environmental effects of floating mariculture in Puget Sound

[From the U.S. Government Printing Office, www.gpo.gov]

THE ENVIRONMENTAL EFFECTS OF FLOATING MARICULTURE

IN PUGET SOUND

U S - DEPARTMENT OF COMMERCE NOAA
COASTAL SERVICES CENTER
2234 SOUTH HOBSON AVENUE
CHARLESTON , SC 2905-2413

Donald P. Weston, Ph.D.1

School of Oceanography, WB-10
University of Washington
Seattle, Washington 98195

Prepared for:

Washington Department of Fisheries
and
Washington Department of Ecology

August, 1986

Prc'phrtY of cc .

Current address: Science Applications International Corporation,
af~t 13400-B Northup Way, Suite 38, Bellevue, Washington 98005

The preparation of this report was financially
a~ided through a grant from the Washington State
Department of Ecology with funds obtained from
the National Oceanic and Atmospheric Administration,
and appropriated for Section 306 of the
Coastal Zone Management Act of 1972.

Table of Contents

Acknowledgments .............................. v
Abstract. . . . . . . . . . . . . . . ...................vii
1.0 INTRODUCTION . . . . . ... . . . . 3
2.0 FLOATING MARICULTURE IN PUGET SOUND
2.1 Salmon net-pen culture. . . ............. 7
2.2 Shellfish culture . ...................... 12
3.0 POTENTIAL ENVIRONMENTAL EFFECTS
3.1 Water circulation . . . . . ................... 17
3.2 Water quality . ......................... 22
3.3 Phytoplankton .......................... 39
3.4 Sedimentation ........... . . . . . . . . ......48
3.5 Benthic macrofauna. ....................... 64
3.6 Fish and megafauna . . . . . .................. 74
3.7 Introduction of exotic species .................. 78
3.8 Diseases. . . . . . ......... 85
3.9 Genetic effects ......................... 91
3.10 Toxicants ............ . . . . . . . . . . . ....96
4.0 MODELING OF ENVIRONMENTAL EFFECTS
4.1 Introduction. . . . . . . . . .... 101
4.2 Water quality modeling. . . . . . . . . 101
4.3 Sedimentation modeling . . .................... 106
5.0 ENVIRONMENTAL REGULATION OF MARICULTURE
5.1 Introduction. . . . . . . . . . ...... 115
5.2 State regulations . . . . .................... 116
5.3 Regulations of other countries .................. 118
6.0 RECOMMENDATIONS.. . . . . .................. 127
LITERATURE CITED .............................. 131

Acknowledgments

The following persons provided information included in this report or reviewed
earlier drafts. I thank them for their assistance and cooperation.

Hans Ackefors - University of Stockholm, Stockholm, Sweden
John Baross - University of Washington, Seattle, Washington
Ed Black - Ministry of the Environment, Victoria, British Columbia
Paul Blake - Center for Disease Control, Atlanta, Georgia
Bruce Burrow - Municipality of Metropolitan Seattle, Seattle, Washington
Ken Chew - University of Washington, Seattle, Washington
David Damkaer - National Marine Fisheries Service, Seattle, Washington
Frances Dickson - Fisheries and Oceans, Vancouver, British Columbia
Marvyn Eddie - Marine Harvest Limited, Edinburgh, Scotland
Magnus Enell - Swedish Environmental Research Institute, Stockholm, Sweden
Richard Finger - Municipality of Metropolitan Seattle, Seattle, Washington
John Forster - Sea Farm of Norway, Port Angeles, Washington
Kunihiko Fukusho - National Research Inst. of Aquaculture, Nansei, Mie, Japan
John Galat - New Haven Salmon Ranch, New Haven, New Zealand
Nicolas Gonzalez - Spanish Institute of Oceanography, La Coruna, Spain
Lee Harrell - National Marine Fisheries Service, Manchester, Washington
Ken Honey - Maine Department of Marine Resources, Augusta, Maine
Eric Hurlburt - Washington Department of Fisheries, Olympia, Washington
Akihiko Hara - National Research Institute of Aquaculture, Nansei, Mie, Japan
Bill Hershberger - University of Washington, Seattle, Washington
Bill James - Washington Department of Fisheries, Olympia, Washington
Bob James - Washington Department of Ecology, Olympia, Washington
Peter Jeffards - Penn Cove Mussels, Penn Cove, Washington
Bruce Kay - Environmental Protection Service, Vancouver, British Columbia
Michael Kyte - Ardea Enterprises, Lynnwood, Washington
Jon Lindbergh - North Bend, Washington
John Liston - University of Washington, Seattle, Washington
Lars-Ove Loo - Institute of Marine Research, Lysekil, Sweden
Conrad Mahnken - National Marine Fisheries Service, Manchester, Washington
Bob Matsuda - Municipality of Metropolitan Seattle, Seattle, Washington
Asha Mahtre - Department of Public Utilities, Olympia, Washington
Louis Mottet - Friday Harbor, Washington
Madelon Mottet - Friday Harbor, Washington
Charles Nolan - Seattle-King County Health Department, Seattle, Washington
Arthur Nowell - University of Washington, Seattle, Washington
Bruce Pease - Washington Department of Fisheries, Point Whitney, Washington
Deborah Penry - University of Washington, Seattle, Washington
Mary Jane Perry - University of Washington, Seattle, Washington
Tommy Petersen - Faroese Sea Breeding Commission, Faroe Islands
Gil Potter - Washington Dept. of Social and Health Serv., Olympia, Washington
Bob Saunders - Washington Department of Ecology, Olympia, Washington
Knut Senstad - Sea Farm of Norway, Bergen, Norway
Lynn Singleton - Washington Department of Ecology, Olympia, Washington
Emil Smith - California Department of Fish and Game, Sacramento, California
James Staley - University of Washington, Seattle, Washington
Richard Strickland - Thalassaco Science Communications, Seattle, Washington
Jack Rensel - Milner-Rensel Associates, Seattle, Washington
Rutger Rosenberg - Institute of Marine Research, Lysekil, Sweden
Barry Uchida - Municipality of Metropolitan Seattle, Seattle, Washington
Dennis Willows - University of Washington, Friday Harbor, Washington
Ron Zebal - Washington Department of Fisheries, Point Whitney, Washington
V

ABSTRACT

Mariculture, particularly net-pen culture of salmon and suspended culture

of shellfish, has grown rapidly in Puget Sound and the potential exists for

this growth to continue. This review attempts to assess the probable magnitude

of environmental changes that may result from mariculture operations in Puget

Sound. The environmental changes which have been addressed can be summarized

in four categories. First, there are some adverse changes which are highly

probable for most salmon or suspended shellfish culture operations. These

would include the accumulation of organic-rich sediments beneath the culture

structure, and the consequent effects on the benthic macrofauna community.

Secondly, the potential exists for mariculture effects on water circulation,

water quality, phytoplankton productivity and the introduction of exotic spe-

cies. However, the likelihood of these effects and their potential magnitude

is highly dependent upon site-specific conditions or the species cultured.

Thirdly, there are several issues for which available data are inadequate for a

conclusive determination of significance. These include: 1) the environmental

effects of antibiotic usage; 2) alteration of the wild gene pool; 3) the capac-

ity for a mariculture operation to serve as a disease reservoir for the infec-

tion of wild organisms; and 4) the proliferation of human pathogens in the
vicinity of mariculture sites. These potential effects of mariculture have not

yet been demonstrated, and the existence of such effects remains largely specu-

lative. Finally, there is little likelihood that mariculture would adversely

affect local abundance of fish and megafauna.

Water circulation - A culture structure placed in the marine environment

will reduce current velocity in the surrounding area, particularly in the down-

current direction. This reduction in current velocity will impair dilution and

dispersal of wastes downstream of the culture operation. However, this effect
is not likely to be significant except in cases of intensive culturing in an

area with very restricted natural circulation.

Water quality - Cultured organisms and culture practices alter the chemis-

try of the water passing through the culture structure, most notably increasing

ammonia concentrations and decreasing dissolved oxygen concentrations. Greater

vii

changes will occur with salmon net-pen culture than with mussel culture, given
the relative production capacities of typical Puget Sound facilities.
The concentrations of nutrients and BOD in the water passing through a
mariculture structure are generally very dilute compared to most other dis-
charges to the marine environment. Field studies have typically observed
little or no changes in water quality outside the culture structure in well-
flushed areas. Adverse effects would be anticipated only in areas of extremely
limited flushing or very intensive culturing activity.

Phytoplankton - Floating mariculture is unlikely to have a measurable
effect on phytoplankton standing stock or productivity in most of Puget Sound.
In the Main Basin of Puget Sound nutrients are not limiting to phytoplankton
growth. Thus, additional nutrient input attributable to mariculture should
have no effect on productivity. Nutrients may be periodically limiting in some
vertically stratified area (e.g., Saratoga Passage), however, in many of these
areas the dilution of nutrients which would occur prior to their utilization by
phytoplankton would probably make any effect on productivity unmeasurable. A
localized and measurable effect on productivity would be expected only in high-
ly stratified and poorly flushed areas (e.g., some embayments of the southern
Sound) in which the water passing through a culture structure could maintain
its integrity with minimal dilution.

Sedimentation - Floating mariculture generates large amounts of solid
wastes in the form of feces and unutilized feed (salmon culture) or feces,
pseudofeces and shell debris (shellfish culture). These materials are gener-
ally deposited in the immediate vicinity of the culture structure. This depo-
sition results in physical and chemical changes to the natural sediments in-
cluding decreased redox potential, increased sediment oxygen consumption and
increased concentrations of total volatile solids, total organic carbon, sul-

fides, nitrogenous compounds, and phosphates. While there are profound effects
on sediment chemistry, and consequently the sediment biota, these effects are
very localized. Visible accumulation of solids and the alteration of sediment

chemistry typically extends no more than 30 m from the culture structure.
Accumulation of organic-rich sediments occurred around most culture facil-
ities reviewed. The depth of this accumulation depends for the most part upon
water depth and current velocity. In general, accumulation of sediments can be

viii

expected beneath any culture facility when less than 15 m exists between the
bottom of the structure and the sea floor. Sediment accumulation is possible
at even greater depths, but little data are available since few culture opera-
tions have been sited in deeper waters.

Benthic macroinvertebrates - The accumulation of organic-rich sediments
beneath culture facilities and the consequent depletion of oxygen in the pore
waters results in changes in the infaunal invertebrate community. Loss of
species intolerant of organic enrichment (typically echinoderms, crustaceans
and molluscs) often occurs. A few opportunistic species, most frequently poly-
chaetes, attain numerical dominance in the community. In cases of extreme
organic enrichment, there may be a total absence of benthic macroinvertebrates.
Benthic community changes have been typically observed to be limited to an area
within about 30 m of the culture structure. These changes can be expected to
persist for the duration of culture activities and for at least several years
following their cessation.

Fish and megafauna - Accumulation of organic material in the vicinity of
a mariculture operation may result in the loss of nonmotile megafauna (e.g.,
geoducks) living in intimate contact with the sediments. However, fish and
motile megafauna (e.g., crabs) living on or above the sediment surface are typ-
ically found in higher densities around floating mariculture structures than in
the surrounding area. The attraction of fish and megafauna to the culture area
is probably due to increased availability of food in the form of feed unutil-
ized by the cultured fish, the high abundances of opportunistic macroinverte-
brates, and epifaunal organisms living on the culture structure or which fall
to the bottom.

Introduction of exotic species - If culture operations require the impor-
tation of live material into the state, the potential introduction of exotic
species probably represents the greatest environmental threat posed by mari-
culture. Introduced species may establish self-sustaining wild populations,
potentially becoming pests or eliminating native species. While there appears
to be little risk of this occurring for those species currently being deliber-
ately introduced for culture in Puget Sound, requests for future introductions
should be carefully evaluated in this regard.

Accidental introduction of pests, parasites and diseases associated with
imported organisms may also occur. Because of this risk the state has placed
many restrictions on the type of organisms which may be imported and the steps
which must be taken before live material destined for aquaculture may be placed
in state waters. The adequacy of these regulations is currently being eval-
uated.

Diseases - Some concern has been expressed that the culture environment
may serve as a reservoir for those diseases which are present in an environment
but demonstrating no clinical symptoms in wild fish. There is a fear that the
disease organism may proliferate among the cultured fish, become more virulent,
and reinfect the wild stocks. However, there is no evidence to indicate that
this scenario has ever occurred. The outbreak of a disease is often associated
with some form of stress. In the culture environment, fish may be stressed by
overcrowding, undernourishment, poor water quality and physical damage assoc-
iated with handling and confinement. Thus, while fish held in culture are
likely to show more frequent appearance of disease than wild fish, disease does
not appear to be transmitted to the wild populations.
The bacterial diseases of salmonids in mariculture are not transmissable
to humans. However, some investigators have questioned whether the organic
enrichment of bottom sediments caused by mariculture could promote the growth
of those species pathogenic to humans. Consumption of shellfish collected in
the vicinity could then serve as a route for human infection. This is an issue
where available scientific evidence is very meager, but experience to date has
failed to show cause for concern. Increased bacterial abundance in sediments
beneath a mariculture facility is quite probable, but it has not been demon-
strated that this increased abundance is of any significance in terms of human
health.

Genetic effects - The issue of the genetic effects of mariculture is
largely speculative. Cultured organisms may be at a competitive disadvantage
with respect to wild individuals. Thus, if escapes and interbreeding with the
wild population occur, there may be some temporary loss of reproductive capac-
ity in the wild population resulting from the production of less fit genotypes.
The potential magnitude of this effect is dependent upon the proportion of the
breeding population comprised of escaped animals.

The potential consequences of the interbreeding of escaped and wild organ-
isms , if any at all, are unclear. However, for salmonids at least, the poten-
tial magnitude of the problem would seem minimal. For decades fisheries man-
agement agencies have routinely been transferring hatchery-reared salmonids

between river systems to improve commercial and recreational fisheries. The
number of fish which might escape from mariculture is negligable in comparison.

