sitated broad searches for economically feasible, alternative dis-
posal techniques.  Seaside communities and industries have found that
a system of barged ocean disposal of hard-to treat solids and liquids
is an economically feasible solution under existing regulations.

     The President's Council on Environmental Quality, in a recent
report on ocean dumping (1) shows the current responsiblity for control
of disposal to be dispersed among several governmental agencies.  The
jurisdiction of these agencies, generally confined to areas other than
those where the actual disposal occurs, results  in what is termed
uncontrolled disposal.  Conflicts of interest are also pointed out
for agencies possessing both regulatory and operational responsibilities
in the same area.  The Council recommends a national policy that would
ban all unregulated disposal and would  place sole responsibility and
control in the hands of a single agency.  This agency, through the es-
tablishment of discharge regulations and evaluation orocedures, would
"regulate" all ocean disposal operations.  The processes of establishing
regulations, and evaluating permit applications  depend on an under-
standing of limitations in existing methods for  analyzing the fate of
materials dumped from barges.

-------
                          Purpose and Scope

     The primary purpose of this report is  to document currently available
methods and approaches for evaluating the physical  fate and distribution
of wastes discharged to the ocean environment.   Emphasis is on the local
as opposed to global distributions.   This is  only one aspect of a waste's
total effect on the marine environment but  one  that must be understood
prior to setting standards, as well  as evaluating the biological and
chemical effects which, of course, may be different for each type of
waste and each bio-geographical coastal  province.  Where general mechanisms
are involved or can be approximated, they are included.

     This report first classifies the wastes  giving typical characteristics
and reported effects of current operations.  A discussion of the theoretical
transport mechanisms is then presented.   A  Methods of Analysis Section
explores solution techniques and presents typical examples, followed by
a section discussing barging costs.   The report is concluded with recommend-
ations for operation and for research, supplemented with an extensive
bibliography.
                                     10

-------
                              SECTION  IV

                 WASTES:   CHARACTERISTICS AND EFFECTS

     Wastes are  discharged from barges at over 250 U.  S.  coastal  locations
with the majority of sites in nearshore waters less than 100 feet  deep.
Table 1, taken from a Dillingham Corp. report (3) shows the breakdown for
1968 by area and waste type.

     Projections for the year 1980 (1) show expected increases in  bulk
quantity of most wastes to exceed 100 percent of the 1968 values and,
without a major change in policy, most of this is expected to be barged
and dumped at sea.

     For the purposes of this report wastes will be classified as:

     1.  Dredge spoils

     2.  Sewage sludge

     3.  Industrial wastes

     4.  Radioactive wastes

     5.  Fly ash  and incinerator  residue

     6.  Garbage  and refuse

                             Dredge Spoils

     The major  disposal  of wastes to  the  oceans  is  in  the  form  of dredge
 spoils,  a  practice  in  existence  since man  first  found  a need  to maintain
 and improve harbors.

-------
ro
                                                      TABLE 1

                                 OCEAN DUMPING:  TYPES AND AMOUNTS (in tons), 1968*

Dredge spoils
Industrial wastes
Sewage sludge
Construction and demolition debris
Solid waste
Explosives
Total
* Reference (3)
Atlantic
15,808,000
3,013,200
4,477,000
574,000
0
15,200
23,887,400
Gulf
15,300,000
696,000
0
0
0
0
15,996,000
Pacific
7,320,000
981 ,300
0
0
26,000
0
8,327,300
Percent
Total of total
38,428,000
4,690,500
4,477,000
574,000
26,000
15,200
48,210,700**
80
10
9
<1
<1
<1
100
         ** A recent study by the F.D.A. has shown this total  to be closer to  62  million  tons.  The report will be
             published in early 1971.

-------
     Dredging operations are usually carried out in estuaries  where the
primary sediment source is the adjacent watershed and its  drainage system;
however, additional  sediments may be deposited due to:

     1.  Littoral drift

     2.  Incoming tides

     3.  Estuary banks

     4.  Mud flats

     5.  Man made waste discharges

     Estuarine sediments range in size from finely divided colloidal clays,
a fraction of a micron in size, to larger particles of a few centimeters.
These sediments may include variable amounts of organic and inorganic solids.
The sediments are indigenous to the area and reflect the history of sediment
sources.  Development and utilization of lands in the drainage basin have
resulted in additional problems.  These problems are in the form of sediment
contamination from fertilizers, chemicals and pesticides as well as a variety
of industrial wastes which  create a polluted sediment that when dredged or
otherwise disturbed may adversely affect the marine environment.

     To determine the physical fate of the  sediment, the following waste
characteristics  must be specified:

     1.  The size distribution of the solids

     2.  The density distribution

     3.  Chemical flocculation tendencies
                                    13

-------
     Table 2 gives some freshwater dredge  spoil  characteristics  as  reported
by the University of Wisconsin (4) for the Great Lakes.   It should  be
emphasized that these characteristics  may  be  entirely  different  for marine
sediments.  Some characteristics  of marine sediment  are  given in an extensive
survey of the literature on  ocean sedimentation  and  deposition near structures
conducted by Einstein and Weigel  (5).

Reported Effects of Dredging Operations

     A report of a cooperative study between  the Corn  of Engineers  and
FWPCA (6) described a two-year study of  harbor dredging  operations  as they
affect water quality in the  Great Lakes.   The authors  found that the effects
of dredge spoil  dumping in open lake areas remain open to question, but
concluded that in-lake disposal  of heavily polluted  dredgings must  be
considered presumptively undesirable pending  further study.

     The U.  S. Fish and Wildlife  Service (7)  has just  published an  investi-
gation of the effects of dredging operations  in San  Francisco Bay on the
fish and wildlife resource.   The  report indicates that spoiling results
in a temporary reduction in  fish  abundance.  Hopper  dredge spoiling was
found to create an oxygen sag of a temporary  nature  with measured values
as low as 0.1 ppm in bay waters.   The  effects of increased turbidity were
investigated in a series of  laboratory experiments the results of which
indicate that fish exposed to high turbidity  levels  may exhibit a weight
loss and an  increase in the  level of pesticide concentration.

     The effects of dredging on the waters of Chesapeake Bay were studied
by Briggs (8) who found that the spoils were  spread  over an area 5 times
that of the  defined disposal area and  that the total phosphate and nitrogen
in the overlying waters was  increased  50 to 100 times normal values.
Material collected from bottom cores was found to be more than 90 percent
silt and clay in the area to be dredged but was reduced to 75 percent in
the spoils area.
                                    14

-------
                                                     TABLE  2



                                    GREAT  LAKES  DREDGING  SPOIL  CHARACTERISTICS*
en
Location
Buffalo
Calumet
Cleveland
Green Bay
Indiana
Rouge River
Sodus Bay
Toledo
-/ Based on
Average
Percent Density
Solids gm/ml
37
40
44
43
35
43
53
39
30 minute
.0
.7
.9
.0
.2
.7
.1
.0
settl ing
1.27
1.33
1.36
1.37
1.23
1.28
1.51
1.30
Settling
Velocity
ft/hr a/
0
0
0
0
0
0
0
0
.068
.144
.201
.103
.150
.290
.506
.023
Average Percentage by Weight
Gravel Sand Silt Clay
d>2 mm 63u
-------
     Servizi, et al., (9) reported on  the  effects  of a  proposed dredging
operation in Puget Sound on the salmon fishery.  Two types  of sediments
were involved:  a highly organic,  putrifying  pulp  fiber with a high con-
centration of hydrogen sulfide and a natural  silt  of low organic concen-
tration.  The authors concluded that because  of  the highly  toxic nature
of the sediments to the salmon and because various methods  of dispersal
appeared impractical, land disposal of the pulp  fiber sediments would be
necessary to protect the fish  stock.   It was  determined that a 1000 to
1 dilution would otherwise be  necessary to protect and  prevent toxicity
problems to the salmon.

     Brehmer (10)  outlined some of the detrimental effects  of suspended
and sedimenting solids in estuaries and concluded  that turbidity and siltation
reduce the quality of estuarine waters and degrade the  system as a biological
habitat.  O'Connor and Craft (11)  in a study  of  the Mersey  Estuary in England
found fine sediments that settled  as flocules.  An isolated investigation
of individual  flocules revealed an inorganic  core  surrounded by organic
material.  The Mersey has a strong salinity gradient over a tidal cycle  with
a net landward movement of water near  the  bed.  This action was found to
result in a buildup of organic material in the estuary with the presence of
the large amounts  of fine suspended material  creating a buoyant system for
larger solids and  thus fostering and encouraging siltation  of this material.
Brown and Clark (12) found that dredging  for  navigational purposes in Arthur
Kill  near Raritan  Bay resuspended  bottom  sediments having a relatively high
BOD.

                            Sewage Sludge

     The inherent public health hazard and the potential for buildup of
organic solids on the ocean bottom makes  the disposal of sewage sludge
one of the most significant wastes considered.

     The characteristics of the sludge necessary  to predict  its physical
fate are essentially the same as those of dredge  spoils.  The  bulk specific
                                    16

-------
gravity can be controlled to some degree through watering processes  or
through the use of additives such as incinerator residue or fly ash.

     Normally a primary sludge will  have a solid content of between  2
and 5 percent with 70 to 80 percent volatile matter.  A well-digested
sludge will contain 5 percent solids which can be increased upward to
10 percent upon dewatering with a 40-50 percent content of volatile  matter.
Typical values of sludge characteristics for both aerobic and anaerobic
processes are given in Table 3.

     The settling velocity distribution of digested sludges in the sea
water was investigated by Brooks (13) and Orlob (14) for Santa Monica
and San Diego, California, respectively.  Their results indicate a high
proportion of solids have very low settling velocities, with 90 percent
at Santa Monica and 100 percent at San Diego settling at rates less  than
one centimeter per second.

                               TABLE 3

               TYPICAL DIGESTED SLUDGE CHARACTERISTICS*
    Characteristics	Anaerobic	Aerobic	
Total solids                34,000      68,000     	           	
Percent volatile                48          48.6      47             48.9
Specific gravity                 1.011   	     	           	
PH                               7.7         7.2       6.7            5.6
Max. particle in microns     2,000       	     	           	
*References  (15),  (16), and  (17)
                                   17

-------
Reported Effects of Sludge  Disposal  Operations

     Philadelphia's disposal  of  digested  sludge  to  the  ocean  via  barges
was initiated in 1961  and has been discussed frequently in  the  literature
(18, 19, 20).  The dumping  is confined to an area of  one square mile,  ten
miles off the coast near Cape May, New Jersey.   By  1969 (21)  the  annual
volume had reached 115 million gallons.

     The sludge dumping grounds  in the New York  Bight area, located 11 miles
off the coast in waters no  less  than 72 feet,  have  been described as a vast
desert on the ocean bottom  (22).  Another study  of  the  area stated that,
"the bottom of the area of  the mud,  rubble-excavation and sewage  sludge dump
is so badly fouled that changes  in the dump location  would  be of  little help
to the immediate area."  (17).

     Beulow (23) reported on  a bacterial  study made in  the  New York and
Philadephia sludge dumping  grounds.   While it  was  noted that the  coliform
concentration decreased quite rapidly in  the waters receiving the sludge,
high levels of fecal  coliform contamination were found in surf clams, forcing
the closure of affected  areas to further  harvesting.   It was also noted that
there was considerable sludge covering  the bottom.

     Beyer (24) reported  on an investigation of  sludge dumping in the Oslo
Fjord with primary emphasis on the  spreading of  sinking particles.  It was
reported that heavy sludge  particles sank to the bottom rapidly and adhered
to meshes of shrimp trawlers  in  the  area, while  a cloud of polluted water was
visible at the surface for  long  periods  of time.

     There have also been numerous  studies of ocean outfalls for sewage
sludge disposal on the West  Coast which are pertinent  because  of the  reported
effects.  Orlob  (14) reported on the effects of digested sludge discharged
approximately two miles off  the San  Diego shore in 200  feet of water.   The
author's analysis estimates  a sludge accumulation rate  of  one  tenth inch per
year near the outfall with 40 percent of  the solids  settling at such  slow
                                    18

-------
rates that their accumulation within a five mile area could be considered
negligible.  Grease and floatables were identified as potential  problems
but were not defined quantitatively.  Apparently, little consideration
was given the ultimate fate of the solids which are carried away from the
discharge site.

     Brooks (13) studying the sedimentation and dilution of digested sludge
in Santa Monica Bay for Hyperion engineers, concluded that sludge accumulation
rates should average 2-3 inches per year within a 500-foot radius decreasing
to 0.25 inches per year at a two-mile radius from the outfall.  This analysis
assumed a constant current of 0.2 knot with equal frequencies in all directions.
These rates of accumulation were considered unobjectionable based on 1956
standards.  Other studies of this area (25, 26) have reported that the dis-
posal of approximately 4000 tons of solids per day has had no apparent effect
on fish abundance.  Another study (27) however, has  indicated that California's
giant kelp is  being adversely affected by  increases  in sea urchin population
apparently fostered by waste disposal operations along the coast.

      In the Puget Sound area, the disoosal of  digested sludge through out-
falls has  also  prompted studies on  effects and fates.  Brooks,  et al, (28)
studied the outfall design  and  gave  predictions  for  sludge accumulation.
An earlier study by Sylvester  (29)  in 1962 also  discussed  this  problem  and
pointed out the potential  problems  that  could  arise  from:

      1.   The  increased nutrient content  of the water

      2.   Sludge accumulation

      3.   Floatable  materials

      4.   Effects  on the marine  ecology
                                    19

-------
                          Industrial  Wastes

      Industrial wastes vary greatly in both physical  characteristics
and toxicity.  The severe effects that may result from the disposal of
these wastes present the greatest potential hazard to the total marine
environment of any of the wastes discussed.

     For industrial  wastes, the solid concentration,  bulk specific gravity
and solubility in sea water are the minimum characteristics necessary for
the evaluation of the local physical  fate of the material.  Characteristics
of some industrial waste components may be determined from standard
references.

Reported Effects of Industrial Waste Barging Operations

     Hood (3) reporting on the disposal of chlorinated hydrocarbons by
the Shell Oil Company, stated that the organisms endemic to the disposal
area were either killed or seriously imparied immediately upon contact
with the waste.  The area was found to return to near normal in three to
eight hours.  The author concluded that toxic wastes  could be disposed of
beyond the littoral  zone of the sea, resulting in only a slight effect on
organism biomass.  Dispersal was found to be slow, hence to avoid contam-
ination of fishing grounds and areas of upwelling, disposal within the
Gulf of Mexico was recommended only beyond the 2400-foot depth line.

     The toxicity of ferrous sulfate and sulfuric acid was investigated
by the National Lead Company in relation to dumping operations they
carried out in the New York Bight area.  Reported conclusions  (30) claim
no permanent effect on plant, fish, or animal life.  The report mentions
similar studies and conclusions for both containerized and liquid wastes
discharged by U. S. Steel, Champion Paper and Fiber, and National Aniline
Division of Allied Chemical and Dye.
                                   20

-------
                          Radioactive Wastes

     Radioactive wastes  are a potential  hazard to man because  of radiation
received from the immediate environment and by substances  taken into the
body by ingestion, inhalation, or absorption through the skin.  It is
feared wastes may reduce the life span, impair the functioning of parts
of the body, or increase the mutation rate altering the inherited charac-
teristics in future generations (31).  The disposal of radioactive wastes
is generally accomplished either by containment--allowing for natural
radioactive decay, or by dispersal--diluting the radioactivity to permissible
levels, or a combination of the two.  In the past, ocean disposal has re-
quired containment and placement in waters exceeding 6000 feet in depth.
The increased use of radioactive materials by universities, hospitals, and
research facilities has resulted in a corresponding increase in low-level
radioactive wastes.  The AEC licenses commercial firms for the disposal
of these wastes in coastal waters.

     The NAS-NCR  conducted study  (31) considers  the disposal of radioactive
wastes  into  the Atlantic and Gulf coastal waters and reports safe levels
for radioactive wastes, containment  requirements and recommends specific
disposal sites.   This study  also  discusses  the  hazards  associated with  dis-
posal  in relatively  shallow  water,  e.g.,  recovery  by fishermen,  buildup  of
radioactivity  in  marine organisms,  and  washup on beaches  of contaminated
materials.

      Joseph (32)  presented a summary of U.  S. disposal  operations  through
 December 1956  stating that the operations were  considered to  be  under controls
 adequate to preclude hazards in handling and disposal.   He estimated that
 8500  drums of  55-gallon  capacity had been dumped in the Atlantic and over
 13,000 drums in the Pacific, representing a combined total  of nearly 16,000
 curies.

      Waidichuck (33) reported on the containment of radioactive wastes off
 the Canadian Pacific coast and Collins (34) discussed container construction
                                     21

-------
laws, possible discharge  limits,  liquid  effluents,  volatile  reduction,  and
biological  concentrations.