Toxicants - Salmon net-pen culture in Puget Sound requires the occasional
usage of antibiotics, most frequently oxytetracycline. The potential environ-
mental effects of this usage are minimized by the high water solubility of the
compound, rapid dilution with the surrounding water and the infrequency of its
use. Although, there is very little information available, the use of oxytet-
racycline does not appear to be a major problem at the present level of mani-
culture development in Puget Sound.

Modeling of Environmental Effects

Mathematical models are not yet available which provide reliable a priori
estimates of the effect of mariculture on the receiving water body. As a

short-term solution, a model is proposed to predict the changes in dissolved
oxygen or metabolite concentration only in that parcel of water which passes

directly through a culture structure. Since the model neglects the role of
dilution by the surrounding water mass, it does not provide a definitive sol-

ution to modeling water quality. However, the model may have applications to
site evaluation or the determination of maximum allowable culture intensity in

a given area.
Sedimentation models would be valuable in predicting the depth and spatial
extent of the accumulation of organic-rich sediments around a culture struc-
ture. Analytical models which predict this accumulation based entirely on
properties of the settling particles and the moving fluid are beyond the cur-
rent state-of-the-art. Empirical models are limited by site-specificity, but
they represent the only modeling approach of immediate utility. A simple em-
pirical model originally developed for net-pen culture of yellowtail in Japan
is described. Given field measurements of particle settling velocity and
current regime, the model is able to predict the depth of sediment accumulation
as a function of distance from the culture structure.

Environmental Regulation of Mariculture

A review of environmental regulations pertaining to mariculture indicates
that the policies of Washington State are generally comparable to those of many
of the other states and countries surveyed. In general, the potential environ-
mental consequences of mariculture are assessed on a site-specific and case-by-
case basis. Applications for mariculture permits are usually subject to a
review process which includes an opportunity for the responsible government

agencies to evaluate potential environmental effects. There are typically few,
if any, formalized criteria with regards to siting and operation.
Importation of eggs or live animals typically requires a permit from the
responsible government agency. The permit will carry with it certain condi-
tions which, depending on the particular state or country, may include restric-
tions on the country of origin, visual and/or pathological inspections, periods
of quarantine, or disinfection requirements. Policies governing importation
are tending to become more restrictive, and in several countries there is a
total ban on the importation of eggs and/or live animals.

xii

THE ENVIRONMENTAL EFFECTS OF FLOATING MARICULTURE

IN PUGET SOUND

1.0 INTRODUCTION

Mariculture, the farming of plants or animals in marine waters, is under-

going a period of expansion in Puget Sound. For nearly a century, mariculture

in Puget Sound has been largely limited to the production of clams and oysters

by bottom culturing techniques. The commercial harvest of these species in the

Sound is now at about 1,800 and 1,700 metric tons for clams (whole weight) and

oysters (meat weight), respectively. Much interest has recently arisen in

using the waters of Puget Sound for the culture of organisms other than clams

and oysters. The culture of marine algae in the Sound has recently begun,

primarily as a result of research by the Department of Natural Resources. The

Department of Fisheries has been developing a hatchery for the production of

geoduck seed which may be used to repopulate harvested beds. Mussel culture in

the Sound began in the mid-1970s and is now practiced by a few private firms.

Current annual production is about 140 metric tons. Several firms are also

engaged in salmon culture, which first began in Puget Sound in 1969, currently

yields about 1,500 metric tons per year, and now shows potential for rapid

expansion.
Floating mariculture, or those types of culture in which organisms are

suspended within the water column, shows the potential for the greatest growth

in Puget Sound in the near future. Puget Sound lacks expansive areas of

shallow water, thus the useable area for bottom culture is very limited. In

addition, floating mariculture permits utilization of the three-dimensional

nature of the water column, thus allowing a greater production per unit area.

The types of floating mariculture with which this review is concerned are

net-pen culture of salmon and suspended (i.e., raft or longline) culture of

molluscs, particularly mussels. The culture of noni, a marine alga, is also a

type of floating culture conducted in Puget Sound. It is not included in this

review, however, since the technique is currently under extensive study by the

Department of Natural Resources.

With the growth of mariculture in Puget Sound, increasing public attention
has been focused on how culture activities may affect other uses of the Sound.

This report provides a review of available information on the environmental

changes which may occur as a result of floating mariculture. An effort has

been made both to identify the potential effects and to assess their potential

magnitude given the physical, chemical and biological conditions found in Puget
Sound. The purpose of this study is to assist the public, regulatory agencies
and the industry in planning the development of mariculture in Puget Sound.
This report is intended to assist in the review of proposed mariculture ven-
tures by identifying the issues of greatest concern and by providing the data
base necessary to evaluate these issues. The information presented has been
assembled through site visits to several mariculture operations, interviews
with recognized authorities, and an extensive review of the scientific
literature.
The placement of salmon net-pens or suspended mollusc culture in marine
waters may affect the surrounding physical, chemical and biological environment
in many ways. This report addresses the following potential environmental
effects:

1) Changes in water circulation;
2) Sedimentation beneath the culture operation, particularly the accumulation
of feces and excess feed;
3) Changes in water chemistry;
4) Alteration of phytoplankton biomass and productivity;
5) Effects on the benthic macrofauna;
6) Changes in species composition and abundance of fish and megafauna;
7) Introduction of exotic species;
8) Disease transmission from cultured to wild animals;
9) Proliferation of bacteria pathogenic to humans;
10) Changes in genetic fitness of wild stocks;
11) Use of antibiotics and effects on the surrounding biota.

Following a general description of salmon and shellfish culture practices
(Section 2.0), each of the above potential effects is evaluated (Section 3.0).
Each effect is discussed independently in order to simplify the presentation,
but it must be recognized that many of the effects are interrelated. For
example, changes in phytoplankton biomass and productivity are a function of
water chemistry changes (e.g., input of nutrients). Benthic community changes
are a result of the accumulation of sediments beneath the culture.
Section 4.0 describes mathematical models which may be useful in predict-
ing the extent of environmental changes resulting from establishment of a

mariculture operation. Models are presented which could be used to predict the

extent of anticipated water quality changes and the dispersion of particulate
material falling from the culture structure.
Section 5.0 reviews the environmental regulations of other states and
countries which have been established to minimize the environmental effects of
mariculture. These regulations include restrictions on siting, operational
practices and the importation of live animals.

2.0 FLOATING MARICULTURE IN PUGET SOUND

2.1 SALMON NET-PEN CULTURE

The culture of salmon in floating net-pens in Puget Sound was begun in

1969 by the National Marine Fisheries Service (NMFS) at Manchester. Techniques
were further refined in the early 1970s in cooperative experiments between NMFS

and Ocean Systems, Inc. (now Domsea Farms). This early work demonstrated the
feasibility of rearing both chinook and coho salmon to a marketable size within
the confines of a net-pen.
There are presently 9 sites in Puget Sound where salmon are commercially
grown to a marketable size (Figure 1, Table 1). A permit for an additional
net-pen facility (Passage Silver at Bainbridge Island) has been granted, but
pens are not yet in place. Many other net-pen applications are pending, par-
ticularly in Island, Jefferson, Kitsap and San Juan counties. Three net-pen
operations in Puget Sound have failed (Aqua Seafarms, Weyerhaueser Co., Mari-
culture Northwest) because of market/finance problems, or the selection of
sites with inadequate water exchange or recurrent phytoplankton blooms.
Some of the earlier net-pen operations in Puget Sound were located in
sites that would now be quickly dismissed as unsuitable for salmon culture.
Over the years operators have developed criteria to evaluate the physical,
chemical and biological conditions at prospective sites. Some of the major
criteria include:

Minimum current velocity - Current velocities adequate to supply oxygenated

water to the net-pen and remove metabolic wastes, feces and excess feed are
-1
required. In general, a flow of at least 10 cm-sec outside of the net-pen is
desirable throughout most of the tidal cycle (Kennedy, 1978; Leavens, 1983;
Sutterlin and Merrill, 1978). Although occasional and short-term decreases
below this velocity are acceptable (such as at slack tide), sites with consis-

tently low current velocity are avoided.

Maximum current velocity - Excessive current velocities will cause the sides of
the net-pen to distort and stress both the fish and the mooring system. Maxi-
-1
mum current velocities of 50-100 cm-sec have been suggested (Kennedy, 1978;
Leavens, 1983; Nyegaard, 1973; STOWW, 1974; Sutterlin and Merrill, 1978).

* S~~~~~~~~~~~~~~~''ELLINOHAM

RCA~~~~~~C

-- ISLAND. -

VHTORIA

SrRAir oF JUAN orf FIJcA

SEATTLE

Figure I Location of salmon net-pen facilities in puget Sound

Table 1
Salmon net-pen facilities in Puget Sound
(Adapted from James, unpub.; Rensel, unpub.)

Map No. Name Location Facilities
1 Sea Farm of Port Angeles Culture of full-size Atlantic
Norway salmon (2-5 kg). Currently with
24 pens. Permitted for 50 pens.

2 Scan Am Fish Farms Deepwater Bay, Culture of Atlantic salmon.
Cypress Island Originally located near Hadlock,
Jefferson Co.

3 Cypress Salmon Deepwater Bay, Culture of coho and Atlantic
Cypress Island salmon. Recently acquired by
Scan Am.

4 Olympic Seafarm Deepwater Bay, Currently culturing coho.
Cypress Island Culture of Atlantic salmon planned.

5 Skagit System Swinomish Channel Currently culturing coho and
Cooperative Atlantic salmon. One net-pen and
several seawater raceways.

6 Dirk Nansen and Port Blakely, Culture of coho salmon. Recently
Tom Hamilton Bainbridge Is. received county permission for
culture of Atlantic salmon at
Port Townsend.

7 Passage Silver Fort Ward, Recently permitted for 16 pens
Bainbridge Island of coho salmon.

8 Domsea Farms Clam Bay, Manchester Culture of pan-sized coho
salmon. 160 net-pens moored
in a single 4 acre array.

9 Domsea Farms Fort Ward, Culture of pan-sized coho.
Bainbridge Island

10 NMFS Clam Bay, Salmon held for research purposes-
Manchester only. No commercial culture.

11 WDF Hale Passage, Coho and chinook salmon
Fox Island reared for fisheries enhancement.
No commercial culture.

12 Squaxin Island Squaxin Island Culture of coho to market size
Seafarm and for ocean ranching. Began
Atlantic salmon culture in 1986.
Experimental steelhead culture.

Five major and eight minor net-pen facilities used by Indian tribes or
sportsmen's clubs for delayed release of salmon are not shown on Figure 1
nor listed individually here.

Water depth - The water depth should be sufficient to maintain the net-pen
above the bottom throughout the tidal cycle in order to promote the dispersal
of feces and excess feed. A clearance of at least 1-5 m beneath the net-pen
has been suggested to minimize the possibility that these wastes could affect
the health of the cultured fish (Kennedy, 1978; STOWW, 1974; Sutterlin and
Merrill, 1978). Maximum depth is determined by mooring considerations.

Wind and waves - Areas sheltered from severe wind and wave action are preferred
net-pen locations. Some of the earlier net-pen designs could withstand waves
of up to I m in height (Kennedy, 1978; STOWW, 1974); net-pen systems capable of
withstanding much greater wave heights are now available.

Dissolved oxygen - A supply of dissolved oxygen adequate to meet the respira-
tory needs of the fish is necessary. A dissolved oxygen concentration of 6-7
mg. 11C is generally regarded as a minimum acceptable level (STOWW, 1974;
Sutterlin and Merrill, 1978; Zook, et al., unpub.).

Temperature - optimum growth of salmon occurs at temperatures of 10-15 0C. A
range of 5-16 0C is acceptable (Kennedy, 1978; STOWW, 1974; Sutterlin and

Merrill, 1978).

Salinity - There are no generally accepted standards for salinity in net-pen
culture. Pacific and Atlantic salmon have been reared successfully in both
marine and freshwater.

Pollution - Net-pens should be located far from polluted waters, such as indus-
trial outfalls, densely populated areas or similar areas where water quality

could threaten the health or marketability of the fish.

Permitting - Net-pen applications are being subject to increasing public
scrutiny, particularly at the local level. Sites must be selected to comply
with local zoning restrictions and state environmental policies, and to mini-

mize conflict with other water uses.

Accessibility - If an operation is to be economically successful, the site must
be readily accessible for the transport of staff, supplies, feed, and fish.

Currently, commercial net-pen operators in Puget Sound culture either coho

(Onchorhynchus kisutch) or Atlantic (Salmo salar) salmon. While culture of

chinook salmon (0. tshawytscha) is possible, the species is more susceptible to

disease than the coho and less willing to accept pelleted dry feed (Mahnken,

1975) . Steelhead (Salmo gairdneri) culture in net-pens is currently in an ex-

perimental stage at the Squaxin Island net pens in a cooperative venture be-

tween the Squaxin tribe and the Washington Department of Fisheries.

Fish are held in net enclosures with 10-30 mm mesh, depending on the size

of the fish. The dimensions of the enclosure vary among facilities. A small

net-pen may be only a few meters on each side, whereas a large net-pen may be

up to 12x12 m square. The depth of the net-pen is typically 2-4 m. At most

facilities many net-pens are moored together to form a single large unit, with

the individual pens separated by walkways.

Eggs are hatched and the fry reared in a freshwater facility. Salmon

smolts are generally transferred to the net-pens in the spring of each year.

The timing of this transfer is deliberately spread over several months in order

to provide a more consistent supply of product. The fish are held in the net-

pens until they reach marketable size. Stocking density in net-pens depends on

several factors (e.g., fish size) but is typically about 15 kgm- 3in Puget
Sound net-pen operations. Fish intended for the "pan-sized" market are held

6-11 months until they attain a weight of about 0.3 kg. Fish intended for the

"full-size" market are held 18-24 months, attaining a weight of 2-5 kg.