     Koczy (35) discussed the distribution  of  radioactive materials in  the
sea providing the information on  the  variation in dispersion rates.  Isaacs
(36) discussed the magnitude of disposal  of low level  radioactive wastes
into Pacific coastal  waters  as did the Coast and Geodetic Survey (37).
The Pneumo Dynamics Corporation conducted a survey of radioactive waste
disposal  sites (38),  and  an  evaluation of sea  disposal  containers (39).
This problem has been studied in  the  United Kingdom (40, 41) and, no doubt,.
elsewhere.

     Sabo (42) discussed  the river and tidal characteristics of the Savannah
Estuary and described the accumulation of nuclides by organisms.  Pursuhathaman
and Gloyna (43) reported  on  the effects of sedimentation on the transport
of certain radionuclides.  Sheh and Gloyna (44) also reported on this subject
and presented a mathematical model to predict the influence of sediments on
the transport of solubles in open channel flow.

                   Fly Ash and Incinerator Residue

     Fly ash generated by fossil  fuel power stations, represents only  a
minor percent of the total volume of wastes discharged  to the ocean but
is worthy of discussion  for  possible desirable  properties.

     Tenny and  Cole  (45) reported on  the use  of fly  ash as  a  sludge conditioner
and examined  the  subsequent  effect on vacuum  filtration methods  of dewatering.
They found that the  addition of  fly  ash  to  a  sludge  reduced the  volume of
filtrate due  to the  absorptive capacity  of  the  fly ash.  Table  4 summarizes
the typical  characteristics  of the fly  ash  generated from pulverized coal
fired  plants.   The particle  size distribution of the material  appears  to be
describable  as  a log normal  distribution.
                                     22

-------
                                 TABLE 4



                    TYPICAL CHARACTERISTICS OF FLY ASH*
Parameter
Silica (Si02)
Alumina, A1000
2 3
Iron Oxide, Fe^O-
Calcium Oxide, CaO
Sulfur Tri oxide, SO.,
Percent Vol . Matter
Particle Sizes
Bulk Density (computed)
s.g.
Units
mg/1
mg/1

mg/1
mg/1
mg/1

y
PCF

Range
34-48
17-31

2-26.8
1-10
0.2-4
0.37-36.2
0.5-100
70-80
2.1-2.6
* Reference 45
                                       23

-------
     Fly ash possesses an adsorptive capacity which has also been studied
 for its potential for removing soluble COD.   Deb et al., (46) studied the
 effects of adding fly ash to the sludge at a treatment facility and found
 it would adsorb soluble COD.  The effectiveness of the fly ash reached an
 upper limit when the concentrations approached or exceeded 4000 mg/1.
 The effects of the treated sludge when diluted with sea water are not
 known, therefore it is safe only to consider the physical consequences
 resulting from the use of fly ash or incinerator residue as an additive
 to sludges.

     The immediately obvious consequence of such additives is the increase
 in bulk specific gravity that occurs.  This control can be put to good use
 in barging operations as the penetration depth and the dilution of the dis-
charged waste sludges are both sensitive to changes in specific gravity.
These effects may be positive or negative dependent upon the actual en-
vironmental and discharge conditions.

                          Garbage and Refuse

     The disposal of processed refuse and garbage to the ocean is receiving
considerable attention.  This is not a common practice yet, but it is ap-
proaching an economically feasible status relative to land disposal.  One
process now being considered involves sinking of properly compressed bales
and depends on the increased pressure with depth to maintain a density
sufficient to keep the material on the bottom.

     A five-year study conducted jointly by the Harvard University School
of Public Health and Rhode Island's Graduate School of Oceanography (47,
48, 49, 22) investigated waste incineration at sea and the subsequent dis-
 posal  of the non-floating residue.  The results of this study indicated
 little or no toxicity to a series of marine organisms and concluded that
 a depth of 200 feet was sufficient to keep the material from reaching the
 beaches in the test area.  Kinsman (50) however, reports that waves alone
                                    24

-------
have been responsible for sediment movements in water of this depth.   To
date, the burning of garbage and refuse at sea has not proven to be eco-
nomically justifiable and current and future air pollution regulations may
prevent it from becoming technically acceptable.

     Gunnerson (3) notes that in 1968, 26,000 tons of garbage and refuse
were dumped into the Pacific Ocean off San Diego and San Francisco.  San
Diego, however, discontinued this process in November of that year.
                                    25

-------
                              SECTION V

                         TRANSPORT MECHANISMS

     The transport of waste materials dumped into the sea depends, in
general, upon:

     1.   What is introduced - its physical, biological, and chemical
properties.

     2.   Where it is introduced - its position with respect to local
ambient-density and velocity distributions.

     3.   How it is introduced - its residual buoyancy and momentum.

     This paper emphasizes both the immediate mixing and dispersion of
wastes over periods of time that are relatively short when compared to
the circulation times of the oceans as a whole and does not consider
the physical oceanographic processes, whereby wastes can be diluted and
dispersed from one part of the ocean to another.  These are known  to
continually vary with both time and  space  as well as with changes  in
boundary condition.  This  aspect will  be discussed only qualitatively to
aid in  visualizing the applicability and limitations of the analyses
subsequently presented.

      It is  often  assumed or  theorized  (51)  that  although the  ocean is in
continuous  motion  the rates  of motion  and  exchange cover such wide ranges
that  they can  be  separated into  nearshore  horizontal and vertical  exchange,
intermediate  and  deep circulation  exchange and  the exchange associated
with  coastal  and  enclosed  basin  circulation.

      In most coastal  and  open waters  agitation  and turbulence generated
 by wind stresses  on  the  surface  result in  a surface  or mixed  layer

-------
 characterized by a near uniform density gradient.  This layer varies
 between 60 and 1200 feet and is separated from the colder deeper waters
 by a stable layer exhibiting a sharp density gradient - the thermocline
 or pycnocline.  The magnitude of this gradient can vary in both time and
 space and characterizes the relative stability or strength of the layer.

     Wastes introduced into the mixed layer generally will be rapidly
 distributed vertically throughout this layer due to convection, wind-
 stirring or mixing, density differences, and internal currents.  If
 they fail  .to penetrate the pycnocline they will  be transported from the
area of introduction primarily by wind-driven surface currents which, in
general, extend throughout this layer.  The analysis of wastes which do
penetrate  the pycnocline will be influenced, if not controlled, by large-
scale global  currents such as the Gulf Stream and the Kuroshio.  The
average location, magnitude, and direction of these currents has been
documented (52, 53, 54, 55) and in lieu of on-site determinations their
use would  produce approximate but reasonable results.  Estuarine and
nearshore  currents have also been studied, although to a lesser degree,
and typical  values are given by Ippen (56), Orlob (14), and Brooks (11).

     The presence of eddies resulting from turbulence can act to vertically
disperse waste materials in addition to mean current dispersion.  The
rapid increase of density with depth in the thermocline inhibits vertical
transfer,  and eddy diffusion is small compared to that of the mixed surface
layer with its near uniform density gradient.

     There are other localized phenomenon that can influence the exchange
of material  between the surface and sub-surface layers.  This occurs in
areas where:

     1.  The pycnocline is shallow and subject to disturbances, usually
wind generated.
                                    28

-------
     2.   Offshore transport of surface waters results in an upwelling of
colder sub-surface waters.

     3.   Downwelling exists caused by an increase in the density of surface
waters due to evaporation or cooling.

     The first step in a complete analysis of the effects of a waste
discharged to the ocean is to predict its physical fate.  The objective
of the analysis usually dictates a time scale that varies as a function
of the waste material itself.  For example, the time required to reduce
a toxic waste through dilution to a non-toxic concentration may be on the
order of hours, if it is susceptible to chemical and biological destruction
but on the orders of weeks if it is refractory.  The subsequent analyses
utilize a variety of simplifying assumptions and are limited to environmental
conditions variable only with depth and totally exclude biological and
chemical effects.

     The total transport of waste materials  can be  divided  into four
basic transport  phases.  Using the  terminology  of Koh  (2),  these  are:

      1.  Convective  descent

      2.  Collapse

      3.  Long-term  dispersion

      4.  Bottom  transport  and resuspension
 These phases are graphically presented in Figure 1.  It can be noted that
 the first two phases, convective transport and collapse are of short duration,
 when compared to long-term diffusion and, as will be shown later, are
 important in determining the initial conditions for the long term diffusion
 stage.
                                      29

-------
u>
o
                      Long  lerm  (Months]
Collapse
(Minutes)
                                                                             jot torn Transport
Convective
Descent
(Minutes)
                                          FIG. 1.  Basic Transport Phases

-------
                          Convective Descent

     A waste material  dishcarged to the ocean from the surface generally
possesses an initial downward momentum and a density greater than that
of the receiving fluid.  These result in forces that cause the waste to
settle in the form of a "cloud."  As the cloud settles shear stresses are
developed at the interface between the moving cloud and the receiving
fluid.  These stresses result in a dispersion of momentum and in the
creation of turbulent eddies that entrain ambient fluid.

     Entrainment of the less dense ambient fluid reduces the density
differential and tends to slow the descent of the cloud.  The descent
speed is, at the same time, being reduced as solids with settling velocities
greater than the descent speed of the cloud settle out, further  reducing
the cloud's density.  The waste cloud may, in a stably stratified fluid,
eventually reach an equilibrium level where the descent velocity is zero
and the density of  the cloud is in  approximate equilibrium with  the  ambient
fluid.  This zero velocity  state is considered the end of the convective
descent stage.

     To solve this  problem  accurately  the  size and density  distribution
of  the constituent  waste  elements  would  have  to  be  known.   If solids are
considered  to settle  continuously,  the end of  the  convective  descent stage
would  theoretically never be  reached  for more  than  an instant because,
due to continual  particle settling, the  cloud  would  become  positively
buoyant  and begin to  rise.   Many  studies of this  phenomenon have shown
that when both  the  concentration  and  size of the solids  are small  the
waste slurry will tend.to act as  a pure  liquid.   For the wastes commonly
dumped into the ocean this  assumption is currently made for most classes
excepting dredge spoils and industrial wastes having high solid concen-
 trations.

      Morton, Taylor and Turner (57) solved the problem for a point source
 or slug release in  a uniformly stratified fluid assuming:
                                     31

-------
     1.  Wastes to act as pure liquids.

     2.  Velocity and buoyancy in the cloud to be of the same form.

     3.  The mean velocity throughout the cloud to be described by K(Ut)

     4.  Entrainment proportional to the mean velocity and given by

     By writing equations for conservation of volume, momentum and
buoyancy the authors developed the following equations to predict the
final depth of penetration, the final cloud radius and the time of
maximum descent.

     Final depth Yf = bQ  1.682  (aK)"3/l*E~^  	[1 ]

                                     —^"  *~^f                          f~oT
     Final radius bf = bQ  1.632 (aK)  H E     	UJ
     Max. time   tf = 3.14 10s E"
     The solution of these equations requires the constants  of  proportionality
a, and K to be determined experimentally while relying  on  a  linear  density
gradient to determine E defined as follows:
     E = e0b0/ps-p0
Where,
     e  = Density gradient  (3p/3Y)
     b  = Initial cloud radius
      o
     p  = Density of waste
     p  = Ambient fluid density
     g  = Acceleration of gravity
                                    32

-------
     Morton's  experimental  value for the product (otK)  is  0.285 and  upon
substitution allows equations 1  and 2 to be solved.   The  author gives  only
a best fit value for a equal  to  0.093 which by using the  (aK)  relationship
above, allows  equation 3 to be solved.

     Baumgartner, et al., (58) used Morton's work to determine the
ultimate trap  level for a diluted bolus of slurry discharged from a barge
as,

     W=3-8b°E"*	C5]

     Koh and Fan (2) have developed a model designed to predict the
distribution subsequent to a deep underwater nuclear explosion.  This
model, however, can be used for analyzing the convective  descent stage
of a surface discharged waste and, under the previous assumptions, will
produce results identical to those attributed to Morton,  et al.  This
model is general in nature and was derived under somewhat different governing
assumptions.  The entrainment coefficient a was assumed to be proportional
to the vector difference of the mean  cloud and ambient velocities.  The
momentum equation was expanded to account for entrained momentum and the
effects of drag and added mass with  the resulting set of equations programmed
for numerical  computer solution.  The  program allows for arbitrary, horizontal
velocity distributions and can accommodate any depth-dependent density
structure desired.  Koh and Fan  (2),  when  analyzing the deep  radioactive
debris cloud, concluded that  the  inherent  errors in the entrainment coefficient
 (a) more than overshadowed any effects  resulting from the inclusion or drag
or added mass effects.  They  also concluded  that ambient ocean currents were
negligible; however,  this was in  comparison  to relatively high cloud velocities,
initially 600 ft/sec.  When applying this  deep water model to  surface dis-
posals one must assume  that a cloud  of finite dimensions has  been  formed and
can  be described in terms of  known  surface  dimensions and velocities.  It
should also be  noted  that effects of solid  constituents were  not present in
the  radioactive cloud which  requires the  further assumption of wastes that
act  as liquids.
                                    33

-------
     The effects of drag  and  added mass  can  be  removed  by choosing CD = 0
       =1, then the maximum
     m
quiescent receiving body  as:
 and C  =1, then the maximum depth of penetration can be described for a
     m
where, F1 the densimetric Froude number is given by:

     F' - V/tg'b)*5 ............................................. [7]
with g', the reduced gravitational potential defined as:
           Po(o)
     Koh and Fan (2) chose (a) equal  to 1/6 assuming a range of 1/3 to
1/8 and ran solutions for various values of e.  These solutions were
normalized and plotted as functions of the densimetric Froude number and
are presented here as Figures 2 and 3.  Slight differences will result
when comparing these solutions to those using equations 1, 2, and 5
for two basic reasons:  (1) the difference in the defined level of
descent and (2) the variance in methods and coefficients used to account
for entrainment.

                               Collapse

     The second phase of transport, the collapse phase, is the transition
between convective descent and long term dispersion.  The analysis of this
phenomenon assumes that the cloud has come to rest at some equilibrium
position and that a dynamic vertical  collapse characterized by horizontal
spreading occurs.  This collapse is driven primarily by a pressure 'force
and resisted by inertia! and frictional forces.  Many complex actions may
be occurring simultaneously here and an analogy to a density underflow is
useful in attempting to describe them in qualitative terms.  A three-
dimensional form of surge head accompanied by the corresponding reverse
                                    34

-------
    10  _
                            10
10
 FIG. 2 - Convective Descent Terminal Depth [after Koh & Fan  (2)]
o
                                                                            =  0
   FIG.  3  -  Convective  Descent  Terminal  Size  [after  Koh  &  Fan  (2)]




                                    35

-------
 flows should exist, with the circulation increasing the diffusion in the
 area of the spread.  The internal  density structure of the cloud relative
 to the ambient density structure will  exercise control over the magnitude
 of this spreading rate and instability criteria should exist, similar to
 that of the two-dimensional  case,  that will predict breaking interfacial
 waves for some velocity level.  If this occurs, mixing and entrainment
 will further increase.  The  near zero  vertical velocity of the cloud should
 allow for an increase in the number of solid elements that can settle
 out interjecting another action that may foster not only a decrease in the
 driving force but also an upward motion in the cloud itself.

     To date, there is no analysis specifically designed to analyze this
 problem, however, if one realizes  the  shortcomings, the analysis presented
 by Koh and Fan (2) can be applied—providing a feel, at least, for the
 limiting values.  It should  be pointed out here that the method described
was not intended to describe the collapse phase of a waste slurry and the
 shortcomings are to some extent a  function of its extrapolation.

     To apply this analysis  one must assume the following:

     1.   The cloud has an axi symmetric shape at the end of the convective
descent stage.

     2.   The ambient density structure is linear.

     3.   The internal density structure of the cloud is described by an
equation that differs from that for the ambient density through the inclusion
of an internal density distribution term (y).  This equation can be written
 as:
           (y,r)   P0
     4.  No entrainment occurs during the collapse
                                   36

-------
     From the equation for the cloud density structure it can be seen that
when y = 0 the driving force would be the greatest and when y = 1  no spreading
would occur because the cloud and the receiving fluid would exhibit equal
densities at all depths.  The validity of this approach, as pointed out by
the authors, has not been verified and the results presented here are for
theoretical examples designed to explore the interrelationships between
the defined terms.
     A balance of forces in the horizontal directions resulted in the following
equation using the coordinate system shown in Fig. 4.
                          y
                      r2/b
                                                 2 =
         16
                        FIG.  4.
p Tiab:
~T6
         Coordinate System for
         Collapse Phase
                                 d2b
                                      de
                           2a
                                 ()  da
                                           .[9]
      The term on  the left is  the pressure force term equated to respect-
 ively, a local  inertia!  force, a convective inertial force, and a friction
 force.  The equation is  simplified by defining two new terms Kx and  K2  as:
         = 4 CI/TT
      K  =
                                          [10]
                                          [11]
                                    37

-------
     In lieu of measured values for C, and C« the authors reasoned that
they should be near unity and, therefore, attention was focused on the
interrelationship of K, and Kp.  Figure 5 is typical of the results of
this analysis and shows a series of S-shaped curves for a constant value
of K,.  This indicates a pool  acceleration at small values of time which
are independent of K? and are interpreted by the authors to indicate the
domination of the inertial and pressure forces which gradually and contin-
ually decrease until the viscous forces described by K« become dominant
and collapse ceases.  Figure 5 can be used to predict maximum time and
cloud dimensions for the case where y is equal to zero and K, is taken
equal to 0.1.  Examples will be presented in a subsequent section.

     It should be pointed out that no ambient horizontal currents are
assumed to be acting during this phase, a condition not likely to be found
in the shallow waters where barge dumping is common.

                         Long Term Dispersion

     Standardized differential equations are available that describe
dispersion and convection in turbulent flows.  To date, only simplified
solutions  have been used.  These are usually applicable for open ocean
conditions only and presuppose a waste that acts solely as a liquid.

     To apply these solutions to a waste cloud it is again necessary to
assume a neutrally buoyant cloud acted upon by molecular diffusion, eddy
dispersion and mean convective processes.  These simplifying assumptions
result in  a series of differential equations for each phase.

     The basic diffusion equation may be written as:
     If
                                   38

-------
              102i-
CO
         -Q LQ
                                                         t' = t /^
                              FIG,  5,   Collapse Size and Time for y = 0  [After  Koh  & Fan (2)]

-------
where C denotes the concentration of the waste, D the molecular and K  .
                                                                     Xj
the eddy diffusion coefficients.  The last term (m) represents either a
source or a sink.  If the waste is subject to decay, flocculation or
chemical and biological reactions, these processes might be appropriately
estimated by incorporation into a first order reaction function, viz.,
dc/dt = -cK.  The last term of equation 12 would then include (	-cK)
for each process included.  If there is only one source or sink term, it
need not be included in the differential equation, however, the resulting
values'of c(t) the so-called conservative values must be multiplied by
 -kt
e    to obtain the correct liquid concentrations.

     For ocean diffusion problems the molecular diffusion can be shown
to be insignificant when compared to eddy diffusion; therefore, the term
of interest is K . which is a function of the flow field rather than of the
                xj
fluid itself.
     The general  solution, for instantaneous releases of conservative
wastes, can be described by the following equations:
 ,    z t) =          M	       r(x-ut)2+ (y-vt)2 + U
              ( A \ /2 f (/ (/ t/ \ "^-l- /2       H" Ix t     *T ix t      T"ixw
              1 ^;  ^ x yxzj             X        y
where M is the mass  of waste released.

     If the source is fixed and  continuous and is being discharged into
a uniform flow field the concentration  for any time at a point can be described
as follows:
                                                                    C14]
where q is a measure of the volumetric discharge and has units of [L /T],
and c  is the initial waste concentration.  For the case given here the
current is assumed to be in the X direction with the eddy diffusion in that
direction assumed negligible.
                                    40

-------
     Orlob (11)  used the two-dimensional  form of equation 13 to predict
sludge accumulation rates adjacent to an  ocean outfall.  He redefined the
eddy diffusion terms K - to represent the greater effects of current velocity.
Glover (59, 60)  also used this relationship to analyze the dispersion of
solid and suspended materials in open channel situations.

     Simplified versions of these equations have been used by the Atomic
Energy Commission to analyze the diffusion of radioactivity subsequent to
the sudden rupturing of cubical cannisters residing on the ocean bottom.
The assumed conditions reported (61) provided for an instantaneous rupture
in the absence of currents.  Equation 13 under these assumptions reduces to:

     c = M/4UK t)3/2  Where M E cQV  	[15]

     The assumptions made should be  reviewed to reinforce the limitations
of these solutions.  A constant flow was assumed removing shear effects thus
limiting the use of these equations  near boundaries.  The turbulent diffusion
coefficients were  assumed constant,  an assumption that will subsequently be
shown not to be valid for many ocean situations.  These  solutions cannot be
expected to predict the  transport of any wastes with  high solid concentrations
due to the assumed zero  settling velocities.   It should  also  be pointed out
that in the case of surface  dumping  of waste materials the  initial conditions
at the beginning of this phase are  described  by the waste cloud at the end
of the collapse  phase.   These  conditions,  at  present, cannot  accurately be
predicted which  further  effect the  reliability of derived results.

      Solutions  recently  presented by Carter  and Okubo (62), Okubo  (63, 64)
and  Okubo  and Carweit (65)  have  included  shear effects for  both continuous
and  slug  releases. These  solutions are  applicable  only  when  the  settling
velocity  can  be assumed zero.   The  eddy  diffusivities have  also been
considered constant with respect to both  time and  space  based on  a division
of the  turbulence  into  large and small  scale eddies.   The  small scale eddies
                                     41

-------
were felt to exhibit small time and length scales relative to those
of observations, thus allowing the use of mixing length theory to describe
this diffusion process. The large scale eddies were assumed to create an
inhomogeneity in the flow that could be described by defining the mean
velocities as:
     Vy = Vz = 0 	[17]

where, ft ,  fl  denote constant horizontal and vertical  shears.  This
        A   £.
assumes a mean velocity along the x axis with the z axis vertical and
leads to a  diffusion equation for their model of:
     Computer programmed solutions to these equations have been reported
(63) which provide  as functions  of time:

     1.   Families  of isoconcentration surfaces

     2.   Dimensions of the contaminated region

     3.   Volume  and quantity of  material

     Koh and Fan (2) have presented an even more general  solution allowing:

     1.   Turbulent  eddy diffusion coefficients variant with both scale
         and or  depth.

     2.   Current generated shear in both vertical  planes.

     3.   Non-linear density gradients, when applicable.
                                   42

-------
     The environmental  conditions  are somewhat limited as  they are  independent
of the horizontal  coordinates.   This limitation is  minor in light of the  un-
certainty that exists around the proper choices for coefficients  to describe
entrapment, diffusion, drag, and added mass.   The  analysis does  however
provide the gross  characteristics of the "cloud" which include:

     1.  The total material distributed over each horizontal plane.

     2.  The location of the centroids.

     3.  The standard deviations of  the centroids.

     4.  Estimated characteristic concentrations.

      5.  Horizontal  cloud  dimensions.

                   Bottom Transport  and Resuspension

      This  final transport  phase,  contrary  to  the basic assumption  made
 earlier, assumes  that  the  solid constituents  of the  waste slurry do indeed
 settle and reach  the bottom.  This  could be accomplished  entirely  during
 the convective descent stage if the water depth were less than the predicted
 or theoretical total penetration depth.  The resulting distribution- of the
 solid constituents for this case would become a function of the  residual
 momentum of the cloud as well  as the existing density disparity.  When
 the residual momentum is high, a dynamic turbulent rebound effect might
 be expected, but for near zero momentum the density disparity would cause
 a spread similar in nature  to that  of  the collapse phase.  Two questions
 are raised  regarding the  solids once  they reach the bottom:   [1] will
 movement occur?  and [2] if  movement occurs, what  form will occur - bed  load
 or  resuspension?

       In general, it is  accepted  that  motion will  be  initiated in a flow
  field when the shear  stress on  the particle  creates  a  lift force  in  excess
                                     43

-------
of the submerged weight of the particle.  The direction of the initial
particle movement will be nearly perpendicular to the plane of the applied
shear stress and for a horizontal bed will be near vertical.  This force
will lift the particle off the boundary and it will resettle, subject to
currents and eddies, either to a position in the flow where the resultant
lift on the particle just equals its settling velocity (suspended transport),
or when the settling velocity exceeds the lift force, to another position
on the boundary (bed load transport).  Suspended transport is most common
when the current shear is nearly constant with bed load motion resulting
when an additional shear resulting from turbulent eddies is superimposed.

     A method is available whereby one can predict what form of transport
will occur.  This method assumes steady uniform flow, a logarithmic velocity
profile and a linear shear stress distribution with depth.  An equation
which relates the settling velocity of the particle, the diffusivity of
the system and the sediment concentration c is given by:


     H  •  ^
where K = pUJcz under the above assumption.
     This equation can be rearranged and integrated over a  region of interest
with the following results:
                                                                   [201
     /•••l   \ -t I  II !/• 	'	*»."-*'J
Defining r   as (q) we can relate the type transport to  the magnitude of
this parameter.  By plotting the In  (c/ „) against the  In  (z/,)  it can  be
                                       ca                    a
seen that 1/q is the slope of the line and is descriptive  of the concentration
gradient.
                                    44

-------
            ln(f)
               FIG. 6.
                                  1/q
                   In(f-)
                       a
                   Sediment  Concentration
                   Gradient  Definition  Sketch
     When the magnitude of q is less than unity the effects of the shear
stresses  are always greater than the gravity effects keeping the material
in suspension with what is termed wash load transport resulting.  Studies
have shown that when q is greater than unity but less than three suspended
load transport should be expected with bed load transport predominant when
q exceeds three.

     To use this relationship the settling velocity of the particle must
be determined and related to the existing shear velocity U*.  The shear
velocity is governed by the following relationship:
u =
                                                                    [21]
where Z equals the point of interest in the flow--measured from the bottom--,
Zn is the depth where the assumed logarithmic profile goes to zero and k is
Von Karman's constant equal to 0.4.  Solving for U* would require an iteration
process if the profile is unknown.  If shear stress measurements are available
U* can be determined from the following relationship:
         Tb  =
                                                              ,[22]
                                   45

-------
     The settling velocity of a particle can also be determined by an
iteration process using the following equations and Figure 7 which relates
the coefficient of drag to Reynolds number defined using the settling
velocity of the particle:

        _  4  PS  - p gdA	[23]
    ws  '  T L  p  CD   J

     After determining both U* and (w)  a direct substitution into the
relationship for (q)  will  give a prediction of the type transport that
should be expected.

     Another approach can  be used which compares the critical shear stress
to the calculated or  measured stress on the boundary.  The critical shear
stress is that stress that will just initiate motion for a particle of
given diameter (d).   This  stress can be determined using Fig. 8 and the
following equations where  the critical shear stress TC is given by:

    T.  =  f(R*,n) = T*(p -l)gd	[24]
      C                 *>

with n given by:
                   n = g[(0.1)(ps-l)gd]
                                 P
For a given U*,n can be determined and T* can be read directly from the
graph and a simple calculation using Eq. 24 will give the critical shear
stress for that combination of shear velocity and particle diameter.
Comparing T  to T.  given by Eq. 22 will reveal if motion should be expected.

     These equations are admittedly simplified, and do not represent a
complete literature review but within the accuracy of any of the other
                                  46

-------
      io ;
10
         8

        lor
  FIG.  7.
                   Reynolds Number R = —
                  Drag Coefficient for Spheres  as
                  Function of  the Reynolds Number
in
O)
s_
 1.0


0.6

0.4
S-  Q.
ro  i
O)   to O.2
^Z  Q.
oo —
(O
o
 O.I


.06

.04



.02
       .01
d
v
0.1
                                                4661002    468 1000
        O.I  0.2  0.4 0.6 I.O   2    4  6  8 10    2    4  6 8100   2   468 1000
                  Boundary  Reynolds Number
                                                    V
          FIG.  8.  Shields  Diagram, as  Modified by Vanon£ (1964)

                               47

-------
transport phases allows the prediction of motion as well  as type transport.
These predictions are subject to the several  assumptions  made, namely:

     1.  A steady uniform flow

     2.  A logarithmic velocity profile in lieu of actual bottom shear
stress measurements.

     3.  A horizontal  bed of uniform roughness (i.e., no  dunes, growth,
outcroppings, etc.)

     Partheniades (66) presents a State-of-the-Art summary of the behavior
of fine sediments in  estuaries and documents  a number of  research needs in
this field.   The A.S.C.E. Task Force on Sedimentation (67) also presents
a summary of applicable approaches and presents relationships which account
for non-uniform turbulence.

     Other approaches can be found in the work of Schmidt and O'Brien (68, 69),
Souther!in (70), Brooks (13), Shields (71), While (72), Vanoni (73), and
Anderson  (74).
                                    48

-------
                              SECTION VI

                         SOLUTION TECHNIQUES

     There is no single analytical method currently available that will
completely and accurately predict the dispersion and dilution of a waste
material  discharged into the marine environment.  The general nature of
the approach used by Koh and Fan  (2) does, however, offer a good means of
analyzing these effects for surface discharges under a variety of en-
vironmental conditions.  This solution technique has been programed for
general  use and the following section will explore hypothetical case
solutions for each of the three separately defined transport phases.

                          Convective Descent

     The convective descent phase allows the initial short term distribu-
tion of the waste to be explored  for any combination of density and
velocity profiles.  Two separate  cases will be  investigated; one with
a linear density gradient and the other  representative of a  strong
pycnocline.  For these conditions the penetration depth  (Y.r) and the
dilution (DILN) will be investigated under  the  existence of  a  two-dimen-
sional shear flow.  Discussions will be  two-fold  in  nature;  describing
the procedure to be followed to use the  program,  while pointing out
critical parameters and some of the errors  inherent  in simplifying assump-
tions commonly made.

     The input requirements are listed  and  discussed in  Appendix  I
complete with a sample problem.   The convective descent  solution  gives
incremental  penetration depths, dilutions,  X  Y  Z positions  of  the cloud
center all  as functions of descent  time.  The analysis is  terminated  either
when the cloud has  reached the  bottom or where  the  cloud becomes  neutrally
buoyant  and its descent is stopped.  This  latter case  is characterized  by
a  reversal  in  the direction of  the  cloud velocity with the  analysis  ter-
minated  after  one complete cycle.

-------
Discussion of Example Problem Results

     A classic example is  revealed  in  Fig.  9 where  the dilution is
high for high values  of the  initial  densimetric Froude number with  a
characteristic curvilinear decrease  with increases  in the penetration
depth.  For example,  an initial  densimetric Froude  number of 0.09 would
tend to concentrate  the material  at  a  depth of 185  feet with a dilution
of 425.  Increasing  the Froude number  one order of  magnitude would  de-
crease the descent to 95 feet but would increase the dilution approximately
eight times for a total dilution of  3400.  The presence of the strong
pycnocline, Fig.  10,  at Froude numbers between 0.4  and 2.0 appears  to
significantly reduce  the penetration and dilution with the waste material
retained in the oycnocline.   This figure also indicates two methods by
which the dilution can be increased.  An increase in the Froude number
between 0 and 0.2 results in higher dilution and it appears that a  decrease
in large Froude numbers to 0.2 will  allow the waste cloud to break  through
the pycnocline, after which the penetration and dilution rapidly increase.

     These results are not descriptive of the general case, but were
designed to represent a typical operational evaluation.  A series of computer
runs were made where the discharge radius was varied while maintaining a
flow of 100 cubic feet per second (cfs).  This limiting criterion in effect
establishes a unique relationship between the radius of the discharge and
the initial densimetric Froude number.  Two sets of environmental parameters
were employed; the first descriptive of the linear density profile, and
second, a strong pycnocline.  The penetration depth and the resulting di-
lution were then compared against the  initial densimetric Froude number.

     Figure 9 for the  linear gradient  exhibits low penetration and  high
dilution for  Froude numbers of order 10, which, for this example, correspond
to a small diameter discharge.  This result is predictable  if  one considers
that the entrainment  is high and, when  applied to a small waste  volume,
dilution proceeds rapidly and buoyant  equilibrium is  attained  at a  shallow
depth.   The  same argument applied at the other end of  the scale  where  the
discharge  velocity is  low,  and the  radius  is  large, would predict a biower
rate  of dilution and  a greater penetration, again as  shown  by  Figure  9.

                                    50

-------
O.
QJ
Q
o

tQ

tu


Q.
     -60-
    -120
    -180-
    -240
    -300
    -360
    -420
    .480
    -500
                                                                                       DILN

                                                                                       Penetration
                                                                                                     600
                                                                                                    12CO
                                                                                                    1800
                                                                                                    2400
                                                                                                    3000
                                                                                                    3600
                                                                                                    4200
             J	I	I  I  I I I I
                                            '    I   I  I  I  I I  1
I    1   I  I  I  I
                                     Initial  Densimetric Froude Number F'
                                                                                         10
4800

5000
          FIG. 9.
                  Penetration  Depth  and  Dilution  as  Functions  of the Initial  Densimetric Froude
                  Number under  Linear  Density  Gradient  (E  =  1.17xlO"Vft)

-------
cri

r°
   -C
   -p
   O-
   
-------
     When the density profile is complicated through the inclusion of
a pycnocline, the results take on a different form.  For Froude numbers
of order 10, the same trend seen for the linear case exists (Fig. 10)
with the cloud's descent terminated above the level of the pycnocline
and its existence is really of no importance.  At the low end of the
Froude number scale (F 0.1) entrainment is also reduced and in essence
the descent is governed by the linear gradient that exists below the
pycnocline.  The major difference occurs in the intermediate Froude
number range, where entrainment and waste volume are sufficiently
interrelated for the cloud to obtain buoyant equilibrium in the pycno-
cline, but, driven by a high momentum, is carried beyond this position,
with both the velocity and the buoyancy becoming positive.  The sub-
sequent dilution and ascent are relatively rapid, with the oscillations
damped almost immediately.