Salmon may be fed on a variety of diets including pelleted dry or moist

feeds and wet feed (e.g., minced fish). Growers in Puget Sound typically use a

pelleted dry diet. The feed may be provided either by hand, through demand

feeders (feed delivery triggered by fish behavior) or through automatic feeders

(feed delivered at preset times and in predetermined quantities). Salmon are

typically fed 2% of their body weight per day, although this may vary by an

order of magnitude depending on size of the fish and water temperature. A feed

conversion ratio of 2:1 is commonly achieved in Puget Sound, meaning that two

kg of feed are required to produce I kg of salmon.

The net-pen operator must protect against loss of fish through both di-
sease and predation. Mortality due to disease can be minimized by proper

husbandry practices (e.g., reducing stocking density), vaccination or addition

of therapeutic agents (e.g., antibiotics) to the feed. Predators include

birds, otters and marine mammals. Birds and otters are excluded by the in-
stallation of nets over the top of the net-pen. Predation by seals and sea
lions may be prevented by use of a double net around the cultured fish.
Nonlethal seal and sea lion deterrents, such as acoustic devices, are also
occasionally used.

2.2 SHELLFISH CULTURE

Most suspended shellfish culture in Puget Sound is dedicated to the pro-
duction of the blue mussel, Mytilus edulis. With the exception of a few small
operations, oysters are cultured on bottom tracts rather than by suspended
techniques. Suspended scallop culture in the Sound is still In the experi-

mental stage, and is not yet commercially feasible.
Culturing of mussels in the Sound began in the mid-1970s by Penn Cove
Mussels. Although Penn Cove Mussels is still the largest firm, five other
firms have entered the market (Figure 2, Table 2). Much of the culture ac-
tivity is concentrated around Whidbey Island, particularly in Penn Cove,
although there are culture operations in both the San Juan Islands and in the

southern Sound.
Mussel culture of mussels may be conducted on the bottom, as in the Neth-
erlands, on posts, as in the "bouchot" culture of France, or on strings sus-
pended from a rack which rests on the bottom. Suspended culture from rafts or
longlines has proven to yield the greatest production per unit area. In Spain,
where mussels are grown suspended from rafts, a maximum production of 500
metric tons per hectare (in English units, 250 tons per acre) has been
achieved. This figure is 1,000 times greater than any other form of aqua-
culture in which animals are grown without supplemental feeding (Ryther, 1969).
Suspended mussel culture in Puget Sound involves rafts or longlines. The
size of these structures varies depending on the operator. Using Penn Cove
Mussels as an example, rafts at this facility measure 10x12 m, and longlines
are 91 m long. The mussels strings suspended from these structures extend

about 7 m below the surface of the water. At the Penn Cove Mussels facility up
to 400 strings may be hung from a single raft or longline, and the unit will
typically support 9 metric tons of mussels (P. Jeffards, Penn Cove Mussels;

pers. comm.).