     The sudden  shift or increase in dilution that  occurs  in Figure  10
for  Froude  numbers of approximately 0.2 results when a cloud has  in-
sufficient  positive  buoyancy  to  reenter the  pycnocline and remains
subject  to  the  nearly uniform-gradient  below.  A  long period oscillation
 is  developed  which  results  in the  rapid increase  of the  time required
 for this cloud  to  reach  a  stable position  (from  five minutes to  twenty
minutes) with an obvious  increase  in  the  dilution.

      Comparing  the  results of the  two examples  reveals  that  using a
 linear assumption where, in fact,  a pycnocline  exists would  result
 in a general  overestimage in both the depth of penetration and the di-
 lution.  This conclusion is valid only under the previously  described
 conditions; however, generalized plots can be made and  presented in a
 manner similar to that shown in Figures 2 and 3.

      The penetration depths as predicted by the various methods dis-
 cussed in  the  previous chapters are compared in Fig. 11.  Curves 1 and 2
                                    53

-------
              o.
C/1
-fa.
        Q.
        OJ
        Q
        O)
        C
        
-------
utilize the same basic equation, however in case 2 the relative effects
of introducing cloud drag and current generated shear are responsible for
the net differences.  The penetration depths for all methods, over the
full range of Froude numbers tested, varied a maximum of only fifty per-
cent compared to a variance in dilution which at times exceeded 800 per-
cent.  For this reason alone it would seem that dilution, as well as
penetration should be given consideration in all predictive convective
descent models.  The predictive nature of Morton's approach is such that,
on the basis of penetration alone, it could be erroneously interpreted
as a conservative model over the full range of environmental conditions
that might exist.

     The  penetration depth  (Yf) and  the dilution  (DILN)  have been shown
to  be  conditionally related to  the initial densimetric Froude  number.
Changes in the  magnitude of the Froude number  can  be  effected  by  altering
the  magnitude  of  the velocity,  the diameter, or  the method of  the dis-
charge or by modifying  physical waste  and  discharge characteristics.   With
the  exception  of  the waste  specific  gravity, physical modification  will
directly  affect only  the  initial  dispersion  that occurs  during the  con-
vective descent stage.   Once  a cloud attains buoyant  equilibrium within
 the water column  or reaches bottom,  the  subsequent dispersion  is entirely
 dependent on its  relationship to  the natural  environment conditions that
 exist. This in essence means that any control over the  long term dispersion
 must be  exercised during the discharge process in such  a way as to alter
 its convective descent.

 Waste Characteristics:

 The waste characteristics are  infinitely variable; however, in analyzing
 the initial distribution, only the  physical characteristics need be con-
 sidered.  The  bulk specific gravity of a waste  slurry is one  characteristic
 that, within  physical  limitations,  can be controlled.   Changes in  the
 specific gravity of a  waste slurry  or sludge, whether through the  use of
 a  thickening  process  or through  the use of additives, can affect both the
                                     55

-------
 resulting dilution and penetration.  Such changes may result in changes in
 either the concentration, density or size of the solid constituents or any
 combination of the three.  The effects of solids on diffusion in fully de-
 veloped turbulent liquid flows has been investigated by Katta and Hanratti
 (75), Rouse (76), Vanoni (77), Householder and Goldschmidt (78-79), Singamsetti
 (80), McNown and Lin (81), and by Ahmadi and Goldschmidt (82).  From these
 studies it is generally concluded that suspensions with grain sizes less
 than 60 microns can be assumed to act as pure liquids.  Inter-relating effects
 are shown between such parameters as settling velocity and particle concen-
 tration with an increase in particle concentration increasing the diffusion
while causing a decrease in the settling velocities of the particles.

     From data presented by Ahmadi and Goldschmidt (82) the turbulent Schmidt
number can be calculated independently of concentration.  Such calculations
for typical  values for dredge spoils and sewage sludges reveal ratios of mass
to momentum transport that vary between 1.0 and 1.1, indicating that the
motion of the solids is not significantly different from that of the liquid
phase.  Such analyses provide the justification needed for a pure liquid
assumption in waste dispersion studies, and, when made, reduce to one the
number of physical characteristics of the waste that must be known; namely
the specific gravity.

     The approximate specific gravity of waste sludges or slurries can be
determined from a knowledge of the percentage of solids and volatile matter
or the percent moisture alone when the solid concentration is high.  Equations
 25 and 26 after Fair and Geyer (83) are commonly used for this purpose.
            100SQ S,,
       _        b  W
         _
       ~ P Ss + (100-P) Sw

       = 100 Sf Sv/100 Sv + Pv(Sf-Sv) .............................. [26]
                                    56

-------
where:
     S  = Specific gravity of waste

     $<-= Specific gravity of waste solids
      O

     Su= Specific gravity of water in waste mixture

     Sf= Specific gravity of fixed solids  (2.5)

     S = Specific gravity of volatile solids  (1.0)

     P = Percent moisture by wt.

     P = Percent volatile matter  in  sludge

 Physical modifications:
     Several  other  controls  are available and are of a ohysical  nature.
 Included among  these  are the depth,  size, and orientation of the discharge
 outlet,  as  well  as  the barge speed and direction, which can be used to in-
 fluence  and control  the final  distribution to some degree.  However, the
 optimum discharge  method must be seoarately determined for each waste and
 should include  consideration of possible  biological and chemical effects.
 The general influence of changes in density, discharge radius and velocity
 on penetration  and dilution can be evaluated in  terms of the changes in the
 densimetric Froude number.  The Froude number can be  shown to vary as bQ,
 the initial cloud radius, to the minus two-thirds power, and directly with
 the discharge velocity.  These relationships are not  generally  independent
 and a change in one of  the  parameters without a  compensating change  in  the
 discharge  rate will necessitate  the  use  of a continuity  relationship.

       Pump  discharges  into the  wakes  of  propeller streams  of  barges cannot
 be analyzed by  the  Koh  technique and must be  handled  separately,  and
 in such a  manner  that the  increased turbulence  created  by  the  passage
 of the  barge is  accounted  for.  These discharges are  usually employed
                                      57

-------
when the waste is toxic, taking advantage of the increased turbulence
to maximize the rate of dilution.

     If this method of discharge is considered for cases where stable
density stratification occurs, the liquid slurry may not settle directly
to the bottom upon discharge.   This action can then be described in terms
of three separate flow regimes as  shown in Figure 12.

     The longitudinal dimensions of Zone 1 can be assumed to coincide
with that distance behind a barge  (X1) where a fully turbulent wake is
established.  Schlichting (85) states that this is reached, for a cylin-
drical object, when the following  relationship is realized:

     X '  = 50 CQW .................................................. [27]

where W = barge width and
      CD= drag coefficient

     Redfield (86), Ketchum and Ford (87) and Hood (30) have analyzed the
discharge of wastes into the wakes of moving barges where mixing was ob-
served to be instantaneous in the vertical direction.  Thus, by neglecting
vertical dispersion, and transverse dispersion when the barge travel and
the ambient currents are along the same axis, the problem becomes one di-
mensional.  With such a simplification the solution in the established flow
zone becomes one of two variables--the turbulent diffusivity and the dis-
charge rate.

     Equation 12, the basic diffusion equation, applied to the dispersion
of a waste  in a barge wake has been solved by Ketchum and Ford (87) and
Pearson, Storrs and Selleck (88) with the respective solutions describing
the median  and average concentrations per unit cross section along  the
centerline  of the wake.  Both solution equations are identical:
              cAq
ci  =
              o
                                                                    [28]
                                    58

-------
             Zone
              III
on
                                                              mixed surface
                                                                 layer
                  FIG.  12.   Schematic Presentation  of Fate  of  Material  when  Discharged  in
                            Barge Wake

-------
where A. = 0.493 describes the median concentration and A. = 0.282 the
average concentration.  The other terms in the equation are:

                                              o
     q = volumetric discharge per unit time [L /T]

     h = mixing depth [L]

     U, = barge speed [L/T]

                                                               2
     K = "turbulent diffusion coefficient for combined system [L /T]
      X

     T = the length of time since the barge has passed the point of
         interest [T]
     c = initial concentration of the discharge

     Turbulent eddy diffusion coefficients in wakes have been shown,
                                                             2
through field determinations (87, 86), to vary between 1.0 ft /sec and
       2
30.0 ft /sec and should be expected to fluctuate.
     Schlichting (85) equates the wake diffusion coefficient to the
wake centerline velocity and the defined half width of the barge as
follows:

     K  = 0.0235 Ur  (X'CnW)1'5	[29]
      A           Li      U

The turbulence is known to decrease over the length of the wake.  This
relationship is contained in the above equation since the velocity de-
creases as the length of the wake to the minus two-thirds power.  The
wake turbulence in a practical  sense, has a lower limit equal to the
natural turbulence of the ambient environment.  Pearson et^ al_., (88)
used the  water depth as the length scale (L) and determined a constant
average diffusion coefficient as defined by the following relationship:
               V
                3 	[30]
                                   60

-------
     The  mixing  depth  (h),  the last unknown, is a difficult parameter
to determine.  Omission  of  this term from Eq.  28 would result in a de-
scription of  the concentration per unit cross  section of a water column
of unknown depth.   This  form of the equation was the one used by Ketchum
and Ford  (87)  in analyzing  the concentration of an iron waste discharged
to the  wake of a barge.

     Results  of  full  scale  experiments where a waste was discharged into
the propeller stream  of  a ship are given in a  recent report by Abraham
and Eysink (84)  which  discusses the applicability of jet theory to this
method  of discharge and  concludes that the minimum dilution can be determined
using jet theory calculations.

     When the results  of the convective descent stage analysis indicate
a situation where the cloud comes to rest at some equilibrium position in
the water column, both collapse and long term dispersion must be considered.
These are sequential  analyses with the output of the collapse phase serving
as the  initial conditions for long term dispersion.  The only method for
the analysis  of  the collapse phase appears to be the one presented by Koh
and Fan  (2) which is at present hypothetical.

     The collapsed dimensions of  the cloud can  be determined for the case
where the constants C-, and C2 are assumed near  unity and  K-j is assigned a
value of 0.10.  The initial shape of the waste  cloud must  be assumed an
axisymmetrical ellipsoid with  the major  horizontal  radius  (bQ) equal to the
predicted cloud radius at the  end of the convective descent stage and the
minor radius  (a ) defined as  (bo/2).   The approach  is  limited  by a requirement
that the density gradient be  linear over the  depth  of  the  cloud which pre-
cludes its use  in cases where  the cloud  collapse  transcends a  pycnocline.
Conceding  this  deficiency and  choosing a proper set of initial  conditions,
the  cloud  dimensions  can be  determined for  the case of a  complete collapse.
                                    61

-------
     For the purpose of isolating the effects  of a collapse phase  on the
long term dispersion of a waste,  fixed value results for a convective descent
stage definitive of a set of intial  discharge  conditions were chosen where:

     Bf = radius at end of convective descent  state = 66 feet

     Yf = 159 feet

     E  = 1.17 ft x 10~5/ft.
     For these values the initial  collapse phase dimensions are b  - 66 ft,
and a minor radius a  = 33 feet.   The collapsed dimensions can be determined
directly from Fig. 5 and for this  example were taken as:
                              b  7
     bf = 264 feet   af - aQ (^-)   - 2.06 feet

The time for this collapse to reach completion was determined as 568 sec.

     The question, now, is:  What is the effect of such a collapse on the
long term dispersion solutions as predicted by this analytical technique?

     It was decided to explore these effects for two sets of environmental
conditions:  Case I investigates  the effects of cloud collapse under en-
vironmental conditions consistent with a linear density profile and Case
II investigates the same cloud effects for parameters descriptive of a
strong pycnocline in which vertical transport is suppressed.

     The use of this analytical technique requires that the initial distri-
bution of the waste be assumed axisymmetric about the Y axis and confined
in an ellipsoidal region with a horizontal distribution of material defined
by:

     C(y1fO) = (1 -sV) **	[32]
                                   62

-------
where 6 is  a dimensionless ratio of the depth of the cloud to its half
thickness.   The horizontal diffusion coefficients are assumed to follow a
4/3 law relationship and have been defined in terms of the geometric mean
of the standard deviations along the principal axes as:
Sc - Kz = AK V - °xzl
                                                                   [33]
The standard deviations of the distributions QX and az in the above
equation are taken as equal to 1/4 the size of the cloud as shown in
Appendix II.  The coefficient A will be taken to be constant over the
depth of the cloud but the solution does permit this to be varied if
desired.  This coefficient can be determined by considering the 4/3 law
governing horizontal diffusion where:

      K  .K   - A l>            	[34]
      Kx'Kz  " \L    	

      A,   is a constant  dissipation  parameter  and  (L)  is  a  length  scale.
 For the  problem  at  hand  (L)  is  taken equal  to  the  length  of the  cloud
 which is related to the  standard deviation  of  the  distribution by:

      ,     ^                          	[35]
      Lc '  40x 	

      Substituting we have
      Where A = 6.34 A
                      U)
 The choice of a value for A(fi/) was  taken  for  this  example  as  1.5x10"
 ft2/3/sec which makes A equal  to  0.001  ft'/Vsec  the  quantity which will
 be used for this  example.   The vertical diffusion  coefficient KyS  as  used
                                      63

-------
in this analysis,  must  be  determined  and  related  to  the  existing  density
structure.  These  coefficients  are, in  general, smaller  than  those  for
horizontal diffusion  with  the magnitude decreasing with  depth to  near
molecular values  in  pycnoclines.

     There is  no  apparent  universal law or  value  for K   but a functional
relationship can  be  shown  between  vertical  transport and the  Richardson
number (89)  which  implies  a  dependence  on one  or  all  of  the following
parameters:  density  gradient,  current  generated  shear and the type of
flow.  Reported values  for K range from  1.075 x  10" to 3.215 x  10"
ft /sec with exoected variances from  flow or density changes.  Koh  and
Fan (2) have presented  estimated coefficient ranges  for  several  areas of
interest:
                                        _2              1          2
K 	Mixed  layer   2.2x10    to  2.2 x  10    	ft  /sec
                                         3              5          2
K 	Pycnocline    1.1 x 10"  to  1.1 x  10"   	ft  /sec
                                        _2              4          2
K 	Deep layer    1.1 x 10    to  1.1 x  10"   	ft  /sec
These are suggested values based on limited past information.  For the
example problem being considered here a vertical diffusion coefficient
equal to 0.1  ft /sec will  be used for the mixed layer, 1.0 x 10   for
a strong pycnocline and 1.0 x 10"  for deep layers.
     The output from this analysis describes the concentration distribution
in terms of the initial  concentration at the beginning of the phase and a
normalized elaosed time.  The X, Y size of the contaminated area, the X,
Y, Z position of the concentration centroids and the distribution of con-
centration with depth can also be determined.  These allow predictions for
both the concentration and the environmental exoosure time to be made
thus aiding in the development of field sampling and monitoring procedures
and techniques.  The results subsequently oresented assume an initial
concentration at time zero equal to unity and the cloud at a real depth of
                                    64

-------
159 feet assigned an X, Y, Z position of (0,0,0).  The analyses is ter-
minated when a diffusion time equal to that required for the material
to diffuse to surface under a constant vertical diffusion coefficient is
reached.

     The basic problem being considered here is limited to only the collapse
and long-term dispersion phases.   For an arbitrarily selected set of initial
conditions four individual sets of solutions are being sought.  These solu-
tions compare the results for a cloud that exhibits no collapse to one that
completely collapses under two sets of environmental conditions.  These
effects can best be compared through an examination of the magnitude of
various parameters such as the maximum surface concentration, the maximum
concentration or the location of the cloud or concentration centroid.  These
same parameters can also be compared on a time base.  Such comparisons have
been made for the example considered with the real value results  given in
Table 5 and Figures 13 through 16.  By using both  the figures and the table
trends  should be identifiable and  observed deviations relatable to either
a  cloud collapse or a change  in environmental conditions.  The effects of
a  cloud collapse does represent the  two extremes  possible  and can be con-
sidered a measure of the  confidence  one can  afford any  individual prediction.

Discussion  of Example Problem Results

     Figure 14  relates  the  maximum concentration to exposure  time.   Inclu-
sion of either  a collapse mechanism or a  pycnocline or  both  acts  to  de-
crease  the  rate at  which  this dilution occurs  thus increasing the exposure
 time for  any  given  concentration.   For example,  if the  time  required for
 the maximum concentration to reach 1/1000th of the concentration  at  t=0
 the graph can  be used  to  predict  30 hours for the uncollapsed linear case
 and 100 hours  for  the  collapsed cloud in  the pycnocline.

     The  surface concentration-exposure time relation is given in Figure
 13 and reveals  that a  linear uncollapsed cloud reaches  its maximum value
 in 8.5 hours.   If this cloud is collapsed and subjected to the same  analysis
                                    65

-------
                                              TABLE 5


                                      Example Problem Results
Parameter
Cs (Max)
T (a Cg max
Lv @ C max
A 5
CEx 9 Cs max
Snax at T=7° hrs
L @ C
x max 70
CCx @ Cmax 70
C$ P T=70
x(s)
Ci- / \ @ T=70
Ex(s)
Linear Gradient Pycnocline
No Collapse Complete Collapse No Collapse Complete Collapse
6.2xlO"4
7
940
3420
l.OxlO'4
31 ,400
5,300
2.5xlO"5
22,200
32,500
3xlO"6
70
27,400
21 ,600
6.5xlO"4
20,600
440
3.0xlO"6
27,400
21,500
1.5xlO"7
25
3200
1900
3.7xlO"4
15,400
6,400
7.8xlO"8
12,700
34,000
0
70
-
-
4.42xlO"3
12,500
386
0
0
-
C  = Maximum Surface Concentration


L  = X Dimension of Cloud (ft)
 X
Cr  = X Centroid Location of Concentration Distribution (ft)


T = Time in hours

-------
                                                                                                 -iio"
cr>
                                                               C = collapsed cloud
                                                               U = uncollapsed cloud
                                                               L = linear density gradient
                                                               P = pycnocline
                                                        Cactual   CR(Cinitial)
               /
                           I I
L
                                                           1st appearance of surface concentration^
                                                                                                     4-5
                                                                                                           i_
                                                                                                           3
                                                                                                           (/)
                                                                                                    10
                                                                                                      -6
                                                                                                    10
                                                                                                      -7
                                                                    10
                                                                                                         -P
                                                                                                         0)
                                                                                                         o
                                                                                                         o
                                                                                                         o
                                                                                                         a;
                                                                                                         o
       3
       OO
       -a
       Q)
       +j
       u
       -a
       
-------
CTi
CO
              u
              c
              L
              P
uncollapsed cloud
collapsed cloud
linear density gradient

pycnocline
                                                                                   I   I  I  I I  I i
                                                                                                          10
                                                                                                            - 1
                                                                                                         10
                                                                                                          10
10
                                                                                               100
                                                                                               X
                                                                                               03
                                                                                                               c
                                                                                                               o
                                                                                                               •I—
                                                                                                               -M

                                                                                                            -2  £
                                                                                                            - 3
                                                                                                               O)
                                                                                                               u
                                                                                                               (U
                                                                                             to

                                                                                             O)
     X
     re
                                                   Hours
              FIG. 14.  Maximum  Predicted  Concentration for Long-term Dispersion Stage

-------
m
U
C
L
P
    uncol lapsed  cloud
    collapsed  cloud
    linear density gradient
    pycnocline
FIG.  15.
                                  1              Hours
                             Predicted X Dimension of Cloud at Level  of Maximum Concentration
                                                                           00

-------
                C = collapsed cloud
                U = uncollapsed cloud
                L = linear density gradient
                P = pycnocline
                                        Hours
                                                                                                  - 6000
                                                                                                  - 5000
                                                                                                 - 4000
                                                                                                 - 3000
                                                                                                 - 2000
                                                                                                 - 1000
                                                                                                            X
                                                                                                            (O
                                                                                                          C
                                                                                                          O
                                                                                                         o
                                                                                                         o
o

•M
C
CD
                                                                                                         CD
                                                                                                         4->
                                                                                                         U
                                                                                                         •r—
                                                                                                         -o
FIG.  16.  Predicted Relative Concentration Centroid Location at  Level  of  Maximum Concentration

-------
the time of maximum surface concentration is increased to 70 hours with
a resulting reduction in the magnitude of this maximum concentration.
The collapsed cloud for the pycnocline case never reaches the surface.
The uncoilapsed pycnocline cloud does, however, reach the surface at
approximately 35 hours but with a much reduced concentration when refer-
enced to the uncoilapsed linear case.

     The restrictive nature of the pycnocline essentially appears to  trap
the contaminants and prevent the majority from ever reaching the surface.
The maximum concentration for the collapsed cloud in the pycnocline occurred
                                                                      _1 0
at a depth of approximately 80 feet and reached a maximum of 8.9 x 10~
at 35 hours decreasing approximately 1/3 in the following 35 hours, thus
 indicating that for that particular example the surface would have been
 essentially  free of contamination for all time barring any  great changes
 in the  boundary or environmental conditions.

      The maximum  concentration  centroid  locations  (Fig. 16) for the collapsed
 clouds  are found  to  be relatively  stable moving a maximum distance of only
 440 feet compared  to  a 5-6000 foot  movement associated with the uncoilapsed
 clouds.  The surface  concentration  centroids  follow the same general  trend
 with the centroid  locations after  70 hours  having  moved  6 miles for  the
 uncollapsed clouds and closer to 4 miles for the  collapsed  cases.  These
 values  are both for  the case of a  linear gradient  and  therefore  represent
 a different set of extremes, this  is necessary because  of  the failure of
 the collapsed thermocline case to  penetrate the  thermocline and  reach the
 surface.

      Only one representative cloud dimension is  examined  and that  is  !_x
 the cloud length  in  the direction  of the flow (see Fig.  15).  The  trend here
 follows that seen for the maximum concentration-exposure  time curves  with
 the cloud length  at t=70 hours decreased for either the  inclusion  of a
 collapse mechanism or a pycnocline or both.  If  the conservation  of  mass
 is considered, where:

      L = f (M/C)      M = Mass     C = Concentration     L  = Length
                                    71

-------
    This trend can  be explained where  as  shown  in  Fig.  13  (t=70  hours),
the concentration  increases  with  both  the inclusion  of  a collapse  mechanism
or a pycnocline or  both.   This relationship  applied  to  the  above equation
predicts the same  trend as was found for  the cloud dimensions.

    The cloud dimensions  are directly  determinate only for X, Z directions
as the analysis, in its present form,  gives  only  the concentration distri-
bution over two non-rigid boundaries and  does not  include  buildups asso-
ciated with the cloud reaching the  bottom of the  surface.

    The distributions of  concentration with  depth  at t=70  hours  are shown
in Fig. 17 for all  the cases considered.   From this  figure the wide variance
in surface concentration  can readily  be realized  as  well  as the  variety  of
profiles that must  be considered  when  these  concentrations are  being sampled
for in the field.   This figure probably best typifies the  problem  at hand
and allows a generalization  that  shows collapse to flatten this  vertical
distribution.  It can also be concluded that the  collapse  mechanism represents
a phenomenon as powerful  in  its effect on some parameters  as the inclusion
of the pycnocline over a  linear gradient.  It is  perhaps appropriate here  to
point out that while the  linear uncollapsed  cloud distribution  appears to
overestimate the resulting concentrations it underestimates the  possible
exposure times by a factor of three.   This should not be overlooked as the
effects evidenced in marine plants  and organisms  are dependent on  a combination
of the two.

    A need is clear for research  on the collapse mechanisms and the forces
which drive it.  Emphasis should  be placed on the possible entrainment and
dilution that may occur during this phase and on the effect of  particle settling
and its influence on the internal density structure of the cloud.

    The deviations that may occur from the  neglect of  cloud collapse have
been  shown by this analysis to result in predictions which describe:
                                   72

-------
—i
CO
     0


    20


    40


    60


    80


   100


   120


£  14°
Q.
0)
Q  160


   180


   200


   220


   240


   260

   280


   300
                                    UP
                                                                            collapsed cloud
                                                                            uncollapsed cloud
                                                                            linear density gradient
                                                                            pycnocline
                          10"
                        10
                                      -7
10~6      10       10""     10
   Relative Concentration C/R\
                                                                           -3
10
                                                                                    -2
10
                                                                                              -1
                               FIG.  17.   Concentration  Distribution with Depth at T = 70 hours

-------
     1.  Higher surface concentrations

     2.  Smaller contaminated surface areas

     3.  Shorter times to the realization of maximum surface contamination

     4.  Faster dilution rates for the maximum concentration

     5.  Larger areas over which the maximum concentration is distributed

For these reasons, the only course of action open at this time is to conduct
analyses that assume both cloud stability and collapse providing the maximum
and minimum conditions expected.  These values are, of course, influenced
by the no entrainment collapse mechanism presented herein.
                                   74

-------
                              SECTION VII

                         BARGE CHARACTERISTICS

      Wastes are discharged from barges either in bulk or containerized
 form.  Containerized methods have been used for toxic, radioactive,
 and a variety of industrial wastes as well  as the more notorious surplus
 poisonous war gases.  Generally the most popular waste container is the
 55-gallon steel drum which can  be carried to the site and simply dropped
 overboard.   Reclaimed drums have an expected life of 10 years  (90), but
 this can  be extended by filling with a concrete mixture containing  the
 contaminants.   Scientists  have  indicated that the voids in the  concrete
 may result  in  implosion of the  drum and fracture may occur at depths
 between 100 and 1000 meters.

      Two  types  of  barges,  towed  or self-propelled,  employing either
 pumped or gravity  discharges  are  usually used for bulk  disposal.  Until
 recently  the self-propelled barge  had  been  limited  to  the  hopper dredge,
 however,  the "Glen Avon,"  an  automated  sewage  disposal  vessel was recently
 purchased by the city of Bristol,  England.   It  has  the  following features
 (91):

      1.   900 ton load capacity

      2.  Maximum speed  of  12  knots

      3.  Discharge time of  15 minutes

     4.  Low pressure air-gravity discharge system

     These characteristics may be compared with those of the Hopper  Dredge
(Table 6)  which operates in the  following manner.  During the  dredging
process the bottom material is pumped in a diluted state into  hoppers

-------
                       TABLE  6



SPECIFICATIONS OF CORPS OF ENGINEER'S HOPPER DREDGES*
Name

Essayons . . .
Goethals ...
Biddle
Comber
Gerig
Langfit ...
Harding ...
Markham . . .
Mackenzie. . .
Hains
Hoffman . . .
Hyde
Lyman
Davison ...
Pacific ...
* Reference
Length,
beam, and
depth
Feet
525x72x40
476x68x36
352x60x30
352x60x30
352x60x30
352x60x30
308x56x30
339x62x28
268x46x22
216x40x15
216x40x15
216x40x15
216x40x15
216x40x15
180x38x14
(92)
Maximum
hopper
capacity
Cu. yd.
8,270
6,442
3.060
3,422
3,060
3,060
2,682
2,681
1,656
885
920
720
920
720
500

Maximum
draft
loaded
Ft. in.
31-0
29-0
24-33A
24-33/.
24-3'A
24-33/t
20-3
20-0
21-0
13-0
13-0
13-0
13-0
13-0
11-3


Number

2
2
2
2
2
2
2
2
1
1
1
1
1
1
1

Dredcje j>um
Size
Inches
32
30
28
28
28
28
20
23
26
20
20
20
20
20
18

ps
Horsepower

1,850
1,300
1,150
1,150
1,150
1,150
650
2650
900
410
410
410
410
410
340


-------
equipped with  overflows.   The hoppers can be emptied in 3-15 minutes
dependent  upon the  volume  and nature of the material.  Generally, rela-
tively fluid materials  are dumped quickly whereas sticky clays and certain
granular materials  may  require the washing of the hoppers with large
volumes of water,  a process known as monitoring.

     Towed barges  are by far the most numerous and are quite variable
in characteristics.  They range from the simple bottom release mud scow
used in small  dredging  operations to specially designed, automated tank
barges for sewage  and industrial sludges, toxic liquids and gases, and
pressurized liquids.  Creeman,  (93) in a discussion of the loading, un-
loading and transport of tank barges illustrates three basic configu-
rations (Fig.  18):

      (a)  single skin
      (b)  double skin
      (c)  double skin with  independent containment  vessel

      Most petroleum products  are carried in single  skin  barges  (a) with
 poisons,  acids  and cargos  requiring heat or insulation utilizing  double
 skinned  (b) vessels.  The cylindrical  tank  barges  (c)  are  generally used
 to carry  liquids  under pressure, however, it is not uncommon  for  pressurized
 liquids  to be transported in double skinned vessels.

      Barges can carry  large waste loads  and discharge  the  contents  quickly.
 Servizi,  et al. (9), reported on a proposed dredging operation  that would
 use 300  cubic yard (approximately 350 tons) bottom dump barges  with bottom
 openings  (16  by 65 feet).  A self-dumping unmanned 5000 ton barge was
 described (94) for dumping chemical insecticide wastes at  sea at distances
 not less  than 125 nautical miles.  All  valves, pumps,  and  other machinery
 were contained in square tanks that comprised the interior of the 298  foot
 x 50 foot barge.

      Eberman  (95) reported on a deep-sea disposal  barge with a rated capacity
 of 1150  tons  in rough  weather, and 1300 tons in fair weather.  The  barge
                                    77

-------
CARGO
      SPACE
CARGO
      SPACE
             (A) Single Skin

	 - —
CARGO
SPACE
	 • 	 	 .
CARGO
SPACE



                                               VOID
              (B)  Double Skin
       CARGO
           SPACE
                                               VOID
 (C)  Double Skin  with Independent  Cargo  Spaces
     FIG. 18.   Basic Barge Configurations
                     78

-------
has six  compartments  for the cargo, two rows of three each and a loading
manifold for  pumping  the liquid material  into any of the compartments.
A set of spring-closed check valves on the bottom of each compartment are
opened only  to  permit flow out of the tank when a reduction of pressure
occurs on the discharge side of the valve.  A vertical turbine pump with
a capacity of 2200 gpm against a head of 33 feet discharges through an
8-inch pipe,  extending 12-feet below the deck level of the barge.  In
practice, 4 hours are required to pump out the barge at a pumping rate
of approximately 1000 gpm.

     The National Lead Company at Sayreville, N. J., uses a barge for the
disposal of spent acids and ore washing sludge from a titanium processing
plant (96).   The system consists of rubber-lined shore storage facilities
including dock  and dredged harbor and two barges for the transport to the
disposal area located 38 miles from the plant dock.  The two barges have
capacities of 5400 and 3200 tons.  The larger barge is used regularly while
the  smaller one  is on a standby basis.  The 5400 ton barge is 289 feet  long
with  a  22 foot depth and a load draft of  17 feet and 5 feet when empty.  All
8  flat-bottomed  waste acid tanks and two  cone-bottomed mud tanks are  rubber-
lined.  The disposal operation takes place  while the  barge is underway  at  a
speed of 8.5 knots.  The waste acid  is pumped  through  two  12-inch discharge
pipes at the keel level past  specially designed  skegs  approximately 50  feet
apart.  A load can be discharged in  70 minutes  at  a  rate  of  78  tons per minute,
The  mud is discharged by  gravity from  the air-agitated  cone-bottomed  tanks.
The  smaller  standby  barge  has two  12-inch diameter discharge  pipes  43 feet
apart at  keel  depth,  which  for this  barge varies from 15  feet when  fully
loaded  to 6  feet when  empty.   The  rate  of discharge  has  been  reported (86, 87)
to vary from 16  to  39 tons  per minute  while being  towed  at a  speed  of 6 knots.

      New York  City  barges digested sludge with an  automated  6300 ton  barge
capable of  handling  cargoes of liquids,  acids, or  suspended  materials.   The
 226 feet long, 56 feet wide, and 20 feet deep barge  is radio-controlled and
 has an  unloading time of 30 minutes.
                                    79

-------
                                                       TABLE  7

                                                BARGE CHARACTERISTICS*
oo
o
(Tons)
Capacity
5,400

3,200
5,000
1,200
1,100
8,000
6,300
350
*Reference
Type of Waste
Iron-acid
Ore washing mud
Iron-acid
Chem-Insecticides
Chlorinated
Hydrocarbons
Chlorinated
Hydrocarbons
Philadelphia
Digested Sludge
NYC Digested Sludge
Dredge Spoil
(34)
Average Discharge Characteristics
Depth of Pipe Number Towing Discharge
Discharge Type Size Spacing Speed Rate
Feet Inches Knots Tons/mi n.
10 Pumped 12 2 @ 50 ft. 8.5 78
10 Gravity - 2 @ 50 ft 8.5
10 Pumped 12 2 @ 43 ft 6.0 16 - 39
10 Gravity - - -
12 below
deck Pumped 81 6.0 5
8 Pumped 41 6.0 4
Gravity 24 8 267'
210
17-20 Gravity Bottom dump - 100 - 200

-------
     The  City  of Philadelphia uses an 8000 ton barge for hauling digested
sludge with  a  2700 ton unit for backup (18, 19, 20, 21).  The sludge is
hauled approximately 110 miles where it is dumped in 30 minutes through
eight 24-inch  bottom valves.

     Table 7 summarizes towed barge characteristics from the literature
readily available on this subject.

                          Barging Economics

     Many factors have been shown to influence the dispersion and dilution
of wastes dumped into the ocean.  The physical parameters over which some
control can be exercised to minimize the effects of waste disposal on the
ocean environment are the same parameters which influence the associated
costs.  These include discharge rate, water, depth, barge capacity and
distance to the disposal area.