BEL LINGHAM

RCAS ISLAN

'A �>QfIlA

SrRA~r OF SIAN D JUCANkS4

* ~~~~~~~~~~~~.. - ISLAND

.'EERETT

Kr. LYMPIA

Figure 2 -Location of mussel culture facilities in Puget Sound

Table 2
Mussel culture facilities in Puget Sound

Map No. Name Location Facilities
1 Scott McCullough Lopez Island One raft and two longlines
as of 1984. Production that
year of about 1000 kg.

2 Penn Cove Mussels Penn Cove, Twelve longlines and four
Coupeville rafts. Annual production
approx. 50,000 kg.

3 West Coast Blue Penn Cove, Nine rafts
Mussels Coupeville

4 Sea Forms Penn Cove, Ten acre site with twelve rafts.
Coupeville

5 Race Lagoon Mussels Race Lagoon, Seven acre intertidal site
Whidbey Island with longlines.

6 Kamilche Sea Farms Totten Inlet Small facility - no specific
data available.

Mussel seed for Puget Sound mussel culture comes entirely from natural
populations. There are very few places in the Sound with a good, reliable seed
set; Penn Cove is one of the best areas. The empty mussel strings are placed

in the water in June or July, the major period of mussel set in Puget Sound.
Following collection of the mussel seed, the strings may be either left In
place or transferred elsewhere in the Sound for growth. The mussels are grown
for 6-10 months until reaching a harvestable size, and may remain on the
strings for almost 2 years until the last of a given set is harvested.
Throughout the culture period it is necessary to periodically lift the strings
from the water for removal of fouling organisms and thinning of the mussels.

3.0 POTENTIAL ENVIRONMENTAL EFFECTS

3.1 WATER CIRCULATION

Question

Water circulation is critical in minimizing sedimentation or promoting
removal of metabolic wastes from a mariculture operation. To what extent will
structures such as net-pens or mussel rafts reduce the current velocity in the

culture structure and in the surrounding area?

Discussion

Effects on water circulation within the culture structure -

Any structure placed in a moving fluid will impede the flow of that fluid.
The reduction in current flow induced by a net-pen is dependent upon the size
and shape of the pen, the diameter of the twine used in the net, and the size
and type of mesh (e.g., knotted or knotless, diamond or square). Textural

differences in the type of material used for net construction can also affect
current velocity (Milne, 1970). The effects of variables such as stocking
density of fish and extent of fouling growth on the net are difficult to quan-
tify, but they can have a dramatic impact on the resistance of the net-pen to
water flow. Even the swimming movements of the fish can alter flow patterns
(Hisaoka, et al., 1966).
There are several empirical studies of the effect of net-pens on water
circulation; the most extensive being that of Inoue (1972). Current velocities
inside floating net pens stocked with yellowtail were found to be 35-81% of the
current velocities outside the pens. In addition, equations were developed to
estimate the effect of multiple pens placed parallel to the current. As would
be expected, current velocity was reduced with passage through each succesive
pen. For pens having dimensions of 3x3x3 m, with 2.4 cm mesh and without fish,
it was estimated that current velocities would be reduced to 70-80% of initial
velocities after passage through the first pen, and reduced to 10-25% of ini-
tial velocities after passage through three pens. Comparable results were
obtained with various stocking densities, cage sizes and mesh sizes.

Hisaoka, et al. (1966) measured current velocities associated with float-
ing net-pens stocked with yellowtail in Daio Bay, Hiroshima Prefecture. Cur-
rent velocity was reduced inside a net-pen with a 5 mm mesh to an average of
60% of' upcurrent velocity, while a 3 cm mesh reduced inside velocity to 70% of
outside velocity. These measurements were made after the net had been placed
in the water only a few days prior to the investigation, thus providing insuf-
ficient time for growth of fouling organisms. Thus, the reduction in current
velocities was attributable entirely to the net and the enclosed fish. Further
reduction in water flow could be expected with growth of fouling organisms on
the net.
Reduction of current velocity has also been observed in the vicinity of
shellfish culture operations. Arakawa, et al. (1971) measured current veloc-
ities amidst culture strings of the Pacific oyster, Crassostrea gigas, under a
raft in Hiroshima Bay. Within the raft current velocities were generally
12-14% of the velocity outside the raft, but 1 m below the lower end of the
oyster strings there was no consistent effect on current velocity.

Effects on water circulation beyond the culture structure -

No in situ measurements are available on alteration of water flow in the
environment surrounding mariculture facilities, but principles of fluid dy-
namics can be used to estimate the magnitude of this effect. Much of the
available information pertains to flow around solid objects rather than a
porous structure, such as a net-pen or shellfish raft, however, a rough approx-
imation of effects is possible. A large structure placed in the marine envir-
onment will reduce current velocity upcurrent of the facility for a relatively
short distance. For a solid object this upstream distance is equal to about
twice the dimension of the structure perpendicular to current flow (hereafter,
a "diameter") (Nowell and Jumars, 1984). For a porous structure, such as a
net-pen, the extent of upstream influence is probably on the order of about 1
structure diameter. To either side of the structure perpendicular to the
current flow, alteration of flow can be observed for 2-5 diameters from a solid
body (Eckman and Nowell, 1984; Nowell and Jumars, 1984). Since a mariculture
facility is porous and since the effect of a structure on current velocity is a
logarithimic function with distance, current velocity should be about 95X of
the free stream value at distances of approximately 2 structure diameters. The

influence of a structure on current velocity is measurable for the greatest
distance downcurrent. Solid bodies will typically influence current velocity
for up to 50 diameters downcurrent. For a porous structure current velocity
should return to 95% of the mean unaffected value within 20 diameters down-

current (A. Nowell, Univ. Washington, pers comm.). As an example of the poten-
tial areal extent of influence, if a net-pen 12x12x4 m were moored with a 12 m
face perpendicular to the direction of water movement, there may be alteration
of the flow field 12 m upcurrent of the pen, 24 m on either side and 240 m
downcurrent. Multiple net-pens moored together would function as a single
larger body, and the effects on circulation must be scaled upwards accordingly.
These figures must only be considered as order-of-magnitude approximations,
since important variables such as mesh size and extent of fouling are unspec-
ified. It is also important to recognize that current flow patterns are rarely

uniform and unidirectional as is implied in this discussion. The complex flow
patterns actually found in most estuarine environments severely complicate
efforts to quantify the effects of a mariculture structure.

Sedimentation -

One potential effect of a reduction in current velocity is sediment de-

position in the area of the mariculture structure. Sedimentation resulting
from current velocity reduction as discussed here is considered as distinct
from sedimentation directly attributable to culture practices such as the de-

position of feces and excess feed, potential effects of mariculture which are
discussed separately in Section 3.4. The ability of flowing water to transport
sediments is largely determined by current velocity. A reduction in current
velocity may allow the sediment particles to settle out of suspension with the
larger, denser particles settling first. This potential effect is not likely
to be of significant magnitude for floating mariculture facilities in Puget
Sound. First, floating structures are much less prone to problems associated
with sedimentation than are bottom culture operations. Unless the structure is
located close enough to the bottom to restrict flow of water beneath it, the
current beneath the structure should prevent sedimentation of particles attrib-
utable to impedance of water flow by the structure. Secondly, the suspended
sediment load of most Puget Sound waters is very low. Sedimentation would be
expected to be greater in areas of major riverine inflow; areas that are

generally unsuitable for mariculture for other reasons (e.g., phytoplankton

blooms, variable salinity).

Circulation and water quality -

A reduction in current velocity caused by mariculture structures may af-
fect water quality by preventing adequate supply of oxygenated water or inhib-
iting removal of metabolic wastes. The cultured organism is typically the
first to show the effects of inadequate water exchange. Chinook salmon held in
Squaxin Island net-pens experienced heavy mortalities when fouling of the nets
prevented adequate exchange of water. Dissolved oxygen concentration in the
pens dropped to as low as 2 mg.1-1 while concentrations outside the pens were

6.5-9 mg.-l (STOWW, 1974). When the nets were replaced the dissolved oxygen
concentrations in the pens returned to near ambient levels. Thus, it is in the
operator's best interest to minimize reduction in current speeds caused by his
operation. A detailed discussion on mariculture effects on water quality is

presented in Section 3.2.

Conclusions

Like any structure placed in the marine environment, a mariculture facil-
ity will alter water circulation. Within a net-pen itself, reductions in
current velocity to 35-81% of the upcurrent velocity have been reported. There
is also a "wake" associated with a mariculture facility which alters current
patterns and velocities in the vicinity of the structure. The effect of this
wake is observable in all directions from the structure but its influence
extends for the greatest distance downstream. In some situations the effect of
a mariculture structure may be considered beneficial if the facility functions
as a breakwater, dampening wave action on inshore structures.
The reduction in current velocites induced by a mariculture structure
could potentially promote sedimentation of suspended particulate matter or en-
hance degradation of water quality as a result of poor flushing. However,

deposition of the suspended sediment load in the vicinity of floating marn-
culture facilities is likely to be of negligable importance in most of Puget
Sound. The effect of reduced current velocity on water quality could be sig-
nificant under certain conditions, such as a high density of mariculture units

in an embayment with very restricted natural circulation. Such mariculture
development is unlikely given the need of the operator to provide good water
quality for the cultured organisms.
Operators of mariculture facilities can take certain measures to minimize
the effect of their structures on water circulation. The fluid dynamics dis-

cussion above has implications for mooring arrangement of multiple pens or
rafts. In strong currents, multiple pens or rafts should be arranged parallel
to the direction of water flow. Not only would this arrangement minimize the
alteration of water circulation, but it would simplify mooring requirements.
Under weak current regimes, such a parallel arrangement would be undesirable
because of potential reductions in water quality for those pens or rafts far-
thest downcurrent. Under these conditions, the individual units would be
better arranged perpendicular to current flow with the optimal spacing between
pens or raf ts of 2 diameters or more. This spacing will minimize each unit's
effects on water circulation, and consequently water quality, in neighboring
units. These guidelines should be considered as optimal configurations strict-
ly from the standpoint of minimizing effects on water circulation. Local

conditions or operational considerations may dictate alternative arrangements.
Other mitigative measures which an operator might take include wide spac-
ing of the strings in shellfish culture, use of the largest mesh possible for
net enclosures, and frequent cleaning to remove fouling organisms. In most
cases these measures would be in the best interest in the operator to minimize
water quality degradation which could affect the health of the cultured
organisms.

3.2 WATER QUALITY

Questions

How much organic matter and nutrients are introduced into the water during
culture activities in the form of unutilized feed and metabolic wastes? Could
these inputs result in a deterioration of water quality in the vicinity of the
culture site, including high concentrations of substances such as of ammonia,
nitrates and phosphates and reduced levels of dissolved oxygen?

Discussion

Characterization of wastes -

Dissolution of substances from unutilized feed and the metabolic breakdown
of ingested food both result in some alteration of water quality. The general

inputs are:

Ammonia - Ammonia is an end-product of protein metabolism and the principal

nitrogenous excretory product of marine and freshwater organisms, both verte-

brates and invertebrates. It is also released from sediments by the anaerobic
microbial decomposition of nitrogen compounds. Ammonia can be present in ei-
ther the ionized (NH4 ) or unionized (NH ) form. The unionized form, which
comprises 2% or less of the total ammonia under typical marine conditions of
temperature and pH, is the more toxic of the two.

Nitrite and Nitrate - These compounds are produced by microbially mediated con-
version of other nitrogenous compounds. Elevated concentrations of nitrite

(NO2) and nitrate (NO3 )might be expected in the vicinity of mariculture fac-
ilities due to bacterial oxidation of ammonia.

Organic Nitrogen - Urea is typically the principal component of this group.

Like Aiminnia, it is produced as an end-product of protein metabolism.

Phosphate - Phosphate is excreted in urine. A portion of the phosphate input
from salmon culture facilities is also the result of leaching from feces and

food.

Oxygen - Mariculture reduces dissolved oxygen in the water through both respir-
ation by the cultured organisms and consumption of oxygen by decomposition of

feces and excess feed. The latter utilization is quantified as the biochemical
oxygen demand (BOD).

Waste loadinig from salmonid culture -

Much work has been attempted to quantify the nutrient load from freshwater
fish hatcheries. In contrast, work in the marine environment is extremely lim-
ited: I can find no measurements of nutrient or BOD loading from marine net-
pen culture. Freshwater systems have a number of characteristics that often
make them more susceptible to adverse effects of nutrient enrichment, and
therefore, these systems have been the subject of more attention. Freshwater
systems are often characterized by high stability of the water column, lack of
tidal-induced flushing, and culture discharge volumes which are relatively

large in proportion to the volume of the total water body. Freshwater systems
are also very different from marine environments in that the availability of
phosphorus is often the factor which limits phytoplankton production. Nitrogen
is typically the limiting nutrient in marine systems.
A summary of estimated nutrient and BOD loadings from freshwater salmonid
hatcheries and net-pen aquaculture facilities is presented in Table 3. Load-
ings have been quantified on the basis of the weight of fish, annual fish pro-

duction and amount of feed provided. This analysis does not discriminate
between dissolved nutrients which will directly influence water quality and
those nutrients which reach the bottom as feces or excess feed. The vast ma-
jority of the nitrogenous wastes will be released in the dissolved form, where-
as most of the phosphorus will be incorporated into the particulate material
(Enell and Ldif, 1983b). Most of the ROD is generally associated with the solid
material (Speece, 1973; Willoughby, et al., 1972).
It is readily apparent from Table 3 that loadings are extremely variable
among or even within studies. This variation is attributable to many factors
including type of feed provided, quantity and frequency of feeding, tempera-
ture, fish size, and proportion of suspended solids in the effluent. This
variability makes it very difficult to predict the nutrient and ROD loading
from any given facility. Attempts to estimate potential loadings from proposed
facilities have been frustrated by the highly divergent results obtained

Table 3
Loading of nutrients and BOD from freshwater salmonid culture

Organic Total Total
Reference Ammonia Nitrite Nitrate Nitrogen Nitrogen Phosphate Phosphorous BOD

Loading expressed as g.kg- fish-day-1

Bergheim and 0.3-0 -- 0.05 1.6-4.6a
Selmer-Olsen, 1978

Bergheim, et al., 0-1.3 -- -- -- 0.13-3.8 0.02-0.27 0.005-0.43 1.6-2.7a
1982

Butz and -- -- --1.4-8.1b
Vens-Cappell, 1982

Clark, et al., 0.3-0.8 -- 0.13-0.21 -- -- 0.067-0.17
1985

Korzeniewski, 0.032 -- 0.053 0.040 0.12 0.033 0.10
et al., 1982

VKI, 1976c 0.125 -- 0.063 -- 0.38 0.05 0.1 1.8b

Loading expressed as g-kg- fish produced-yr1

Alabaster, 37-180 ---- 0-548e .... .. . 22-110 510-990b
1982

Ackefors and -- - 71f 1.9f
Sodergren, 1985

Ackefors and -- -- -- -- 87g -- 13.5g
Sodergren, 1985

Penczak, et al., -- -- -- -- 100 -- 23
1982

Solbe, 1982 55.5 1.8 10.2 -- -- -15.7 285b

Warrer-Hansen, 45 -- -- -- 83 -- 11 350b