     The literature contains  reports on many  individual barging operations,
few  of which  include  enough  information for meaningful  comparison.  A  paper
by Gunnerson  (3)  has  summarized the  reported  costs  and  presents average
disposal costs  on a dollar  per wet  ton basis.  These figures are  presented
in Table 8, and are representative  of  the  following geographic areas:
Philadelphia, New York  City, Elizabeth, New Jersey, Baltimore, and Washington,
D. C.
                                TABLE 8
          REPORTED  COSTS OF BARGING OPERATIONS IN  $/WET TON*

   Waste                     Total     Pacific   Atlantic      Gulf
 Industrial  (a)  bulk          1.70        1.00        1.80       2.30
   (b)  containerized         24.00       53.00        7.73      28.00
 Refuse and garbage           15.00       15.00        	       	
 Sewage sludge                1.00        	      (.8-1.2)     	
 *Reference (3)
                                  81

-------
     To explore and present a method that analyses the towing costs for
various barge sizes, sludge quantities and towing distances the follow-
ing assumptions are made.   It is assumed that:

     (1)  The community owns its own barges and contracts for tug services
thus removing the hidden costs of profit and overhead margins.

     (2)  A standby barge, similar in design to the primary barge, is
required.   The capacity of this barge is set at 1000 tons.

     (3)  The availability of tug services is limited to 260 days per
year allowing for holidays, strikes, repairs, etc.

     (4)  Loading costs are included in the operating cost of the facility.

     (5)  There are three  major cost categories, namely:

          (a)  Capital  Costs

          (b)  Maintenance Costs

          (c)  Towing Costs

                            Capital  Costs

     The capital  costs  used are based on a figure of $170.00 per ton (14)
and represent the purchase of new, specially constructed barges with radio
controlled rapid discharge systems.   The average discharge time will be
taken at 90 minutes.

     Service life of ocean going barges varies between ten and twenty years,
For this example a barge life of ten years will be assumed for the primary
barge and 20 years for the standby barge.  Annual costs will be calculated
                                   82

-------
using an eight percent  interest  rate,  equal  replacement cost and no salvage
value or benefits  from  depreciation.

                          Maintenance  Costs

     Eberman  (95)  estimated a maintenance cost of $800.00 per trip for a
1500 ton barge making six trios  per year or  a total  yearly cost of $4800.00.
This is  approximately 12 percent of the annual capital  costs of a barge, as
described  for this example, using the  capital cost format itemized.  This
practice of calculating maintenance costs as a percentage of the annual capital
cost is  used  in  the Portland, Oregon area by several firms (95, 97, 33).  This
approach,  where  maintenance is independent of both frequency and duration of
use, is  felt justifiable as salt water corrosion accounts for the major portion
of required maintenance and repairs, usually accomplished during one annual
dry docking.   The capital and maintenance are combined to provide one fixed
yearly operating cost.   Table 9 shows the distribution of these costs for
this example.

                               TABLE 9
         CALCULATED ANNUAL FIXED COST FOR BARGING OPERATIONS

                   Capital Costs
                                                          Annual  Fixed  Cost
                                                           (Nearest  $1000)
                                                              48,000
                                                              76,000
                                                             105,000
                                                             161,000

                              Towing Costs

     The  greatest amount of variability is  expected to occur in the towing
 costs.  This cost can  be influenced by discharge rate, tug  speed, distance
Capacity
(tons)
1000
2000
3000
5000
Primary

25,340
50,680
76,020
126,700
Secondary

17,330
17,330
17,330
17,330
Maintenance

5.120
8,161
11,202
17,284
                                    83

-------
traveled and the existing environmental  conditions (i.e., wind, sea state,
etc.).  Costs in the Oregon, Alaska and  Washington areas were found to
range between $75.00/hr and $85.00/hr and on this basis an average cost
of $80.00/hr was selected as representative of the cost of a tug capable
of handling barges up to 5000 ton capacity at an average net speed of
six knots.   The towing costs on a per trip basis can be calculated from
a simple relationship of the following form:

     Ct = Tc {^+ t} 	[36]

where:
     C.  = total towing cost in dollars/trip

     Tc = towing charge in dollars/hr

     x  = round trip distance traveled in miles (point to point)

     v  = tug speed in miles/hr

     t  = unloading time in hrs.

     From this relationship the total costs per trip can be shown to in-
crease directly with travel time for constant discharge times.

     Solving Eq. 36 for various values of (x) reveals a cost/trip mile
relationship as given in Fig. 19.
                                   84

-------
  20



  18

(D


iie
Q.

±14
o
            J	L
            50       100       150       200      250


                     Round Trip Distance in Miles


           FIG. 19.  Cost per Trip Mile as a Function of

                     Round Trip Haul Distance
                                300
         Assuming  a  loading time of 3 hrs/ton the limiting number of possible

    trips for a  single barge operation can be determined for the stated tug

    availability of  260 days/year.  These limits are shown for this example

    in Figure 20.
                                                                    ton

                                                                    hrs
                                                                                  •:on
                                                                                  •:on
                                                                                  •;on
                                                                                  ton
                50
100
250
     FIG. 20,
                        150          200

                   Round Trip Distance in Miles


Limiting Number of Trips per Year for Preset Barge Sizes
                                       85

-------
     The total  annual  cost (fixed and towing)  can be  presented by a
series of graphs where total  annual  cost is  plotted against  the number
of trips made in a year for given barge sizes, round  trip haul  distances,
and the total annual  tonnage to be disposed  of.   This method of display
(Figs. 21, 22,  23) can provide a method that will aid in the evaluation
of alternatives as well  as predict the barge size that will  accrue the
least cost.

     This example was not intended to be a black box  model that would
reproduce the reported costs found in the literature  but rather an
exercise to expose the controlling variables in barging costs.  The
relationships and figures presented can be used to determine many factors,
for example Fig. 21 for x equal to 20 miles  shows a 1000-ton barge to be
most economic size for yearly sludge loads that do not exceed 100,000 tons
with the 2000 ton barge economically feasible between 100,000 and 300,000
tons and the 3,000-ton barge becoming the cheapest to operate at 600,000
tons/year.  These tonnage figures can be related to peculation of the
community if the percent solids and the percent reduction in solids of
the process are known.  Using 250 mg/1 SS, 250 gpd/cap, 10 percent solids
and a 33 percent reduction by digestion it can be shown that a population
of one million will produce 365,000 tons of sludge per year.   If the haul
distance were  increased to 100 miles round trip, a 5000-ton barge would
be necessary to minimize the costs.  The additional operating  costs incurred
when the standby barge is needed are also given  in Figures 21, 22, and 23.

     The average cost shown in Table 8 for sewage sludges ranged between
 .8 and  1.2 dollars per wet ton.   If an average yearly  sludge load of  100,000
 tons  is  assumed with  a round trip haul distance  of 50  miles, Fig. 23  can  be
 used  to  calculate  the estimated minimum yearly cost  and  barge  size.   This
 results  in a barge of 1610 ton  capacity making 64  trips  per year at a  total
 cost  of $117,000  or  $1.17/wet  ton which  is within  the  average  given by
 Gunnerson.
                                    86

-------
      264
00
                         Annual Tonnage in Thousands
           Operating cost for standby barge
                                                         Trips



                             FIG. 21.  Annual Operating Costs  for  Round Trip  Haul  Distance  of  20  Miles

-------
CO
CO
        288
        264  '
                          Annual Tonnage in Thousands
              Operating cost for standb^
                                        10
                                                          Tri ps
                               FIG.  22.  Annual  Operating  Costs  for Round Trip Haul  Distance of 50 Miles

-------
00
vo
       fO
       rs
       c
      -p
      o
432



396



360



324



288



252



216



180



144



108



 72



 36



  0
                             Annual  Tonnage in Thousands
                \          \
                50UO  ton      \
Operating cost for standby.
   i ••    • i"   "t    i   i  i  i  i  i
                                                                  i     i
                                              10
                                                                             100
                                                                                              1000
                                                             Trips
                              FIG. 23.  Annual Operating Costs for Round Trip  Haul  Distance  of 100 Miles

-------
     The City of Philadelphia  has  presented  the most  complete  set  of
figures (18,  19)  and  will  serve  as  a  test  of the  validity  of this  proposed
analyses.   The conditions  are:

     (1) x =  220 miles

     (2) Barge size  = 2700 tons

     (3) Sludge load  equal to  90 million gallons  @ 220 gal/ton
Example Costs are as  listed below:

     Primary  Capital  Cost           $68,418

     Standby  Capital  Cost            17,330

     Maintenance Cost               10,289
          Total Fixed               96,037
     Operating Costs              $387,000
                      TOTAL       $483,037

     This reduces to $5.36/1000 gallon or $1.20 per wet ton again within
the average cost/wet ton  cited by Gunnerson  (3).  The operating costs
for the Philadelphia operation were given as $3.73/1000 gallons-trip and
can be  calculated at $4.14 by this method,  however the actual average tug
speed was 6.35 knots and  accounts for the majority of the  $0.41 differential
shown above.

     This approach satisfies  the need for a method by which approximate
costs  of  alternative costs of action can  be evaluated.  The effects of
factors such  as  haul distances,  barge sizes, sludge  loads, barge  speeds,
and  discharge methods,  determined  to provide a desired dilution  and penetration,
can  be economically  compared  allowing the cheapest equivalent combination
of design factors to be used  to accomplish  the  desired objectives.
                                   90

-------
                             SECTION  VIII

                             USER'S GUIDE

     This  report  has  established the  fact  that ocean dumping is a
reality  and  that  it embodies  certain  risks to the integrity of the
marine environment.   To  this  point, little has been said regarding
the practical  use of  this  analytical  technique.  The information
derived  provides  a measure of the dilution, spread and travel or drift
of a discharged waste material  under  various sets of initial conditions
and assumptions.   These  data, if properly  interpreted and applied to a
particular operation  and site,  may be used to determine how, when and
where a  waste  should  be  discharged to preclude violating standards
established to protect areas  of high  marine productivity.

     The California  Regional  Water Quality Board, San Francisco Bay
Region,  has established such  standards and they became effective January 1,
1971.  This regulation establishes a protected zone extending seaward
approximately  thirty  three miles.  This action was taken to  protect the
rich marine nursery  which extends  to the  Farallon  Islands, and essentially
bans harmful waste disposal in  this area.

     The action taken in  California was at  this writing  the  most ex-
tensive in the U. S.  and  may in  fact be the  forerunner of  a  more general
trend.  Without examining the  specific  considerations given  in the  es-
tablishment of such  a boundary,  it will be  assumed  that  any  discharges
made outside  this area must  be  accomplished  in such  a manner that the
protective  regulations  are in  no way violated.  Therefore,  the basic
question  is again the one of where, when  and how  such discharges should
be  allowed.   The  approaches  and methods outlined  in  the  main text can
be  used to  determine the  answers to  such  questions  provided  environmental
conditions  descriptive  of the  area are  known.  Seasonal  environmental
changes may be indicative of the when;  dilution,  drift  and spread,  the
where;  and  initial dilution  and penetration, the  how.   These parameters

-------
are all interrelated and a continuous set of evaluations will  have to be
made for each set of conditions  used.

     The data presented in this  report are continuous through  the three
transport phases for only one set of initial conditions for which the
following inputs were used.  These are in no way descriptive of any par-
ticular operation or geographic  area and are used only to explore a use
method.  These conditions are:

     1.  Linear density gradient (Table 10)

     2.  An initial densimetric  Froude number of 0.142 based on an initial
discharge radius (b ) of 8 feet  and an initial velocity of 0.494 feet per
second.

     All other conditions are identical to those for the example problem in
Appendix I.

     From this informational base Fig. 9 can be used to determine the pen-
etration and the dilution that occurs during the convective descent stage.
The computer output, as shown in Appendix  I, provides a final cloud radius
and position both of which may be necessary in either determining the input
for the collapse phase or in determining the total extent of the material
transport.  For this example these parameters are:

     Penetration depth  (Yf)      = 159 feet
     Dilution  (DILN)             = 580
     Final radius  (bf)           = 66  feet
     Travel in  X direction       =1193  feet

     Since the  waste cloud  reaches a state of buoyant  equilibrium both
the collapse  and long  term diffusion-dispersion  processes must  be considered.
The two  possible limits  are  exposed  by  considering both no  collapse  and
                                    92

-------
a cloud  that  undergoes  a  complete collapse.  These values are obtained for
a derived  value  of  K~  given  by Equation 11  and subsequent use of Figure 5.
                               _3
     For this example  K2~3.8xlO~  with the  collapsed radius equal to 264
feet and an assumed height of 2.06 feet.  The collapse time must also be
determined from  Fig.  5 and used to determine the drift.  For the cited K~
value the  time of collapse was 9.4 minutes  and for a net current of 0.2
knots gives a drift of 193 feet.

     Figures  13, 14,  and  15 were constructed for these particular cloud
dimensions and therefore  can be used to predict the concentration and
position as functions  of  time.  The concentrations predicted from this
phase of the  analysis  are relative and assume an initial concentration of
unity.

     Let us  assume that to meet the water quality standards it is necessary
to reduce the concentration at the end of the convective descent phase by
a factor of 1000.  Figure 14 can be used to determine  the time for the max-
imum concentration in the profile to be reduced by this  amount.  These times
are 30 hours  for no collapse and 50 hours for complete collapse.  Figure
15 gives the final length or diameter  of the cloud at  this  time  and only  a
simple calculation is necessary  to determine the  drift,  assuming of course
that the cloud is  carried along  at the same  speed as  the current.  The total
transport can be determined by  simply  adding together  the various components.

                                      (uL)                    (cL)
Convective Descent                1193 feet               1193  feet
Collapse  Drift                        - 0 -                  193  feet
Long  term
  Drift (U)  (t)                   36,000 feet            60,000  feet
  Spread                           5,250                 5,750
                                  42,443                 67,036
                                    93

-------
     Conversion to nautical  miles  gives U/L*  - 7 miles and
miles.  The results of this  approach predict  that for the input conditions
and assumptions, discharge should  not be allowed within 11 miles of the
boundary.   The total  dilution  that has actually occurred during these three
transport  phases can  be shown, assuming a conservative material with no
decay to be:
         3p
      1 =  (CQ)  (1/500)  (10~3
_.         I-..                 —
         ufinal  =
         °final  =

     If we use the San Francisco Bay Protected Zone as an example and apply
an additional  restriction of 11 miles, some feel for the additional costs
involved can be obtained.  Assume that before the regulation became effective
discharges were made at a point 19 miles off the coast with a total haul
distance of 25 miles.  The annual tonnage was approximately 100,000 tons and
the operation utilized a 2000 ton primary barge with   1000 ton barge held
in standby.  Fig. 22 can be used to predict an annual cost, including the
purchase, maintenance, interest and tug fees of $115,000 or $1.15/ton.  The
combination of the two additional restrictions increases the haul distance
to 50 miles and by Fig. 23 predicts an increase in annual costs of $20,000
with the cost/ton increased to $1.35 representing an increase of 17.4 oercent
in cost for a 100 percent increase in haul distance.

     Hopefully, this exercise demonstrates the potential of analytical  ap-
proaches of this form.  This is certainly only one use and only the first
step.  Future research and development should refine and improve the overall
program limiting the number of assumptions that must be made and making the
results more directly  applicable to the real world situations.
                                    94

-------
                             SECTION IX


                             BIBLIOGRAPHY
1.  "Ocean  Dumping:   A  National  Policy."   A Report to the  President
    prepared  by  the  Council  on Environmental  Quality, 45 pages, October
    1970.

2.  Koh,  R. C. Y., and  L.  Fan.  "Prediction of the Radioactive Debris
    Cloud  Distribution  Subsequent to a Deep Underwater Nuclear Explosion
    (U)."   A  Final Report  prepared for the Naval  Radiological  Defense
    Laboratory,  Contract No. N00228-68-C-0684.  266 pages.  October 1968.

3.  Gunnerson, C.  G., R. P.  Brown, and D.  D.  Smith.  "Marine Disposal  of
    Solid Wastes."   A paper  presented at  ASCE Ocean Engineering Conference,
    Miami  Beach, Florida.   December 1969.

4.  "Investigation of Treatability and Effect on  Biota of Dredged Materials
    from Selected Great Lakes Harbors."  University of Wisconsin, U. S.
    Army Corps of Engineers  Contract No.  DACW49-68-C-0042.  September 1968.

5.  Einstein, H. A., and R.  L. Wei gel.  "A Literature Review on Erosion and
    Deposition of Sediments  near Structures in the Ocean."  University of
    California  Report HEL  21-6.  February 1970.

6.  "Dredging and Water Quality Problems  in the Great Lakes."  Buffalo
    District U.  S. Army Corps of Engineers, Pilot Program Summary, 1969.

7.  "Effects on  Fish Resources of Dredging and Spoil Disposal  in San
    Francisco and San Pablo Bays, California."  USDI Fish and  Wildlife
    Service Special   Report.   36 pages.  November  1970.

8.  Briggs R. B.  "Environmental Effects of Overboard Spoil Disposal."
    Journal of  the Sanitary Engineering Division, ASCE, SA3, p. 132-3.
    June 1968.

9.  Servizi, J.  N.,  R.  W. Gordon, and  D. W. Martens.  "Marine  Disposal
    of Sediments from Bellingham Harbor as Related  to Sockeye  and  Pink
    Salmon Fisheries."  Progress  Report No.  23.   International Pacific
    Salmon Fisheries Commission,  p.  1-38.  1969.

10.  Brehmer, M. L.   "Turbidity and  Siltation  as  Forms of  Pollution."
    Journal  of  Soil  and Water Conservation.   Vol  20, No.  4, pp 132-3.
    July  1965.

11.  O'Connor, B.  A., and  J. E. Croft.   "Pollution in a  Tidal  Estuary."
    Effluent and Water Treatment Journal  (Brit.).  Vol  7, No.  7,  p 365-74.
    T96T

12.   Brown, D.  L., and R.  Clark.  "Observations on Dredging and Dissolved
     Oxygen in  a Tidal  Waterway."  Water  Resource Research.   Vol.  4, No.6,
     p. 1381-84.  1968.

-------
13.  Brooks, N. H.   "Predictions  of Sedimentation  and  Dilution of Digested
     Sludge in Santa Monica  Bay."  Report  to  Hyperion  Engineers, April 1956.

14.  Orlob, G. T.   "Effects  of  Digested  Sludge  Discharge  on the Ocean En-
     vironment near the City of San Diego  Outfall."  Report to San Diego
     Regional  Water Pollution Control  Board,  August  1965.

15.  Randall,  C. W., and C.  T.  Koch.   "Dewatering  Characteristics of Aero-
     bically Digested Sludge."   JWPCF.   Vol 41,  No.  5  part 2, p. R215-238.
     May 1969.

16.  Hinesly,  T. D., and B.  Sosewitz.   "Digested Sludge Disposal on Crop
     Land."  JWPCF.  Vol  41, No.  5  part  1, p. 822-830.  May 1969.


17.  Burd,  R.  S.   "A Study of Sludge  Handling and  Disposal."  FWPCA Publica-
     tion No.  WP-20-4.   May  1968.

18.  Guarino,  C. F.  "Sludge Disposal  by Barging to  Sea."  W & SW.  Vol 114,
     p R126-7.  November 1967.

19.  "Sludge Disposal at Sea."   Civil  Engineering.   Vol 38, No. 8, p. 62-3.
     August 1968.

20.  Baxter, S. S.   "Sludge  Disposal  in  Philadelphia."  Journal of the
     Sanitary  Engineering Division.  ASCE, No.  SA6,  Vol 85, p 127.  November
     1959.

21.  "Annual Report."  City  of  Philadelphia.  1968.

22.  University of  Rhode Island.   "Progress Report-Systems Analysis for
     Shipboard Municipal  Incineration."  Contract  No.  U.  I. 00557-03.  1968.

23.  Buelow, R. W.   "Ocean Disposal of Waste  Material."   Transactions.
     National  Symposium on Ocean Science and  Engineering  of the Atlantic
     Shelf. Marine  Technology Society.   Philadelphia,  March 19-20, 1968.

24.  Beyer, F.  "The Spreading  of Sinking  Pesticides in the Sea."  SIW.
     Vol 27, Mo. 9, p.  1073-80.  September 1955.

25.  Hume,  N.  B.,  and W. F.  Garber.  "Marine  Disposal  of  Digested Screened
     Wastewater Solids." Third International Conference  on Water Pollution
     Research, Munich, Germany, Section  III,  paper No.  12, 1966.

26.  North, N. J.   "Ecology  of  the  Rocky Nearshore Environment in Southern
     California and Possible Influences  of Discharged  Wastes."   International
     Conference on  Water Pollution  Research.  Pergamon Press, London, p.  247-
     274.  1962.

27.  Carlson, D.  E., and C.  S.  Zichefooze.  "Factors Affecting Coliform
     Reduction in  Chlorination  of Digested Sludge."  Appendix E,  Disposal
     of Digested Sludge to  Puget Sound.  METRO.   Seattle,  Washington.  July
     1965.
                                    96

-------
28.  Brooks, N. H., D. E. Carlson, and R. C. Y. Koh.  "Function of Outfall
    and Dilution in Puget Sound."  "Disposal of Digested Sludge to Puget
    Sound."  Submitted to METRO, Seattle, Washington.  1965.

29.  Sylvester, R. 0.  "Sludge Disposal by Dilution in Puget Sound."  JWPCF
    Vol 34, No. 9, p. 891-900.  September 1962.

30.  Hood, D. W., B. Stevenson, and L. M. Jeffrey.  "Deep Sea Disposal
    of Industrial Wastes."  Industrial and Engineering Chemistry, Vol 50,
    No. 6, p. 885-888.  June 1958.

31.  Carritt, D. E., et al.  "Radioactive Waste Disposal into Atlantic and
    Gulf Coastal Waters."  NAS-NRC Report No. 655.   1959.

32.  Joseph, A. B.  Sanitary Engineering Conference,  Baltimore, Maryland.
    1954.  USAEC Report.  Washington  D. C.  1955.

33.  Waldichuck, M.   "Containment of Radioactive Wastes for  Sea Disposal
    and Fisheries off Canadian Pacific Coast."  Proceedings of 1959
    Monaco Conference, AEC.  Vol III.  1960.

34.  Collins, J. C.,  (ed.).  Radioactive Wastes.  Wiley & Sons.  1960.

35.  Koczy, F.  F.   "The  Distribution of  Elements  in the Sea."  Disposal of
    Radioactive Wastes.   Vol  2,  p  189-197.   IAEA,  Vienna,  1960.

36.   Issacs,  et al.   "Disposal  of Low  Level  Radioactive Wastes in  Pacific
    Coastal  Waters." NAS-NRC  Report  No.  958.   1962.

37.  U.  S.  Coast and Geodetic  Survey.   "Sea Disposal  of  Low Level  Packaged
    Radioactive Waste."   C&GS  Project No.  10.000-827.

38.   Pneumo Dynamics Corporation.  "Survey of Radioactive  Waste  Disposal
     Sites."   Technical  Report ASD 4634-F prepared under Contract  AT(04-3)-331
     for AEC.  p.  131.  1961.

39.   Pneumo Dynamics Corporation.  "Sea Disposal  Container Test  and Evalu-
     ation."   Technical  Report under AEC contract AT(04-3)-367,  p. 132.   1961.

40.   Mauchline, J.  "The Biological and Geographical  Distribution  in the
     Irish Sea of Radioactive Effluent from Windscale Works, 1959-60."
     United Kingdom Atomic Energy Authority, Harwell, p.  70.  1963.

41.  Preston, A., and D. F. Jeffries.  "The Assessment of the Principal
     Public Radiation Exposure from,  the Resulting Control  of, Discharges
     of Aqueous Radioactive Waste from the United Kingdom Atomic Energy
     Authority Factory at Windscale,  Cumberland."  Health Physics.  Vol  13,
     No. 5, p. 477-85.  1967.
                                     97

-------
42.  Sabo,  J.  J.,  and  P. H. Beckosian, editors.  "Studies of the Fate
     of Certain  Radionuclides in Estuarine and Other Aquatic Environments."
     PHS Publication.  No. 999-R-3.  1963.

43.  Purushothaman-.f-K. and E.?F. Gloyna.   "Radioactivity Transport in Water;
     Transport of    Sr and    Cs under Induced Clay Suspensions."  Report
     ORO-490-13, Environmental Health Engineering Research Laboratory,
     University  of Texas, Austin, Texas.   1968.

44.  Chia-Shun Shin, and E. F. Gloyna.   "Influence of Sediments on Transport
     of Solutes."   Journal of the Hydraulics  Division, ASCE, Vol 95, HY5,
     p. 1347-67.   July 1969.

45.  Tenny, M. W., and T. G. Cole.   "The  Use  of  Fly Ash  in Conditioning
     Biological  Sludges  for Vacuum  Filtration."  JVIPCF.  Vol 40, No. 8
     part 2.  August 1968.

46.  Deb, P. K., A. J. Rubin, A. W.  Launder,  and K. H. Manuy.   "Removal
     of COD from Wastewater by Fly  Ash."   Proc.  21st  Ind. Waste Conference,
     Purdue University,  Ext. Service 121, 848.   1966.

47.  O'Viatt,  C. A.  "The Effects of Incinerator Residue on  Selected Marine
     Species."  Proceedings of the  Annual North  Eastern  Regional Anti-Pollutior
     Conference, July  22-24, University  of Rhode Island.  1968.

48.  Rogers, B.  A.  "A Progress  Report  on Studies  to  Determine  the  Effects
     of Suspended  Solids on Marine  Fishes With  Special  Reference to  the  Effect
     of Finely Divided Incinerator  Residue on Selected  Species."   Department
     of Industrial Hygiene, Harvard School of Public  Health, p.  24,  1968.

49   Mahoney, J. R.  "Feasibility  for Ocean Based  Solid  Waste  Disposal."
     Harvard University, School  of  Public Health,  Department of Environmental
     Health Services,  7  pages,  1968.

50.  Kinsman, B.  Wind Wave.   Prentice  Hall, Englewood,  N.  J.   1965.

51.  Wei gel, R.  L.  Oceanographical Engineering.  Prentice  Hall.   Englewood,
     N. J.  1964.

52   U  S   Navy Hydrographic  Office.  "Atlas of Surface Currents,  North
     Atlantic Ocean, 1947a."   H. 0. Publication 571, First  Edition.  1947.

53   U  S   Navy Hydrographic Office.  "Atlas of Surface Currents,  North-
     eastern  Pacific Ocean, 1947b."  H.  0. Publication 570, First  Edition.
     1947.

54  U  S   Navy Hydrographic Office.  "Atlas of Surface Currents,  North-
     western  Pacific  Ocean, 1950."  H. 0. Publication 569,  First Edition.
     1950.
                                    98

-------
55.  Montgomery, R. B.  "Fluctuations in Monthly Sea Level on Eastern
    U. S. Coast as Related to Dynamics of Western North Atlantic Ocean."
    J. Marine Research.  Vol. 1, p. 165-185.  1938a.

56.  Ippen, A. T. (ed.).  "Estuary and Coastline Hydrodynamics."  McGraw-
    Hill Book Company, New York.  pp. 558.  1966.

57.  Morton, B. R., et al., "Turbulent Gravitational Convection from
    Maintained and Instantaneous Sources."  Proc. Royal Society Academy.
    Vol 234, p. 1-23.  1956.                                   	

58.  Baumgartner, D. J., R. J. Callaway, G. R. Ditsworth.  "Disposal of
    Aluminum Process Wastes in the Ocean."  Working Paper No. 64, Pacific
    Northwest Water Lab, FWQA, Corvallis, Oregon.  March 1968.  (Reprinted
    December 1969)

59.  Glover, R. E.  "Dispersion of Dissolved or Suspended Materials in
    Flowing Streams."  Geological Survey Prof. Paper 433-B, USGPO,
    Washington, D. C.  1964.

60.  Glover, R. E. and C. R. Daum.  "Behavior of Dissolved and Suspended
    Materials in Flowing Steams."  Seventh Hydr. Conf. Proc., lowas
    Inst. Hydr. Research, State Univ. of Iowa.  June 1958, p. 133-142.

61.  Ketchum, B. H., C. S. Yentsch, and N. Corwin.  "Some Studies of the
    Disposal of Iron Wastes at Sea."  Woods Hole Oceanographic Institution
    Reference 58-7, 17 pp., unpublished manuscript, 1958.

62.  Carter, H. H. and A. Okubo.  "A Study of the Physical Processes of
    Movement and Dispersion in the Cape Kennedy Area."  File Report under
    the U. S. Atomic Energy Commission Contract Report No.  NYO 2973-1,
    Chesapeake Bay Institute, The Johns Hopkins University.  1955.

63.  Okubo, A.  "A Review of Theoretical Models of Turbulent Diffusion in
    the Sea."  Journal Oceanpgraphic Society-Japan.  20th Anniversary
    Volume, p. 286-320.  1962.

64.  Okubo, A.  "Some Remarks on  the Importance of the  Shear Effect on
    Horizontal Diffusion."  Journal Oceanographic Society-Japan.  Vol 24,
    p. 2460-69.   1968.

65.  Okubo, A.  and M. J.  Karweit.   "Diffusion from a Continuous Source
    in a  Uniform  Shear  Flow."   Limnology & Oceanography.  Vol 14, p. 514-
    520.   1969.

66.  Partheniades,  E.   "A  Summary of the present knowledge of the Behavior
    of Fine  Sediments  in  Estuaries."   MIT Hydrodynamics Lab., Tech. Note
    No. 8.   June  1964.
                                   99

-------
67.  Task Committee on  Sedimentation,  Sediment Transportation Mechanics:
     Suspension of Sediment,  Progress  Report, ASCE Journal of Hyd. Div.
     Vol. 89, No.  HY5,  Part  1.   September  1963.

68.  Schmidt, Wilhelm.   "Der  Mussemaustausch in  freier Luft und verwandte
     Erscheinungen."  Probleme  der  Kosmischen Physick, Band 7, Haul burg,
     Germany.  1925.
                                                         X
69.  O'Brien, M.  P.   "Review  of the Theory of Turbulent Flow and its
     Relation to Sediment  Transportation." AGU,  pages 487-91.  April
     1933.

70.  Sutherland, A.  J.   "Entrainment of  Fine Sediment by Turbulent Flows."
     Calfornia Institute of Technology,  Report No. KH-R-13.  June 1966.

71.  Shields, A.   "Anwendung  der Aehnlichheist-smechauik und der
     Turbulenzforschung auf die Geschiebebewegung."  Mitt, der Preusss.
     Versuchsanstalt fur Wasserbau  und Schiffbau, Berlin.  1936.

72.  White, C. M.   "The Equilibrium of Grains in the Bed of a Stream."
     Proc.  Roy. Soc.  Vol  174A, pp. 322-34.  1940.

73.  Vanoni, V. N.  "Trans, of  Suspended Sediment by Water."  Trans. ASCE.
     Vol. Ill, pages 670-133.  1946.

74.  Anderson, A.  G.  "Distribution of Suspended Sediment  in a Natural
     Stream."  Trans. AGU.  pages 678-683. 1942.

75.  Kada,  H. and Hanratty,  T.  J.  "Effects of Solids on Turbulence
     in a Fluid."  AICHE Journal.  Vol 6,  No. 4.  December 1960.

76.  Rouse, Hunter.  "Elementary Mechanics of Fluids."  John Wiley & Sons,
     Inc.  Hew York. Pages  245 and 348.  1946.

77.  Vanoni, Vito.  "Some  Experiments  in Transport of Suspended Load."
     Trans. AGU.  1941.

78.  Householder, M. K. and  V.  W. Goldschmidt.   "Turbulent Diffusion
     and Turbulent Schmidt Number of Small Particles  - Part  I."  ASCE
     Nat. Meeting Preprint 692.  Chattanooga, Tenn.  May  1968.

79.  Householder, M. K. and  V.  W. Goldschmidt.   "Turbulent Diffusion
     of Small Particles in a  Two-Dimensional  Free Jet."   Purdue University
     Technical Report No.  FMTR-68-3.  September  1968.

80.  Singamsetti, S. R.  "Diffusion of Sediment  in a  Submerged Jet."
     J. Hydraulics Division, ASCE,  Hy  2, pp.  153-168.  1966.

81.  McNoun, J. S. and  P.  N.  Lin.  "Sediment  Concentration and Fall
     Velocity."  Proc.  Second Midwestern Conference  on Fluid Mechanics,
     Ohio State University,   pp. 401-411.   1952.
                                   100

-------
82.  Ahmadi, G. and V. W. Goldschmidt.  "Dynamic Simulation of the
    Turbulent Diffusion of Small Particles."  Appendix E, Interim
    Annual Report, FWPCA Project No. 16070 DEP.  November 1969.

83.  Diaschishin, A. N.  "Dye Dispersion Studies."  Journal Sanitary
    Engineering Division, ASCE, Vol 89, No. SA1.  January 1963.

84.  Abraham, G. and W. D. Eysink, et al.   "Full Scale Experiments
    on Disposal of Waste Fluids into Propeller Stream of Ships."
    Reprint Rome FAO Conference.  December 1970.

85.  Schlichting, H.  "Boundary Layer Theory."  McGraw-Hill Book
    Company, 6th Edition, pages 692-695.   1968.

86.  Redfield, A. C. and L. A. Wai ford.  "A Study of the Disposal of
    Chemical Waste at Sea."  NAS-NRC Pub #201.  1951.

87.  Ketchum, B. H. and W. L. Ford.   "Rate  of Dispersion in the Wake
    of a  Barge at Sea."  Trans. AGU..  Vol 33, No. 5, pp 680-684.
    October 1952.

88.  Pearson, E. A., P. N. Storrs, and  R. E. Selleck.  "Some Physical
    Parameters and Their Significance  in Marine Waste Disposal."  In
    Pollution and Marine Ecology, T. A. Olson and F. J. Burgess  (Editors)
    Interscience Publishers.  New York.  pp. 297-315.  1967.