1982

Table 3 continued

Organic Total Total
Reference Ammonia Nitrite Nitrate Nitrogen Nitrogen Phosphate Phosphorus BOD

Loading expressed as g.kg-1 feed provided-day-

Butz and -- -- -- 80-300b
Vens-Cappell, 1982

Clark, et al., 31-37 -- 9-15 -- -- 5.2-5.9 ----
1985

Liao and Mayo, 0.3-4.0 -- 0.06-15.0 -- -- 0.04-0.94 --6.4-25.0b
1972

Liao and Mayo, 28.9 -- 24 -- -- 16.2 -- 600b
1974

Querellou, 26-34 0.04-7 3.8-118 -- 9-64 2.6-19 43-75 33-121b
et al., 1982
n _
Speece, 1973 22-31 -- -- -- -- -- 400

Willoughby, 32 -- 87 -- -- -- 340b
et al., 1972

BOD7

Unspecified whether value is BOD5 or BOD7.

c Based on values for dry feed.

Daily loadings of Alabaster (1982) multiplied by 365 days-yr-�

e Sum of nitrite and nitrate

Dissolved fraction only.

g Total of dissolved and particulate fractions.

h Ultimate oxygen demand (UOD).

depending on the method of calculation employed (Tynan, unpub.). It is clear
that loading values from any one aquaculture operation can not be used to ac-
curately predict loadings from another. Thus a priori estimates are reliable
only as order-of-magnitude approximations of actual values.

Waste loadings from mussel culture -

The discussion of nutrient loading has focused on quantifying the loading
from fish culture in which the nutrients are added to the system through the
feed provided to fish. Mollusc culture also has associated nutrient loads, but
in this case such nutrient inputs are derived from the indigenous phytoplankton
and particulate matter, rather than an external food source. Thus, the culture
operation does not introduce any "new" nutrients into the marine environment,
but only promotes the recycling of those which are already present. There is
actually a net decrease in nutrient levels in the system since only about 40%
of the total nutrients removed by the mussels are released directly to the
water column, 30% fall to the bottom as particulate material and 30% are re-
moved in the harvest (Ackefors and Grip, 1985; Ackefors and Sbdergren, 1985).
The nitrogenous excretory products of mussels are composed primarily of
ammonia and amino-nitrogen, with ammonia accounting for 40-90% of the total
excreted nitrogen (Bayne and Scullard, 1977; Bayne, et al., 1976; Kautsky and
Wallentinus, 1980). Oysters excrete a comparable percentage of ammonia, but
the organic nitrogen component is urea rather than amino-nitrogen (Hammen, et
al., 1966). Excretion rates of the mussel Mytilus spp. have been estimated to
be 1.4-75.5 ug inorganic N (Kautsky and Wallentinus, 1980), 5-40 ug NH4-N
(Bayne and Widdows, 1978 for 1 g mussel) and 2-30 ug NH4-N (Bayne and Scullard,

1977), all on the basis of ug N (or NH4).g shell free dry weighthr . The
wide variation in excretion rates within and among studies is primarily a re-
sult in differences in mussel size and study season, both of which dramatically
affect nutrient production.
Mussels excrete phosphorus primarily as phosphate. Phosphate excretion
rates of 0.4-24.1 ug P04-P (Kautsky and Wallentinus,1980) and 0.5-3.6 ug P04-P
(Kuenzler, 1961), both on a ug P04.g shell-free dry weight-hr- basis, have
been reported.

Comparison of mariculture loadings with other discharges -

In order to put the changes in water quality induced by mariculture in
perspective with other discharges to the marine environment, Tables 4 and 5
contrast nutrient and BOD releases from mariculture facilities with the re-
leases from several rivers, wastewater treatment plants and industries in the
Puget Sound area. Concentrations and loadings from mariculture have been based
on expected releases from relatively large facilities. In the case of salmon
net-pen culture, calculations are based on a facility with 50 net-pens, a hold-
ing capacity of 250,000 kg salmon and a current velocity of 8-16 cm-sec- I out-

side of the nets at peak ebb tide. Data presented for mussel culture are based
on a facility with 12 longlines supporting 108,000 kg of mussels. The reader
is referred to the notes following the tables for an explanation of how the
data were derived.
Several observations are readily apparent from Tables 4 and 5. First, the
flow rate of water through either a net-pen or a longline system far overshad-
ows most other discharges. The water flow through a net-pen is 5 orders of
magnitude above that from a seafood processor and one order of magnitude above
the flow of the West Point treatment plant. Only a major river system has a
flow rate of comparable magnitude. Secondly, as a result of this heavy water
usage the concentrations of nutrients and BOD from mariculture facilities are
relatively small in comparison to the other discharges shown. This would sug-
gest that near-field effects (e.g., effluent toxicity) would be much less of a
problem at a mariculture facility than near the other outfalls. Thirdly, the
nutrient load from a mussel culture is considerably less than from net-pen
culture, and in fact is lower than all other discharges shown. It would seem
unlikely that significant water quality deterioration would be found around any
mussel culture of a size typical of Puget Sound facilities. Finally, the rela-
tive nutrient/ROD contribution of net-pen culture in comparison to the other
discharges is highly variable depending on the particular parameter. In gen-
eral, the nutrient or ROD loading from the net-pens is comparable to that of a
small river or small dairy. The ROD load from the net-pens is comparable to
that of the LOTT wastewater treatment plant but the nutrient load of the net-
pens is considerably less.

Table 4
Concentration and loading of nitrogen compounds from various dischargers in the Puget Sound area

Flow Ammonia Nitrates Organic N Total N
3 -1 -1 ) Z1 )1 -L1
Discharge Source (m3 se1) (mg.l ) (kg-day) (mgl ) (kg-day) (mg 1) (kg.day) (mg-l 1) (kg'day )

Riverine discharge
Chambers Creek 3.44 0.037 11 1.53 450 ND ND ND ND
(near Steilacoom)

Stillaguamish River 112 0.043 360 0.27 2600 ND ND ND ND
(near Silvana)

Skagit River 496 0.02 857 0.15 6400 ND ND ND ND
(near Mt. Vernon)

Wastewater treatment plants
LOTT (Olympia) 0.39 14.4 490 2.6 88 1.7 57 19 650

Renton 1.9 14.8 2400 0.04 6 2.9 490 39 6400

West Point (Seattle) 5,0 16 6900 ND ND 7 3024 ND ND

Mariculture
Salmon net-pen 72 0.01 75 0.004 25 0.002 10 0.02 150

Mussel culture 20 0,003 4.4 ND ND ND ND ND ND

ND = No data available

Table 5
Concentration and loading of phosporus and BOD from various dischargers in the Puget Sound area

Flow Phospate Total Phosphorus BOD

(m3 sec 1) (mg.l 1) (kg.day 1) (mg.1l 1) (kg-day 1) (mg- 1 ) (kg-day -)
Riverine discharge
Chambers Creek 3.44 0.02 6 0.05 15 ND ND
(near Steilacoom)

Stillaguamish River 112 < 0.01 < 100 0.03 290 ND ND
(near Silvana)

Skagit River 496 < 0.01 < 430 0.03 1300 ND ND
(near Mt. Vernon)

Wastewater treatment plants
LOTT (Olympia) 0.39 4.7 160 4.97 170 16.6 560

Renton 1.9 2.79 460 3.41 560 9.4 1600

West Point (Seattle) 5.0 ND ND 5.1 2200 87 38,000

Industries
Universal Seafood 0.00055 ND ND ND ND 300 14
(whitefish and crab proc.)

Perfection Smokery 0.00068 ND ND ND ND 975 57
(salmon processing)

Vitamilk Dairy 0.0031 ND ND ND ND 2975 800
(milk processing)

Consolidated Dairy 0.0076 ND ND ND ND 2500 1600
(milk processing)

Rainier Brewing 0.017 ND ND ND ND 2925 4300
(beer brewing)

Mariculture
Salmon net-pen 72 0.004 25 0.004 25 0.1 750

Mussel culture 20 0.001 2.2 ND ND ND ND

Notes to Tables 4 and 5

1) Nitrite loadings are not shown since no reliable values are
available for mariculture, and most other discharges shown either
have no data on the parameter or found concentrations consistently
below detection limits.

2) Riverine discharges - The data presented are from the WDOE water
quality monitoring program and were obtained through Robert James of
WDOE. The data are average values of monthly measurements taken from
January 1984 through February 1986. The rivers shown were selected
so as to encompass a wide range of flow rates. Chambers Creek has
the lowest discharge rate of any riverine discharge to Puget Sound
monitored by WDOE, while the Skagit River has the largest.

3) Wastewater treatment plants - The data presented include:
Lacey, Olympia, Tumwater, Thurston County - Grand mean of the mean
values for 1984 and 1985. Data provided by Asha Mhatre.
Renton - Mean of monthly measurements from January 1984 through
February 1986. Data provided by Richard Finger.
West Point - Mean of daily ROD and weekly nutrient measurements from
January 1984 through December 1985. Organic nitrogen obatined by the
difference between ammonia nitrogen and total Kjeldahl nitrogen
(APHA, 1981). Data provided by Barry Uchida.

4) Industries - All industries listed discharge to Puget Sound via
the Metro wastewater treatment system. Industry types which would be
expected to have an organically enriched waste stream were selected,
and then a range of plant sizes were chosen within each industry
type. Flows represent the mean of the discharge rates over the last
four quarterly measurements. BOD values are the mean of all
measurements made since the last major process changes at the
facilities. No data are available for water quality parameters of
concern here other than ROD. Data provided by Bruce Burrow,
Municipality of Metropolitan Seattle.

5) Salmon mariculture - The data presented are for a hypothetical
net-pen operation, but are based on the permitted capacity of the Sea
Farm of Norway, Port Angeles facility, a moderately large facility in
comparison to other net-pen operations in Puget Sound. Plans call
for expansion of the facility to 50 pens of 12x12x4 m in size. A
maximum probable holding capacity of 250,000 kg of salmon has been
assumed (J. Lindbergh, pers. comm.). In order to simplify the pres-
entation and because of the uncertainty as to the alignment of the
pens in relation to prevailing current at the site, it is assumed
that the net-pens are arranged in a single row oriented perpendicular
to the current direction.
Current veiocity at peak ebb tide at the Sea Farm site ranges
from 8-16 cm-sec (existing and proposed locations - Milner-Rensel
Assoc., 1986). Recognizing that the current velocity will be lower
during other periods of the tidal cycle, and that the net-pens will
further reduce currents in the pen to about one half of their

upcurrent velocity (Inoue, 1972), a current velocity of 3 cm-sec
through the pens has been assumed. With 50 pens, each having a
surface area of 12x4 m perpendicular to the direction of water flow,
the Nolume1of water which passes through the pens can be estimated as
72 m -sec .
Loadings of nutrients ant BOD have been estimated from Table 3
values on the basis of g.kg fish.day . Based on the range of
loading values reported, "typical" loadings have been estimated as:
ammonia - 0.3; nitrate - 0.1; organic nitrogen - 0.04; total nitrogen
- 0.6; phosphate - 0.1; total phosphorus - 0.1; and BOD - 3. These
estimates should not be regarded as anything other than order-of-
magnitude approximations. Concentrations shown on the tables were
back-calculated from loadings and flow. Phosphate concentrations and
loadings for mariculture are based on total phosphate-phosphorus,
whereas data from the riverine discharges and wastewater treatment
plants were reported as ortho-phosphate-phosphorus.

6) Mussel culture - The data presented are for a hypothetical mussel
culture operation, but are are based on the longline culture of Penn
Cove Mussels; the largest in Puget Sound. The facility is permitted
for 12 longlines, each 91.4 m long. Each longline is comprised of
400 strings, 7.3 m in length, and supporting an average mussel weight
of 9,000 kg (P. Jeffards, Penn Cove Mussels, pers. comm.). Given 12
longlines, the entire longline system could support 108,000 kg of
mussels. Of the 9,000 kg on each longline, approximately 15% of the
weight will be comprised of barnacles and other fouling organisms.
No effort has been made to exclude the nutrient contribution of these
organisms from the calculations, and it is assumed that they will
release nitrogen and phosphate at the same rate as the mussels.
Lacking data on curjent velocity in the midst of the longlines,
an estimate of 3 cm-sec has been used as for the salmon net-pens.
The longlines are assumed to be oriented perpendicular to the direc-
tion of current flow and placed directly behind one another. Thus,
the surface area of the longline system on a plane perpendicular to
the flow is 91.4x7.3 m.
Loadings are based on "typical" values given the ranges reported
in the literatuje (see text). Ammonia and phosphate loadings of
20 and 10 ug.g shell-free dry weight-hr , respectively, have been
assumed. Concentrations shown have been back-calculated from load-
ings and flow.
To convert shell-free dry weight to wet weight with shell, the
following factors can be used for Mytilus edulis: shell-free dry
weight is 21% of dry weight with shell; dry weight with shell is 40%
of wet weight with shell (Lappalainen and Kangas, 1975).

Oxygen consumption through respiration -

Oxygen consumption by cultured organisms and the decomposition of waste
products could potentially affect water quality by decreasing dissolved oxygen
concentrations. Salmonid respiration rate depends upon fish size, age, sex,
activity, and temperature but an average value for routine metabolism is about

300 mg 02.kg-1 wet weight.hr-1 (Kils, 1979; Liao and Mayo, 1972). Mytilus
edulis respiration rates are about 0.1-3 mg 02 .kg- shell-free dry weight-hr-
(Kautsky and Wallentinus, 1980). Applying salmon and mussel respiration rates
2 -I -I
of 300 and 1.5 mg 0 .kg .hr ,respectively, to the culture facilities dis-
cussed in Tables 4 and 5 (250,000 kg of salmon and 108,000 kg of mussels), it
2 -1I
can be estimated that salmon respiration may consume 21,000 mg 0 2sec and the
2 -
mussels may consume 45 mg 0 sec This respiration would result in a reduc-
tion of 0.3 and 0.002 mg 02.1-1 in the oxygen content of the water which passed
through the salmon and mussel culture facilities, respectively. Thus, a meas-
urable change in dissolved oxygen would be anticipated around the net-pen cul-
ture, although with dispersion and dilution by the surrounding water concen-
trations should rapidly return to ambient levels. No measurable effect of the
mussel culture on dissolved oxygen concentrations would be anticipated.
These estimates of oxygen consumption do not include the BOD of the feces
and other metabolic wastes, which may consume about 1.5-3 times as much oxygen
as respiration (Kalfus and Korzeniewski, 1982; Liao and Mayo, 1974; Willoughby,
et al., 1972). Since most of the BOD is associated with the solid wastes
(Speece, 1973; Willoughby, et al., 1972), most of the oxygen required to meet
the BOD will be provided by bottom waters rather than the water which passes
directly through the culture structure.

Field studies of effects of salmon net-pen culture on water quality -

Schools of wild fish have been shown to alter the chemistry of water bod-
ies through which they pass. For example, increased ammonia and decreased
dissolved oxygen concentrations have been observed around schools of menhaden
(Oviatt, et al., 1972). Thus, it is not suprising that mariculture has been
shown to alter water quality. In freshwater environments, both lakes and
rivers, there are many examples of water quality changes in the vicinity of
fish hatcheries or net-pens. Increased concentrations of ammonia, organic and

total nitrogen, phosphate, total phosphorus, and BOD and decreased dissolved
oxygen have been observed (Bergheim and Selmer-Olsen, 1978; Beveridge, 1984;
Eley, et al., 1972; Hinshaw, 1973; Korzeniewski, et al., 1982b; Mantle, 1982).
In marine waters measurable changes in water quality from fish culture are
typically not as common or as great as in freshwater. Effects have been noted
in some cases, such as an 8-9 fold increase in ammonia concentration around

net-pens in Norway (Ervik, et al., 1985). Dissolved oxygen concentrations
-1
0.2-2.5 mg 02'1 less than ambient were observed around yellowtail and sea
bream net-pens in Japan, although immediately beyond the boundaries of the cul-
ture area there was no oxygen depletion (Kadowaki and Hirata, 1984; Kadowaki,
et al., 1978a). However, alterations in marine water quality have not been
reported with the same regularity as in freshwater. The principal reasons for
differences between fresh and salt water systems are the generally greater
degree of water-column mixing (and therefore, dilution of wastes) in marine
systems, and the fact that in freshwater systems a large proportion of the
water body is often routed through the culture operation (e.g., 70% of river
flow in Bergheim and Selmer-Olsen, 1978). The dilution capacity of Puget Sound
was clearly evident in studies of the Metro West Point sewage discharge. This

discharge, which has ammonia concentrations over 1,000 times greater than the
salmonid net-pen operation presented in Tables 4 and 5, increases the ammonia
concentration of the receiving water for a distance of only 1.5 km (Collias and
Lincoln, 1977).
Most studies in Washington and British Columbia have found little or no
observable effects of mariculture on water quality. Two water quality surveys
were conducted in 1982 in the vicinity of the Domsea Farms and NMFS net-pens in

Clam Bay, Puget Sound (Anonymous, 1983a; D. Damkaer, NMFS, unpub. data). The
Domsea Farms net-pen operation is by far the largest in Puget Sound. In 1982
it produced over 1 million kg of salmon per year (J. Lindbergh, pers. comm.).
Despite the number of fish present, neither survey found any observable changes
in turbidity, dissolved oxygen, particulate or dissolved nitrogen, or particu-
late or dissolved carbon that could be attributed to net-pen activities (Anon-
ymous, 1983a).
In April of 1986 the Sea Farm of Norway, Port Angeles facility held ap-
proximately 27,000 kg of salmon (J. Forster, Sea Farm of Norway, pers. comm.).
During a water quality survey done at that time ammonia concentrations inside

the net-pens averaged 0.020 mg 1-, approximately three times the concentra-
tions found 30 m upcurrent (0.007 mg-l- ), but well below levels considered
toxic to biota. Within 30 m downcurrent of the net-pens ammonia concentrations
were 0.012 mg.1-1 (Milner-Rensel Assoc., 1986). No changes in ortho-phosphate,
silica, nitrite or nitrate concentrations were observed, even within the net-
pens themselves.
Water quality parameters have been monitored around four net-pen facili-
ties in Sechelt Inlet, British Columbia. During a period of minimal tidal and
wind mixing, elevated ammonia concentrations were observed around all net-pens
and for a distance of at least 27 m from the sites. One net-pen contained no
fish but increased ammonia concentrations were found in the surrounding water,
suggesting that fouling organisms on the nets were at least partially respon-
sible for the increase in ammonia levels (E. Black, British Columbia Ministry
of the Environment, pers. comm.).
Net-pens in Henderson Inlet, Puget Sound have been shown to increase the
ammonia concentrations and decrease the dissolved oxygen in surrounding waters
(Pease, 1977). This effect was limited to periods of water column stratifica-
tion and reduced mixing during the summer months. No measurable differences in
water quality between reference and culture areas were found during the spring,
fall or winter. The Henderson Inlet site could be regarded as a worst case
situation since it was located in a poorly flushed area, with current veloc-
ities further reduced by surrounding log rafts. During a 24-hour observation
period there was no measurable surface or bottom current in the vicinity of the
culture area.

Field studies of effects of mussel culture on water quality -

No water quality data exist in conjunction with mussel cultures in Puget
Sound, but information from a mussel farm in Sweden can serve as a close paral-
2
lel. The Swedish culture area occupied 4500 m and was designed to produce up
to 200,000 kg Mytilus edulis every two years. Water depth was 10-15 m and the
average current velocity in the culture area was 0-2 cm-sec1 (Hagstrom and
Larsson, 1982; Loo and Rosenberg 1983; Mattson and Linden, 1983; Rosenberg and
Loo, 1983). Ammonia concentrations were two times greater and phosphate con-
centrations were four times greater in water samples taken among the longlines
than in upcurrent samples. However, no changes in nutrients or dissolved

oxygen were seen in samples taken about 100 m outside of the culture area
(Hagstrom and Larsson, 1982; Larsson, 1985).
Similar water quality results were reported by Kaspar, et al. (1985) in
their study of a New Zealand mussel farm. No differences in nutrients or other
water chemistry parameters were found in comparisons of the culture site with a

reference area. High current velocities (up to 110 cm-sec-) at the site were
responsible for the rapid dilution of metabolic wastes.

Conclusions

The culture of fish and molluscs results in release of nutrients and con-
sumption of dissolved oxygen. A net increase in environmental nutrient levels
may be expected in salmonid culture because of nutrient input in the form of
feed. Culturing molluscs requires no addition of feed, so no input of "new"
nutrients to the marine ecosystem results. However, molluscs do enhance the

recycling of nutrients in the water column as they ingest phytoplankton and
return a portion of the nutrients to the water column and sediments, making
those nutrients available to primary producers again. The filtering activities
of the molluscs also serve to concentrate nutrients from a wide area into the
area of the culture (Ryther, 1969).
The effects of mariculture facilities on the water quality in the vicinity
of the culture can be summarized as shown in Table 6. This table indicates the
percentage change in several water quality parameters which could be expected
in the parcel of water which passes directly through a large salmon net-pen and
mussel longline culture. It is evident that the only parameters of potential
concern are ammonia and dissolved oxygen from salmon culture and only ammonia
from mussel culture.
It is possible to compare the concentrations of ammonia and dissolved oxy-
gen in the vicinity of the culture to those levels which are considered toxic
to aquatic life. Permissable levels of total ammonia are generally reported to

range from 0.5-2.5 mg.1-l (Ellis, 1948: Liao and Mayo, 1974), and it is obvious
from Table 6 that the ammonia concentrations downcurrent of the culture facili-
ties would not approach toxic levels. The toxicity of ammonia is usually at-
tributable to the toxicity of the unionized fraction (NH3), which at summer
temperatures (12�C) and a pH (8.0) typical of Puget Sound comprises approxi-
mately 2% of the total ammonia (Trussel, 1972). The maximum acceptable concen-
tration of unionized ammonia for the protection of aquatic life is 0.02 mg'l'

Table 6
Effects of mariculture on water quality

Salmon culture Mussel culture
Concentrations Effect of Effect of
in culture culture

Puget Sounda facilityb Changec facilityb Changec
Parameter (mg.-l 1) (mg.11) (%) (mg.1-1) (%)

Ammonia 0.002-0.04 +0.01 25-500 +0.003 8-200

Nitrate 0.3-1.9 +0.004 0.2-1 ND ND

Phosphate 0.5-3d +0.004 0.1-0.8 +0.001 0.03-0.2

Dissolved oxygen 6-14 -0.3 2-5 -0.002 0.01-0.03

a~~~~~~~~
a Annual range in surface waters of the Main Basin from Collias and Lincoln
(1977) and Friebertshauser, et al. (1971). For local conditions the reader is
referred to Collias, et al. (1974), Friebertshauser, et al. (1971) and any
available site specific studies.

b Data obtained from text and Tables 4 and 5 based on a 250,000 kg salmon net-pen
culture and a 108,000 kg mussel culture. Values only apply given the current
velocity and other conditions specified in the notes provided with Tables 4 and 5.

c Estimated change in water quality in the water mass which passes directly

through the culture structure. The effects of dilution with the surrounding
water are not considered.

d The Puget Sound concentrations provided are ortho-phosphate; total phosphate
concentrations would be equal to or greater than this value. Therefore, the
percent change attributable to salmon and mussel culture may be overestimated.

(Environment Canada, 1979; Roberts, 1978; U.S. EPA, 1976). Based on Table 6
the maximum concentration of unionized ammonia downeurrent of the net-pens
would be 0.001 mg1l I; far less than the toxic threshold. Unionized ammonia
concentrations downcurrent of the mussel longlines would be even lower.
Minimum acceptable concentrations of dissolved oxygen have been reported

to range from 5-6 mg1l (Washington Department of Ecology, 1973; U.S EPA,
1976). The net-pen facility on which Table 6 is based caused a relatively

small decrease in dissolved oxygen (0.3 mg-l , 2-5) uigtesme
months dissolved oxygen concentrations in Puget Sound surface waters decrease

to about 6 mg-l I, a value which is at or close to the lowest acceptable limit.
Therefore, the further decrease attributable to the net-pens could be signifi-
cant if there is little or no mixing of the water which flows through the net-
pens with the surrounding water mass. Inadequate mixing should be a signif-
icant problem only in very enclosed embayments.
It was demonstrated earlier that the nutrient and/or BOD loading from a
net-pen facility is generally comparable to that of a small river or small milk
processing plant. Loadings were greater than several seafood processors in the
Puget Sound area and considerably less than that of a major brewery. In con-
sidering mariculture permit requests in Puget Sound, much attention has been
given to comparisons of the water quality effects of mariculture with those of
sewage discharge. In particular, nutrient and ROD loadings from proposed mari-
culture have been expressed in terms of their "population equivalents", and the
siting of the proposed facility evaluated in terms of a comparable sewage dis-
charge. For example the 250,000 kg salmon net-pen facility on which the load-
ings in Tables 4 and 5 are based would have loadings of nitrogen, phosphorus
and BOD equivalent to untreated sewage from approximately 10,000 persons. The
108,000 kg mussel culture facility would have a loading equivalent to approxi-
mately 500 persons. While there may be some utility in this type of compari-
son, putting mariculture effects into some perspective with respect to a more
familiar type of discharge, the determination of a population equivalency tends
to reduce the issue to a single number, an oversimplification which may be
misleading and unwarranted.
There are several factors which complicate comparisons between a sewage
discharge and that of a culture facility. First, most wastewater treatment
plants discharge industrial wastes or storm water along with domestic sewage.
The effluent contains heavy metals, synthetic organics and other toxicants

which pose a threat to the marine environment distinct from the effects of
organic enrichment. With the possible exception of trace quantities of anti-
biotic drugs used for disease treatment (Section 3.10), the discharge from a
mariculture faciliity contains no such toxicants. Secondly, sewage contains
viral and bacterial pathogens which represent a public health hazard. Many
state and federal effluent standards and restrictions on other water uses
(e.g., shellfish harvesting) are a consequence of this health risk. In addi-
tion, the required disinfection of sewage results in the discharge of addition-
al pollutants (e.g., chlorine). Thirdly, during primary treatment the large
particulate material is removed from the sewage discharge so that the nutrient
burden and resulting BOD are primarily associated with the dissolved phase. In
contrast, most of the BOD and phosphates from a mariculture facility are assoc-
iated with the solid wastes such as excess feed and feces. Thus, many of the
effects of mariculture are more localized and gradual release of nutrients from
the sediments may dilute effects over a longer time period. Finally, a sewage
discharge is freshwater. Thus, depending on discharge conditions, the fresh-
water may promote stratification of the water column and inhibit mixing of the
effluent with the receiving waters.
Aquaculture may use more water per unit of product than any other manufac-
turing process (Muir, 1985; Warrer-Hansen, 1982). The quantities of nutrients
and associated BOD from the culture of 250,000 kg of salmonids (example dis-
cussed above) are comparable to those produced by about 10,000 persons, but the
water which passes through the net-pens on a daily basis is equivalent to the
3 -I -I
domestic use of 25 million persons (based on 0.25 m *day *individual ; Muir,
1985). Thus, the concentrations of nutrients and BOD are very dilute compared
to sewage and most other discharges to the marine environment.
Concerns that metabolites would increase or dissolved oxygen decrease to
the point that water quality would be threatened beyond the immediate vicinity
of mariculture operations in Puget Sound are generally unwarranted. Such con-
ditions would be anticipated only in areas of extremely limited flushing or if
culture intensity reached a magnitude far greater than can be expected in Puget
Sound given the many competing water uses. In the few cases where measurable
water quality changes have been noted, the effects have been largely confined
within the culture structure. The cultured organisms would therefore be likely
to first experience any toxic effects. Degradation of water quality resulting
in toxic effects beyond the bounds of the culture area is unlikely in most of
Puget Sound.

3.3 PHYTOPLANKTON

Questions

Could the input of nutrients associated with mariculture increase phyto-
plankton biomass and productivity, thereby inducing the occurrence of blooms
which would have adverse effects on other water uses (e.g., shellfish harvest-
ing)? Could culture of filter-feeding molluscs decrease phytoplankton standing
stock to the point that inadequate food resources would be available to the
natural suspension-feeding community?

Discussion

Enhancement of phytoplankton productivity -

Phytoplankton require nutrients, particularly nitrogen and phosphorus, for
their growth and reproduction. In marine systems phosphorus is typically pres-
ent in concentrations which exceed the demands of the phytoplankton, so the
availability of nitrogen is one of the principal factors which often determines
the growth rate. Mariculture provides nitrogen to the ecosystem principally in
the form of ammonia, which may be used directly by the phytoplankton or indi-
rectly following bacterial transformation of the ammonia to nitrate. In salmon
culture, the feed provided to the fish causes a net increase of nitrogen in the
system. In mollusc culture, there is a net decrease in total nitrogen because
it is removed in the form of harvested animals.
Molluscs can, however, enhance phytoplankton productivity by increasing
the rate of nutrient cycling; ingesting phytoplankton and then excreting a
large portion of these nutrients, making them available again for use by other
phytoplankton. The potential for stimulation of productivity by mussel culture
has often been suggested (Campos and Marifio, 1982; Kaspar, et al., 1985; Tenore
and GonzAlez, 1975). It has been demonstrated in the laboratory (Rosenberg and
Loo, 1983), but at least one study failed to find any effect of mussel culture
on productivity in field measurements (Hagstr5m and Larsson, 1882).
A phytoplankton bloom occurs when physical, chemical and biological fac-
tors interact to provide conditions ideal for a rapid burst of phytoplankton
growth. Blooms are a common occurence in most coastal and estuarine marine

environments including Puget Sound. Intense blooms usually appear in the Main
Basin in late April or early May and recur throughout the summer months, each
bloom lasting about 1-2 weeks (Winter, et al., 1975). Most blooms pass un-
noticed by Puget Sound residents, however, blooms of the diatom Chaetoceros sp.
and the dinoflagellate Ceratium fusus have been responsible for fish mortal-
ities and the subsequent closure of some net-pen operations in Puget Sound. Of
greater public concern are blooms of Gonyaulax catenella, the organism respon-
sible for paralytic shellfish poisoning (PSP).
Mariculture and other sources of nutrients (e.g., sewage discharges) have
been implicated as potentially contributing factors in phytoplankton blooms
outside of Puget Sound. While the input of nutrients may help sustain a bloom,
it is important to recognize that this input may not necessarily trigger a
bloom. In order for a bloom to occur many environmental factors must interact
to provide suitable conditions for rapid phytoplankton growth. The availabil-
ity of an adequate nutrient supply is one of these factors, but of equal im-
portance are light, temperature, low wind velocities, reduced tidal mixing, and
many other factors related to water column stability. The many interrelated
environmental factors which trigger a bloom remain too poorly understood to
permit prediction of bloom events. It is often tempting to implicate an ob-
vious nutrient source without adequate understanding of hydrographic conditions
and recognition of the many factors involved. Along Florida's west coast
blooms were originally thought to be attributable to runoff, pollution and
other land-derived nutrient inputs. Research activities were thus concentrated
in the nearshore zone. Only after several years of research and observation
was it found that the blooms originated 18-74 km offshore and moved shoreward
(Steidinger and Haddad, 1981).
Although not conclusively demonstrated, mariculture may be responsible, in
part, for phytoplankton blooms observed in other countries, with Japan being
the most notable example. Arakawa (1973) correlated phytoplankton blooms with
the culture of oysters in Hiroshima Bay. The frequency of the blooms closely
paralleled historical trends in oyster production within the bay, however, it
should be noted that correlations in time are often spurious. Bottom muds from
shellfish culture have been found to accelerate the growth of red-tide organ-
isms held in nutrient-free laboratory culture (Takagi, et al., 1980). Labora-
tory studies have also implicated yellowtail feces as a potential contributing

factor in phytoplankton blooms (Nishimura, 1982), and yellowtail culture oper-
ations have been adversely affected by blooms, with consequent production
losses. Although the causative factors have not been clearly demonstrated, the
Japanese have found phytoplankton blooms appearing with greater frequency than
in the past. In an effort to minimize the enrichment of nutrients in the
culture areas, Japanese net-pen operators have been reducing the use of wet
feed (which has a high percentage of waste) and moving culture operations from
enclosed, well-protected bays into open coastal waters (Nose, 1985). Shellfish
growers have also been forced to reduce culture densities (Takagi, et al.,
1980).
Outside of Japan there are few reports of mariculture potentially contrib-