89.  Webster, C. A. G.   "An  Experimental Study of Turbulence in a
    Density-Stratified  Shear Flow."  J. of Fluid Mechanics, Vol  19,
    pp.  221-245.   1964.

90.  Fulkerson,  Frank  B.   "Transportation of Mineral  Commodities  on  the
    Inland  Waterways  of the South  Central  States."   U. S. Department of
    the  Interior,  Bureau of Mines,  Infor.  Circular  8431.  1969.

91.  "Sewage Disposal  Vessel  is  Highly  Automated."   Ocean  Industry.  Vol  5,
    No.  2,  pages  55-56.   February  1970.

92.  Mauriello,  L.  J.  and L.  Caccese.  "Hopper  Dredge Disposal  Techniques
    and  Related Development in  Design  and  Operation."  Symposium No. 3  -
    Sedimenatation in Estuaries, Harbors  and  Coastal  Areas,  Misc.  Publ.
    970.  U.S.  Department  of  Agriculture, Paper No. 65,  pp.  598-613.

93.  Creelman, W.  A.   "Pollution Control  in the Barge Transportation of
    Bulk Liquids."  JWPCF.   Vol 41, No.  11, Part 1, pp.  1879-1885.
     November 1969.

94.   Anonymous.  "At Sea About Chemical Wastes."  Chemical Week.,  pp.  133.
     October 14, 1967.
                                   101

-------
95.  Eberman,  J.  W.   Sewerage and  Industrial Wastes.  Vol 28, No. 11,
     pp.  1365-70.  November  1956.

96.  Peschiera,  L. and  Freihers.   "Disposal of Titanium Pigment Processing
     Wastes."  JWPCF.   Vol 40, No. 1.  January 1968.

97.  Waller, Robert.  "Deep  Sea Disposal of Drummed Wastes."  E. I. du Pont
     de Nemouis  & Company, Inc., La  Porte, Texas.  March 1968.
                                  102

-------
                              SECTION X

                             APPENDIX  I

    This section is intended to be used as a guide  for program  use,
the input requirements for the convective descent  stage analysis  are
presented as they are to be entered on  the program control  card.   The
program is currently in conversational  mode; however card  input  re-
quirements will be identical:
PROGRAM CONTROL CARD:
FORMAT:  3F (5.0), 4F (10.0), 2F (5.0,  IF (10.0),  IF (5.0)

     COLUMN                                DESCRIPTION
      1-5	entrainment  coefficient (a)
      6-10	drag  coefficient (CD)
     11-15	added mass coefficient (C^)
     16-25	initial discharge (V ) (enter
                                          as  a  negative  quantity in Ft/sec)
     26-35	initial radius of the cloud  (BQ)
     36-45	initial density of the receiving
                                           body in gms/cc  (p )
                                                            d
     46-55	initial density of waste sludge
                                           or slurry in gms/cc (pw)
     56_60	time increment  for internal
                                           integration (use 0.033 for printout
                                           intervals between 5 and 20 feet)
                                           (DT)
     61-65       	print  out interval	incremental
                                           depths at which solution printout
                                           is desired  (Ym)
     66-75	maximum  water  depth  in feet	entered
                                           as a  negative  quantity (Y^)
     76_80         	number of steps to  be entered  in the
                                           environmental  orofiles (see  discussion
                                           in following  section) (NC)

-------
                  10 Pi        20121        30131        40 141       50151       60 I 61       V0|71       80
             Y           p      '     U      '     W      '
               1°  11        20  21        30  31        40  41        50 51       60 61       70J71       80
         Y       '     p      '     U      '      W                                         '
            1011
                        2 02 1
                                    3031        4041
                                           W
                                                   5051       6061
                                                                            70  71
DEPTH
  Y
1 0|1 1         2 ob 1        3031         40
    DENSITY   ' VELOCITY  '  VELOCITY
     (p)          (U)          (W)
                                              r         soi
5051        6061
                        7071
                                                   Format  4(F10.0)
 6  1011  15
  CD    CM
             1 6
                      25E6
                                  3536       4546
                                       F
                                                          55
                                                             56  60
            1  6 5 |6 I
                                                                                  75
                                              W
                                                      DT    YM      -Yr
                                                                                    76  80
                              NC
                            Format [3F(5.0), 4F(10.0), 2F(5.0), IF(IO.O), 1F(5.0)]
                                 APPENDIX I:   Corrective Descent

-------
DATA CARDS

     There should be one card,  at least for each depth where a change in
any of the environmental parameters  occurs plus one additional card
descriptive of a point one interval  above the waters surface which maintains
the same gradient as the first  interval below the surface.   The program
assumes a linear gradient between entered points; therefore, in areas where
the changes are rapid, closely  spaced points should be entered on either
side of the profile change.   These cards must be ordered so that the deepest
point is read first proceeding  upward to a point one increment above the
water surface.   Values must be  entered for all terms on each card.  The
total number of cards or steps  used  to describe the profile must be entered
in columns 76-80 on the program control card.

Data Card Format:  4(F 10.0)

     COLUMN                                    DESCRIPTION
       1-10	Depth of point.. .entered as a
                                               negative quantity when below the
                                               water surface
      11-20	Density in gms/cc at the depth
                                               entered in columns 1-10
      21 -30	Net hori zontal veloci ty i n X
                                               direction at depth entered in
                                               columns 1-10
      31-40 	Net horizontal velocity in Z
                                               direction at depth entered in
                                                columns 1-10

     The output from this analytical ohase  itemizes the control values entered,
prints the environmental profile and presents the convective  descent solutions
in tabular form.  The output symbols are  defined as follows:
                                   106

-------
    SYMBOL                                DESCRIPTION
    T                                  Descent time in seconds
    Y                                  Penetration depth in feet
    X                                  X position of the cloud center in feet
    Z                                  Z position of the cloud center in feet
    B                                  Cloud radius in feet
    V                                  Cloud velocity in ft/sec
    BETA                               Density disparity
    DILN                               Dilution of the cloud

    The following example problem is provided as a sample of  the  input
and output for the Convective Descent Analyses portion.
    Initial  Conditions:
        a =  .167
       CD =  .250
       p  = 1 .075 gtns/cc
       Yf = 500 feet
        V   =  1975  ft/sec
        o
        DT  =  0.033
        b   =  4.00  ft
        o
        p   =  1 .025 gms/cc
        Y   =10.0  feet
        m
                                    107

-------
     Density profiles  are  given  for  the  case  of linear  gradient  (Table  10)
and a strong pycnocline  (Table 11-13).   The output  is shown  in Tables  12
and 13.   A discussion  of this example  is given  in the main portion  of  the
text.

                              TABLE 10
                  Linear Density Gradient [16 steps]
Y
500.0
300.0
250.0
200.0
150.0
100.0
65.0
60.0
50.0
40.0
30.0
20.0
10.0
• 5.0
0
50.0
Pa
1.029850
1.028510
1.027925
1 .027340
1.026755
1.026170
1.025749
1.025691
1.025584
1.025467
1.025350
1.025233
1.025116
1.025058
1.025000
1.024420
U
0
0
0
1 .0000
2.5400
2.8800
3.0620
3.0880
3.1400
3.1920
3.2440
3.2960
3.3480
3.3740
3.4000
3.6600
W
0
0
0
0
0
0
0
.0370
.1075
.1950
.3750
.7050
1.3150
1 .7000
1 .7000
1.7000
                                   108

-------
          TABLE 11





Strong Pycnocline [23 steps]
Y
-500.0
-300.0
-250.0
-200.0
-150.0
-100.0
- 65.0
- 60.0
- 56.0
- 55.0
- 54.0
- 53.0
- 52.0
- 51.0
- 50.0
- 49.0
- 40.0
- 30.0
- 20.0
- 10.0
- 5.0
0
50.0
pa
1.028900
1.028500
1.028400
1 .028300
1.028200
1.028100
1 .028030
1.028020
1 .028001
1.028000
1.027400
1 .026800
1 .026200
1.025600
1.025000
1 .025000
1.025000
1.025000
1 .025000
1 .025000
1.025000
1.025000
1 .025000
U
0
0
0
1 .6000
2.5400
2.8800
3.0620
3.0880
3.1088
3 . 11 40
3.1192
3.1244
3.1296
3.1348
3.1400
3.1452
3.1920
3.2440
3.2960
3.3480
3.3740
3.4000
3.6600
W
0
0
0
0
0
0
0
.0370
.0655
.0725
.0795
.0865
.0935
.1005
.1075
.1077
.1950
.3750
.7050
1.3150
1.7000
1.7000
1.7000
              109

-------
                  TABLE  12





Linear Profile Convective Descent  Output





    X          Z          B            V          BETA            DILN
3.3
5.2
11.1
18.3
27.1
37.4
50.1
65.2
84.0
109.1
155.4
189.7
372.0
554.4
-6.61
-10.04
-20.10
-30.07
-40.13
-50.02
-60.11
-70.13
-80.11
-90.07
-100.02
-101.84
-92.51
-95.40
4.43
8.28
22.78
41.37
64.42
91.64
124.65
163.89
212.01
275.66
390.83
474.87
915.34
1354.46
2.22
4.08
10.06
15.66
20.57
24.63
28.11
30.94
33.28
35.38
37.85
39.07
41.97
42.78
5.705
6.403
8.312
10.192
12.121
14.059
16.089
18.188
20.398
22.868
26.516
28.942
43.740
59.066
-1.931
-1.838
-1.531
-1.263
-1.047
-0.877
-0.731
-0.598
-0.469
-0.326
-0.114
.000
-0.000
.000
-0.0171812
-0.0121227
-0.0054572
-0.0028725
-0.0016150
-0.009434
-0.0005364
-0.0002732
-0.0000936
.0000332
.0001145
.0001061
-0.0000280
.0000106
2.90
4.10
8.97
16.54
27.83
43.42
65.08
94.01
132.61
186.86
291.32
378.79
1307.58
3219.89

-------
                 TABLE 13





Strong Pycnocline Convective Descent Output





  X           Z           B           V             BETA           DILN
3.3
5.2
n.i
18.2
26.7
36.7
57.5
91.9
118.1
-6.62
-10.05
-20.14
-30.04
-40.00
-50.02
-59.35
-50.19
-54.94
4.43
8.28
22.78
41.05
63.34
89.67
144.01
233.16
300.97
2.22
4.08
10.06
15.57
20.33
24.28
29.62
35,35
38.70
5.705
6.403
8.317
10.180
12.081
14.021
16.557
20.053
22.573
-1.932
-1.841
-1.539
-1.282
-1.083
-0.929
.003
-0.003
.001
-0.0172300
-0.0121896
-0.0055622
-0.0030334
-0.0018147
-0.0010084
.0013284
-0.0016155
.00105910
2.90
4.10
8.99
16.48
27.55
43.07
70.93
126.00
179.70

-------
•" 10 11 1516 20J21 30 31 35 3
DTi NTCi NPTr ' DT2 NTC2 IN
V
INPUT FORMAT 4(E
1 Oil 1 15162021 3031 35364
DYi 'NDYI — DYZ NDYJ —
INPUT FORMAT 4(ElO.
6 "» d if 1 5051
PT2 DT3 NT
5556 60 61
C3 NPT3 DT,,
7071 75J7680
NTCH WT,,
10.5, 215) 4 Points/Card
0 "f 1 50J51 55
DY3 NDY3
5, 15, 5X) 4
56 60E 1 703
— DY4
Points/Card
NDY^ —



                    1011       2021
            Yk2            Yk3
                                             30
                                                3 1
                                                                      5051
                                                                                   6061         7071
                                                                                                           80
                ToTll         20|21         30J31t(J41
                                          4(]41        5051        6061        70

                                                Yr          GAMW          Yw
71        80
                                                 FORMAT  8(E  10.5)
•^      56
 NEND   I  NDY
1 Oil 1   1 5J1 6   20J2 1
Y I NOT  SPRINT'
                                          FORMAT (10I5)1
                                              APPENDIX  II:   Long Term Dispersion

-------
                             APPENDIX II

     Both  the  input and output for the long term dispersion stage analyses
are  normalized,  dimension!ess, values.  There are two basic input control
cards and  three  or more output and program control cards as shown on the
face sheet of  this Appendix section.

     The  input controls are entered in accordance with the following format
and normalization scheme:

Format 8(E 10.5)
Column       Symbol                 Normalization           Reference Fig.
 1-10          A                     hA/Ky(l)bo

11-20          6                     h/aQ

21-30          UQ                    Ush2/Ky(l)bo2/3
31-40          YE                    Y/h                         25
41-50          YE1                   Y/h                         25
51-60          GAMW                  W(max)h2/Ky(l)bo2/3YW       25
61-70          YW                    Y/h                         25
71-80          YK                    Y/h                         26
                  i
Card #2                                                          26
  1-10          YK                    Y/h                         26
                  2
 11-20          YK                     Y/h                         26
 21-30          YK3                    Y/h                          26
 31-40          BETA  1                 Ky(y)/KyO)                 26
 41-50          BETA  2                 Ky(y)/Ky(D

-------
       The initial conditions which must be known, all of which are defined
   and described in the text, are:

       a  = Y dimension of cloud at end of collapse period [ft]
       b  = X dimension of cloud at end of collapse period [ft]
       h  = depth of cloud center at end of convective descent stage [ft]
       U/ \= max x velocity component [ft/sec]
       W  = max z velocity component [ft/sec]
                                                                             2
       K /,» = vertical diffusion coefficient used to describe mixed layer [ft /sec]
        YUy                                                       2/3
       A  = dissipation parameter usually assumed as a constant [ft   /sec]

       The program will accept velocity and vertical diffusion coefficient
   profiles only when they are normalized as shown in the following diagrams:
+1.0
 0.0  -
-1.0  -
, .0
0.0
                                       -1.0
                                                      VI
             DEPTH NORMALIZATION
        VELOCITV CHORD. SYSTFM
             FIG. 24.  Coordinate System and Depth Normalization
                                    116

-------
 1.0;
 0.0
-i.a
 i.o
 0.0
-l.O1-
           U Profile
NOTE:  Both profiles must
       go to zero at same
       point (Y£)

 W Profile
                   FIG. 25.  Velocity Profile Normalization
                                        117

-------
 1.0
 0.0
-1.0-
                      1.0,
                                  Ykl52)
         p(y)
     K(y)
LINEAR
                      0.0
                     -1.0
P(y)
K(y)
                                                     PYCNOCLINE
      FIG.  26.   Vertical  Diffusion Coefficient K(y)  -  Normalization  Method
           The vertical diffusion coefficient profile relates the decrease in
       vertical diffusion with  increasing density to the magnitude of the K^
       value with  depth.

           A  normalized depth  and time  grid  system must be established  for
       program output control.   The  depth grid should  include one interval above
       and  below  the  normalized depths of  (+1.0).  The grid is  defined  as follows
           DY/.^v   =  size of  the individual  step
           NYC,.% = number  of steps  of that size
           NDY     =  total  number of step  size changes
           NPRINT =  number of  Y grid  points  established  between but including
                     (+1.0)  N  print = [(E
                                         118

-------
+1 0

0
-i.o -












\7
DY3
DY2
For tht
MI DY1=
DY2 =
DY,=
DY2
DY3 N Pr
NDY
                                                              j=  8
                                                           NYC2=  6
                                                           NYC3=  6
                  FIG. 27.  Depth Normalization
     The  time  grid is  normalized as  follows:
              t1 = K
                    y(i)t
solving this equation for t'=1.0 will  allow real  time choices to be selected
for output.  The output will then, using the plotting subroutine, plot the
distribution of concentration with depth at the requested times.
     The time grid utilizes symbols similar to those of the Y grid namely:
         DT/.-j  = size of individual normalized time stop
         NTC/.-v = the number of steps of that size
         NOT
= the number of step sizes used
         NPT/-1 = the number of steps required to reach the requested output time
The summation of [(DT,. J(NTC,. J should equal 1.0.
                                  119

-------
     If output is  desired  at  normalized  times  .01,  .1,  15  and  1.0  the
steps could be arranged  as follows:
Example 2

Step change  (i)

     1
     2
     3
NOT  4
                     .001
                     .002
                     .005
                     .010
                            NIC

                             10
                             45
                             80
                             50
                                        (i)
                                                  T. Nor.  time
NPT,
.01
.10
.50
1.00
10
45
80
50
     The output from this  program is also in normalized form and uses  the
following symbols:
     T      = time
      oo
            = total  concentration of material  in XZ plane
     I(C  )  = integral  of C   over the depth
        oo
      max
    a a a
     xx xa
                  oo
   = X centroid location
   = Z centroid location
   = Max concentration in XZ plane
   = normalized Y (as given by grid)
   = standard deviations of the concentration distribution
     These values can be denormalized as follows-:
     (1) The horizontal dimensions of the cloud L/.
where L,^ = 4bQai where i = x,z
     (2) The cloud centroid locations in X,Z where:
CE
  (x)
                     CE(z) • bZc
                                   120

-------