uting to phytoplankton blooms. At one site in Ireland (Doyle, et al., 1984) a
bloom occurred that was localized around a culture operation, and the fish cul-

ture operation was believed to be a contributing factor. However, at another
farm a phytoplankton bloom developed offshore and then moved into the culture

area. Mortalities of the cultured fish occurred in both instances.
Field studies in Puget Sound (Weyerhaueser net-pens in Henderson Inlet,
Pease, 1977; Domsea and NMFS net-pens in Clam Bay, D. Damkaer, NMFS, unpub.
data) have failed to demonstrate any effect of mariculture on phytoplankton.
Chlorophyll a concentrations (a measure of phytoplankton biomass) were measured
in both studies and no effect attributable to the culture facilities was iden-
tified.
Under physical conditions typical of most of Puget Sound, it would be
difficult to demonstrate any localized effect of mariculture on phytoplankton
populations based on field sampling. Individual phytoplankton cells require
several hours to divide even at rapid reproduction rates, and whole populations
require a day or more to double in number (Parsons, et al., 1977). Phyto-
plankton populations require several days to increase to bloom proportions
(Strickland, 1983). Therefore, by the time phytoplankton populations might be
expected to show measurable responses to the nutrient enrichment in a parcel of
water, that parcel is far from the culture site, and nutrient concentrations
have been diluted many times over by mixing with the surrounding waters. Only
in a static system such as a lake or an area with very poor flushing and inten-
sive culture, would any measurable effect on productivity be expected.
There are several other reasons to expect Puget Sound mariculture to have
minimal effect on phytoplankton productivity. As noted earlier, the unavaila-
bility of nitrogen is often a factor limiting phytoplankton growth in most

marine systems. However, in much of Puget Sound nitrogen is usually present in
concentrations which exceed the utilization capabilities of phytoplankton.
Phytoplankton growth is instead often limited by low light levels and instab-
ility of the water column (Campbell, et al., 1977; Collias and Lincoln, 1977;
Welch, et al., 1972; Winter, et al., 1975). Thus, additional input of nut-
rients already present in relatively high concentrations will have little or no
measurable effect on primary productivity. It is only in areas with a high
degree of water column stratification because of freshwater inflow or minimal
tidal mixing would addition of nutrients be expected to influence productivity.
The danger of drawing direct parallels between the embayments of Japan and
Puget Sound is demonstrated by examining nutrient levels in the Japanese cul-
ture areas. The Japanese culture areas are very poor in nutrients in compari-
son to Puget Sound waters, and thus additional nutrient input may have dramatic
effects on phytoplankton productivity in those low nutrient areas. For exam-
ple, various Japanese investigators have recommended upper limits for nutrient
concentrations which, if exceeded, result in high potential for phytoplankton
blooms. Murakami (1973) found that oyster production began to decrease when
the average annual nitrate concentration exceeded 0.1 mg.-l. It has been rec-
ommended that total nitrogen not exceed 0.1 (Iwasaki, 1976) and 0.35 mg.l1
(Takagi, et al., 1980). In contrast, nitrate concentrations in the surface
waters of the Main Basin of Puget Sound are typically about 1 mg1-I (Collias
and Lincoln, 1977), far in excess of the recommended upper limits in Japan.
While nitrogen is abundant and not limiting to phytoplankton growth in the
Main Basin, it should not be assumed that these conditions are found throughout
the Sound. Certain areas of the Sound have been identified where stratifica-
tion of the water column does occasionally permit depletion of available nit-
rogen (Yake, unpub.). These areas are Drayton Harbor, Bellingham Bay, Port
Susan, Port Gardner, Possession Sound, Penn Cove, Holmes Harbor, Saratoga Pas-
sage, Hood Canal (lower three-fourths), Sequim Bay, Port Orchard, Dyes Inlet,
Sinclair Inlet, Liberty Bay, Carr Inlet, Case Inlet, Budd Inlet, Eld Inlet,
Totten Inlet and Oakland Bay. In these regions, nitrate concentrations in
-1
surface waters are reduced to less than 0.1 mg.l for 2-3 months during the
summer.
It is possible to obtain a crude estimate of the potential effects of
mariculture on phytoplankton by estimating the maximum potential increase in
phytoplankton standing stock should all nutrients supplied by a net-pen culture

be used for phytoplankton growth. Tett (1981) estimated that about 3 mg nitro-
gen are required to produce 1 mg of chlorophyll, and Gowen, et al. (1985) used
this factor to estimate the potential effect of salmon net-pen culture in
Scotland. Applying the same calculation procedure to Puget Sound, we note that
the net-pen operation described in Tables 4 and 5 increased the total nitrogen
concentration of the water which passed through the pens by 0.02 mg'l- (or
20 mg-m 3). Assuming that nitrogen is a limiting nutrient, and that all of the
nitrogen released from the net-pens was used for phytoplankton growth, the
phytoplankton standing stock in the water passing through the net-pens could
increase by about 7 mg chlorophyll am 3, or about 30 mg chlorophyll a m2
vertically integrated over the 4 m depth of the net-pens. Phytoplankton stand-
ing stock in the Main Basin of Puget Sound is typically about 10-30 mg chloro-
phyll a-m-2 between blooms and in excess of 100 mg chlorophyll a.m-2 during

blooms (Strickland, 1983; Winter, et al., 1975 - standing stock expressed as a
vertically integrated value through the water column from the surface to the 1%
-2
light depth). The additional 30 mg chlorophyll a-m attributable to net-pen
culture would not initiate a bloom, and, in fact, would be completely unmeas-
urable given the natural spatial variability of phytoplankton biomass. These
calculations have assumed no horizontal mixing, an implausible assumption given
the time required before the phytoplankton population could fully utilize these
nutrients. During this time period the enriched water would be thoroughly
mixed with surrounding water under most conditions, and the increase in stand-
ing stock would be negligible. A bloom might be initiated or sustained only if
the net-pens were placed in an enclosed and poorly-flushed embayment where di-
lution and dispersion were much reduced, or if local geomorphology and hydro-
graphy caused the same water mass to repeatedly pass through the culture area
with each tidal cycle.

Reduction in phytoplankton standing stock -

Mussels and oysters feed by filtering seston (i.e., suspended particulate
matter, both plankton and detritus), and thus the biomass of phytoplankton
downcurrent of a mollusc culture site may be expected to be reduced. The
effectiveness with which molluscs remove particulate material from the water is
evident in the work of Imai (1971) (shown in Table 7). The concentration of
seston in the water was measured after passage through oyster rafts, each

Table 7
Cumulative reduction in seston concentration with passage through
oyster rafts: Kesen-numa Bay, Japan. (from Imai, 1971)

Percentage of initial seston concentration
Raft Number June - Sept Sept. - Nov.

Initial seston concentration 100 100
outside culture ground

Passing through:
Raft 1 88.9 77.4

Raft 2 78.9 60.0

Raft 3 70.1 46.4

Raft 4 62.3 36.0

Raft 5 55.4 27.8

Raft 6 49.2 21.6

Raft 7 43.7 16.7

Raft 8 38.8 12.9

Raft 9 34.5 10.0

Raft 10 30.7 7.7

Raft 11 27.2 6.0

Seston concentration 24.2 4.6
inside culture ground

supporting 50,000-90,000 oysters. The seston concentration was reduced 76-95%
after passage through 11 rafts (although some of this loss may be attributed to
passive settling rather active removal by the oysters).
There is also anecdotal evidence of the filtering efficiency of mussels.
Growers have found that mussels on the upcurrent side of a raft grow faster
than those on the downcurrent side, presumably because of the reduction in food
concentration as the water passes through the raft (Mason, 1976). Divers have
reported the water clarity in the midst of mussel strings to be substantially
improved over conditions beyond the perimeter of the raft, assumedly because
the mussels have reduced the concentration of suspended material (Pease and

Goodwin, unpub.).
While definitive data are lacking from Puget Sound, experiences from other
mussel culture areas suggest that culture intensities anticipated in Puget
Sound are far below the intensities that could be sustained. Changes in seston

concentrations in the vicinity of the culture sites are therefore not likely to
limit feeding of other organisms. Rosenberg and Loo (1983) examined the ef-
fects on seston of a mussel longline culture operation in Sweden that produced
200 metric tons of mussels every two years. On the basis of field experiments
and theoretical calculations of seston concentration, current velocity and fil-
tering efficiency it was estimated that the seston concentration was adequate
to support a culture operation of twice the existing size. It was also consid-
ered unlikely that the culture would remove sufficient quantities of seston so
as to impoverish other hard-bottom epibenthic animals in the vicinity.
The Ria de Arosa estuary in Spain has a surface area of 230 km2 (slightly
smaller than the area of Hood Canal), an average depth of 19 m and a phyto-
plankton biomass and productivity comparable to that of the Puget Sound Main

Basin (Tenore and Gonzalez, 1975). The Ria is an area of intensive shellfish
culture with 2,000 mussel rafts and 200 oyster rafts, with the established
culture areas occupying 10% of the surface area of the entire estuary. The Ria
produces about 100,000 metric tons of mussels per year. In comparison the
annual production of mussels in all of Puget Sound is between 50 and 100 metric
tons - the yield of one or two rafts in the Ria de Arosa. It has been estim-

ated that the mussels filter at least 80% (and at times twice this) of the
water volume of the Ria every day (Tenore, et al., 1982). Yet the seston con-
centration is adequate to support not only the needs of the mussels but the
demands of a dense epifaunal community growing on the rafts. Extrapolating

these observations to Puget Sound suggests that any reduction in seston
concentration in the vicinity of the relatively (in comparison to the Ria de
Arosa) small culture operations in Puget Sound would probably not be a limiting
factor to the naturally occuring filter-feeding organisms.

Conclusions

Floating mariculture is unlikely to have measurable effects on phyto-
plankton standing stock or productivity in most of Puget Sound. With the
exception of some stratified areas, such as some of the inlets of southern
Puget Sound, there is little evidence that nutrients ever limit phytoplankton
growth in much of the Sound. Even the West Point discharge, which releases
about 100 times more ammonia and about 300 times more organic nitrogen than a
large net-pen operation, has not been found to affect productivity, in part,
because growth is limited by factors such as light and water column stability
rather than by nutrient concentrations.
In areas where nutrients are limiting to phytoplankton growth, any input
of nutrients would, by definition, stimulate phytoplankton growth. However, in
most areas dilution of the nutrients released from a mariculture facility would
minimize any potential localized effects on productivity. I calculated that,
even if all nitrogen provided by a large net-pen operation to the water body
passing through or under the pens were utilized by phytoplankton for growth,
the phytoplankton biomass in this parcel of water would only be increased by
-2
about 30 mg chlorophyll a~ m . This increase in standing stock would not con-
stitute a bloom, and, in fact, would be undetectable given the spatial and tem-
poral "noise" of phytoplankton populations. Moreover, given the substantial
dilution that would occur in the time before phytoplankton could use these
nutrients, the actual increase in biomass would be negligible. A localized,
measurable effect would be expected only if the culture operation were sited in
a highly stratified and poorly flushed area where that parcel of water could
retain its integrity for an extended period of time, or if the same parcel of
water were repeatedly passed through the culture area with each tidal cycle.
The potential for changes in phytoplankton community composition as a
result of mariculture activities has not been addressed. However, if observ-
able changes in phytoplankton biomass and productivity are unlikely, observable

changes in community composition seem unlikely as well. Based on observations
in other marine areas, it also seems unlikely that any reductions in seston
concentrations caused by the filter-feeding activities of mussels and oysters
would limit the natural suspension-feeding community in Puget Sound.

3.4 SEDIMENTATION

Questions

Will solid wastes produced by mariculture operations, such as fecal mate-
rial and excess feed, sink to the bottom and accumulate under the pens? In
the event of this accumulation of organic matter, what physical and chemical
changes will occur in the substratum? (The biological consequences of these
changes are addressed separately in Section 3.5.)

Discussion

Quantity of solid waste produced by net-pen culture -

The culture of salmonids in Puget Sound generally results in the produc-
tion of two types of waste: 1) excess food which passes through the pen with-
out being ingested by the fish; and 2) fecal material. Smaller quantities of
waste result from the sloughing of fouling organisms growing on the nets and
associated structures.
The quantity of unutilized feed generated by mariculture is highly vari-
able And depends to a great degree on culture practices including feed type,
feeding methods and frequency. Wasteage is dependent upon type of feed used
since differences in consistency affect capture efficiency and differences in
digestibility affect utilization of the feed ingested. Sedimentation rates
measured beneath net-pens vary up to 4-fold depending on the type of food
supplied to the fish (fresh anchovies vs. frozen mackerel; Kadowaki, et al.,
1980). For land-based trout farming, a wasteage of 1-5% has been estimated for
dry feed, 5-10% for semi-moist feed and 10-30% for wet feed (VKI, 1976). Esti-
mates of feed wasteage for net-pen rearing of trout or salmon in Europe include
15-20% (dry feed - Gowen, et al., 1985), 20% (primarily moist feed - Braaten,
et al., 1983), and 30% (mixture of dry and wet feeds - Penczak, et al., 1982).
The salmon culturists currently operating in Puget Sound typically use dry

feed, which has a relatively low wasteage. It is estimated that no more than
5% of the feed is wasted in Puget Sound net-pen culture (C. Mahnken, National
Marine Fisheries Service, pers. comm.).

Fish defecation is the second major source of organic matter to the sedi-
ments. Butz and Vens-Capell (1982) have shown that under laboratory conditions
25-38% of dry food ingested by rainbow trout is lost as feces. If these data
are applied to net-pen culture of salmon and it is assumed that about one-third
of the feed ingested is lost as feces, it is possible to estimate the quantity
of solid waste which may be generated by a net-pen culture. Food conversion
ratios for salmon raised in Puget Sound are typically about 2:1; the produc-
tion of 1 kg of salmon requires 2 kg of feed. Assuming dry feed is used (the
usual practice) approximately 0.1 kg (5%) of feed will be lost as waste with
the remainder, 1.9 kg, ingested by the fish. With one-third of the ingested
feed lost as feces, the ingestion of the 1.9 kg feed will result in the pro-
duction of 0.6 kg of fecal material. The total solid waste (excess feed and
feces) produced per kg of salmon will be about 0.7 kg (dry weight). A large
operation (e.g., 250 metric tons of salmon per year) will generate about 175
metric tons of solid waste per year.

Quantity of solid wastes produced by mussel culture -

Mussel production generates solid wastes in the form of feces and pseudo-
feces (material which has been filtered from the water by the animal but not
ingested). Shells which fall from the culture structure also accumulate on the
bottom immediately under the raft or longline. Studies of oyster culture in
Japan indicate that the amount of solid waste produced by shellfish culture can
be considerable (Arakawa, et al., 1971; Ito and Imai, 1955; Kusuki, 1977a). A
raft of oysters in Hiroshima Bay holds 350,000-630,000 oysters. During a nine
month culture season a single raft will produce 16 metric tons (dry weight) of
feces and pseudofeces, with an additional 4.5 tons attributable to feces of
fouling organisms growing on the rafts. Approximately 20-30% of this material
is deposited on the bottom under the raft.
Fecal and pseudofecal production by Mytilus galloprovincialis ranges from
14.3-149.3 mg dry weight.day -lindividual -1 (Arakawa, et al., 1971). If a mean
rate of 68 mg-day -lindividual-1 is assumed for Mytilus edulis grown commer-
cially in Puget Sound, it can be estimated that a typical raft supporting
700,000 individuals (P. Jeffards, Penn Cove Mussels, pers. comm.) will produce
17 metric tons of feces and pseudofeces per year.

Sedimentation rates -

Many investigators have attempted to quantify the rate of solid waste pro-
duction from fish and shellfish culture facilities by placing sediment traps
beneath the pens and rafts. Sedimentation rates beneath net-pens have been
estimated for cultures of rainbow trout (Enell and Lof, 1983a; 1983b; Merican
and Phillips, 1985; Phillips, et al., 1985), yellowtail (Hata and Katayama,
1977; Kadowaki, et al., 1978b; 1980), Atlantic salmon (Ervik, et al., 1985) and
Pacific salmon (Pease, 1977). Sedimentation rates beneath mussel rafts or
longlines have been measured by DahlbAck and Gunnarson (1981) and Tenore, et
al. (1982). Sediment traps are of questionable value in obtaining absolute
measures of sedimentation rate since it is difficult to distinguish freshly
deposited material from resuspended sediments, and the amount of material
collected in a trap is highly dependent on trap size and shape (Butman, in
press; Butman, et al., in press). Sediment traps are useful, however, in
obtaining estimates of sedimentation relative to a reference area with a com-
parable current regime. Sedimentation rates beneath net-pens have typically
been found to be 2-10 times greater than in reference areas (Enell and Lbf,
1983a; 1983b; Pease, 1977). Sedimentation rates as high as 100-200 times rates
in reference areas have been reported under net pens in some lakes (Merican and
Phillips, 1985; Phillips, et al., 1985). Rates of deposition beneath mussel
longlines have been observed to be 2-4 times those in reference areas (Dahlback
and Gunnarson, 1981).

Fate of solid wastes -

The fate of solid wastes produced by a culture operation depends in large
part on the current velocity, water depth and sinking speed of the material.
The greater the current velocity or water depth, the greater the probability
that the wastes will be dispersed. Dry feed which sinks more quickly than the
other feed types will tend to accumulate immediately under the pen (Pedersen,
1982). Investigations of floating mariculture facilities have generally found
significant accumulation of solid wastes beneath the pens, rafts or longlines.
This deposit typically consists of a soft, flocculent layer on top of the nat-
ural sea bottom (Kaspar, et al., 1985; Morrison, unpub.; Pease and Goodwin,
unpub.; Tenore, et al., 1982). Table 8 shows the extent of this accumulation

Table 8
Sediment accumulation beneath salmon net-pens or mussel culture facilities

Distance of
Visible
Waterb Depth of Accumulation
Location Animal Depth Current Accumulation from Facility
(Reference) Cultured (m) Velocity (cm) (m) Comments

Weyerhaueser Co. Coho salmon 5-7 Very weakc NDd At least 15 Sediments under pens
Henderson Inlet, (2-4) covered by Beggiatoa
Puget Sound
(Pease, 1977)

Scan Am Farms Atlantic 20 "Relatively 2-8 3-5 Sediments under pens
Hadlock, salmon (13) strong" covered by Beggiatoa
Puget Sound
(Kyte, unpub.;
pers. comm.)

NMFS Chinook and 9-13 5-20 3-4 30
Clam Bay, coho salmon (0-8) cm-sec()
Puget Sound
(Lindbergh, 1972)

Domsea Coho salmon 11-17 5-20(e) 2- 15 Sediments under pens
Clam Bay, (7-14) cm.sec covered by Beggiatoa
Puget Sound
(Pease, unpub.)

Domsea Coho salmon 8 "Strong" see see Isolated pockets of fish
Fort Ward, (4) comments comments feces and Beggiatoa
Puget Sound around rocks and other
(Pease, unpub.) protected areas. Visible
up to 46 m from pens.

Suquamish tribe Coho salmon 8 "Weak to see directly Fish feces and Beggiatoa
Agate Pass, moderate" comments under pens in small, scattered
Puget Sound patches. Site used for
(Pease, unpub.) delayed release; fish not
in pens continuously

Table 8 continued

Distance of
Visible
Watera b Depth of Accumulation
Location Animal Depth Current Accumulation from Facility
(Reference) Cultured (m) Velocity (cm) (m) Comments

Nansen/Hamilton Coho salmon 6 ND 5-8 Only directly Deposit consisting
Port Blakely, (0) under pens primarily of excess
Puget Sound feed
(D. Nansen,
pers. comm.)

Sea Farm of Norway Atlantic 23 8 cm sec-1 0 0 Isolated small patches
Port Angeles, salmon (19) of Beggiatoa
Str. Juan de Fuca
(Milner-Rensel
Assoc., 1986)

Penn Cove Mussels Blue 11-13 "Very weak" 7-30 Shell debris only Sediments under rafts
Penn Cove, mussel directly under covered by Beggiatoa
Puget Sound rafts. Organic
(Pease and enrichment extends
Goodwin, unpub.) less than 37 m.

Hotham Sound, Salmon 21 "Good 15 ND Sediment under pens
British Columbia (12) flushing" covered by Beggiatoa
(Morrison, unpub.)

Nanaimo, Pacific ND ND ND < 20
British Columbia salmon
(Kennedy, 1978)

Table 8 continued

Distance of
Visible
a
Water b Depth of Accumulation
Location Animal Depth Current Accumulation from Facility
(Reference) Cultured (m) Velocity (cm) (m) Comments

Storeb0, Atlantic 10 ND 5-10 ND Significant accum-
Norway salmon ulation of sediment
(Braaten, et al., after 3 years of
1983) operation

Western Norway Atlantic 16-61 ND up to 37 ND
(Ervik, et al., salmon
1985)

Tjarno, Blue 8-13 "Weak- 10-15 ND Beggiatoa under long-
western Sweden mussel (2-7) typically1 lines. 10-15 cm
(Dahlback and 3 cm-sec sediment accumulation
Gunnarson, 1981) developed in 16 months.

Total water depth (at mean lower low water if available) and depth of water beneath lowest point of culture
structure (in parentheses).

bMany studies lack quantitative data on current speeds. The authors' qualitative description is provided if
available.

cNo measurable surface or bottom current in the culture area during a 24-hr. period of observation.

ND = No data available.
e -
Current velocity from Kramer, Chin and Mayo (undated). Reported to reach 100 cm-sec-1 at times
(Lindbergh, 1972).

based on a review of the available literature. Several conclusions can be
drawn from Table 8:

1) Most sites examined were in relatively shallow water with less than 15 m
(and often less than 10 m) between the bottom of the culture structure and the
sea floor, thereby increasing the likelihood of observable solids accumulation.

2) Many sites were located in areas of relatively weak currents decreasing the
probability of solids dispersal. Most investigators provided no quantitative
data on current speeds. This is a result of the fact that many of the studies
identified in Table 8 were simply diver observations from a single survey.

3) Some accumulation of wastes associated with the culture operation was meas-
ured at all but one site examined. The accumulations ranged in depths from
2-37 cm. Even at the one site with no visible accumulation of solid wastes
(Sea Farm of Norway), the presence of -Beggiatoa spp., the odor of hydrogen
sulfide in one sample and the elevation of ammonia concentrations in intersti-
tial water indicated some organic enrichment of bottom sediments. It may be
significant that the Sea Farm of Norway facility has been in operation only 1.5
years. It is unknown if greater accumulation will occur with continued op-
eration.

4) Visible accumulation of wastes was localized, extending at most 46 m from
the facility and typically much less.

5) White patches on the sediment surface consisting of colonies of Beggiatoa
spp. bacteria were observed at most sites. This organism provides a readily
visible indicator of organic enrichment of the sea floor.

The extent of solids accumulation is dependent on current velocity as well
as depth, and both must be considered in project siting, but some of the inves-
tigators identified in Table 8 have used their observations of solids accumula-
tion to suggest minimum depths for location of net-pens. Morrison (unpub.)
recommended that pens not be located in less than 9 m of depth at low water,
with a minimum of 3 m between the net and the sea floor. Braaten, et al.
(1983) suggested that, in general, there should be a clearance of at least 10 m

beneath the net-pen, and furthermore suggested that sites with strong currents
(as indicated by a sand or gravel bottom) be selected. Table 8 suggests that
localized solids accumulation is very likely if net-pens are located where
water depth under pens is less than 15 m, and it should be noted that solids
accumulation is possible at even greater depths. There are insufficient data
provided by Table 8 to suggest optimal current velocities, although obviously
the greater the current velocity, the less will be the extent of solids accumu-
lation (Hata and Katayama, 1977).
Relatively little is known of the time required for observable accumula-
tion of solids under culture operations or for dispersal of such accumulations
following cessation of culture. Measurable accumulation (1-2 cm) of wastes did
not develop until the third year of culture at one Norwegian net-pen site, but

continued accumulation was rapid thereafter (Braaten, et al., 1983). After 16
months 10-15 cm of material had accumulated beneath mussel longlines in Sweden
(Dahlback and Gunnarson, 1981).
The rate of removal of wastes after cessation of culture may be relatively
rapid. Pamatmat, et al. (1973) found that benthic oxygen consumption under a
net-pen in Clam Bay, Puget Sound, which had previously been about 6 times ref-
erence levels, had returned to reference levels within 2 months after removal
of the pens. The pens had rested on the bottom at low tide, so dispersal of
deposits was rapid after their removal. British Columbia salmon farmers report
that deposits disappear within 4-6 months (Morrison, unpub.). Enell and Laf
(1983a) found that wastes beneath a lake net-pen were reduced from 10 to I cm
within one year after feeding rate was halved. Obviously the rates of both
accumulation and removal are going to be highly dependent upon bottom current

velocities at the culture site. Biological processes (both microbial decompo-
sition and macrofaunal activity) will also affect the rate at which the sea
floor returns to pre-culture conditions.

Sediment chemistry -

The solid wastes generated by fish or shellfish culture are enriched in
carbon, nitrogen and phosphorus relative to natural sediments (Kusuki, 1977b;
Merican and Phillips, 1985). Therefore, the accumulation of solid wastes
beneath the culture operation alters the chemistry of the bottom sediments in a
number of ways. There are changes in benthic oxygen consumption, Eh profile,

total organic carbon, total volatile solids, sulfide concentrations, and nut-
rient concentrations (nitrogenous compounds and phosphates). Changes in all
these parameters are typical of marine sediments enriched with organic matter,
regardless of whether this organic matter is derived from a mariculture opera-
tion, sewage treatment plant, pulp mill, food processing plant, or any other
comparable source.
Benthic oxygen consumption - The most extensive survey of oxygen uptake of
bottom sediments in Puget Sound was by Pamatmat, et al. (1973). Typical oxygen
consumption values throughout the Sound ranged from 4-56 ml 02 m -2hr 1. The
respiration of benthic organisms (bacteria, meio- and macrobenthos) generally
accounted for 10-50% of the total with chemical oxidation accounting for the
remainder. Sediments under salmon net-pens in Clam Bay had an oxygen uptake
-2 -1
rate of about 125 ml 02.m m2hr . Nearly two-thirds of this total was attrib-
utable to respiration. The organic enrichment was limited to the immediate
vicinity of the net-pens. Oxygen consumption rates 15 m from the pens (31-42
-2 -1
ml 0 2m .hr ) were only slightly above reference conditions and no effect on
benthic respiration was evident 30 m away. Rates of benthic oxygen consumption
2-3 times reference values have also been reported beneath net-pens or mussel
longlines in Henderson Inlet, Puget Sound (Pease, 1977), New Zealand (Kaspar,
et al., 1985) and Swedish lakes (Enell and LSf, 1983b).
Eh profile - The redox potential (Eh) is a measure of the oxygen content
of sediments. In oxygenated (aerobic) environments the Eh is greater than
zero. In anaerobic, reducing environments Eh is negative. The interface
between the aerobic and anaerobic zones of the sediment (known as the redox-
potential discontinuity) is a zone in which the oxygen demands of decomposing
organic matter are in balance with the supply of oxygen from the overlying
water. These chemical changes are also manifested by color changes of the sed-
iment with oxygenated sediments appearing brown or yellow and anaerobic sedi-
ments appearing black. The depth of this interface can be used as a measure of
the rate of organic material input (Fenchel and Riedl, 1970; Pearson and
Stanley, 1979).
An example of the effect of floating mariculture on sediment Eh is shown
in Figure 3 based on measurements made beneath mussel longlines in Sweden
(Dahlback and Gunnarson, 1981). Eh values become more negative with depth in
the sediment column in all marine sediments due to the limitations on diffusion
of oxygen from the overlying water. However, there is generally a layer of

Eh (mV)
-200 -100 to +100

- '
a8-

10-

12 -

Figure 3 - Redox potential (Eh) profiles in the sediment
column under a mussel culture (squares), and
from a reference area (circles). From
DahlbAck and Gunnarrsson, 1981.

oxygenated sediments which overlies the anaerobic sediments. Such a pattern is
seen at the reference site in Figure 3. Beneath the mussel culture the high
input of organic wastes from the mussels prevents the establishment of a super-
ficial aerobic sediment layer and negative Eh values are observed at the sedi-
ment surface. The fact that the reference sediment has a lower Eh than the
sediment beneath the mussels at depths greater than 3 cm is not a typically
observed condition and was not found in later measurements at the same site.
Total organic carbon (TOC) and total volatile solids (TVS) - TOC and TVS
are both measures of sediment organic enrichment. These two measures are high-
ly correlated and both have been used in documenting organic enrichment which
has occurred beneath mariculture facilities. Concentrations of both TOC
(Dahlback and Gunnarson, 1981; Kaspar, et al., 1985) and TVS (Kaspar, et al.,
1985; Mattson and Linden, 1983; Pease and Goodwin, unpub.) have been found to
be elevated under mussel culture facilities. Pease (1977) found concentrations
of total carbon (TOC unmeasured) to be approximately twice as great under
salmon pens in Henderson Inlet in comparison to reference areas, but no enrich-
ment of total carbon was found beneath the Sea Farm of Norway pens at Port
Angeles (Milner-Rensel Assoc., 1986).
Sulfide concentrations - In the absence of oxygen, the microbial degrada-
tion of organic matter is accompanied by the reduction of the sulfate anion in
seawater to hydrogen sulfide. Figure 4 illustrates the sulfate and sulfide
concentrations in sediment pore waters beneath a mussel longline (Dahlback and
Gunnarson, 1981). Due to the high input rate of organic material and the con-
sequent development of anaerobic conditions, sediment pore waters were found to
have low concentrations of sulfate, a high rate of reduction of sulfate to
sulfide, and high concentrations of sulfides. Ito and Imai (1955) also report-
ed high sulfide concentrations beneath oyster racks. Furthermore, they related
an increasing sulfide concentrations with duration of culture at the site.
Sediments beneath 2-year old rafts had sulfide concentrations approximately
twice those in reference areas, whereas 6-year old rafts had sulfide concentra-
tions nearly 10 times greater than reference areas. Lieffrig (1985) has shown
hydrogen sulfide production beneath salmon net-pens to be 10 times greater than
reference levels.
The generation of hydrogen sulfide beneath culture operations has been im-
plicated as a cause of reduced productivity in shellfish culture (Arakawa, et
al, 1971; Ito and Imai, 1955) or mortality in salmon net-pen culture (Braaten,

sulfate

20- i :::

15- ::. ::::::

E
10 -::::::_

5 -

sulfide
0.12 - - 12

o0.09- -9

0.06- . ....:::-:::-
El

0.06 - . -6 3

E '

o 20.0- : -

E 10.0- n S

E - E

reference mussel
sediment sediment

Figure 4 - Average pore water concentrations of sulfate
and sulfide and sulfate reduction rate in
sediments from beneath a mussel culture and
from a reference area. From Dahlback and
Gunnarsson, 1981.

et al., 1983; Pedersen, 1982). A relationship has also been found between con-
centration of sulfides in the sediments and incidence of disease in cultured
yellowtail (Arizono, 1979). Periodic relocation of culture operations or rota-
tion among several sites has been generally recommended as a means to prevent

loss of productivity or mortality.
Nutrient concentrations - Feed, feces and pseudofeces all contain several
times more nitrogen and phosphorus than natural sediments (Arakawa, et al.,
1971; Kusuki, 1977b; Merican and Phillips, 1985) thus it is expected that there
would be an enrichment of these elements in the solid wastes generated by mari-
culture facilities. This expectation is generally supported by the available
data from both freshwater and marine culture:

1) Henderson Inlet, Puget Sound (Pease, 1977) - Total nitrogen content of sed-
iment beneath salmon net-pens was approximately twice that of the reference
area.

2) Port Angeles, Strait of Juan de Fuca (Milner-Rensel Assoc., 1986) - Ammonia
nitrogen in interstitial and near-bottom water was elevated beneath salmon
net-pens. No effect of pens on total nitrogen was observed.

3) Swedish lakes (Enell and L5f, 1983b) - Interstitial water concentrations of
ammonia and molybdate-reactive phosphorus were approximately ten times greater
beneath rainbow trout net-pens than in the reference area.

4) Lake Letowo, Poland (Trojanowski, et al., 1982) - Phosphorus and several
nitrogenous compounds were enriched in sediments beneath rainbow trout

net-pens.

5) Manchester, Puget Sound (Anonymous 1983a; D. Damkaer, NMFS, unpub. data) -
No elevation of ammonia, total organic nitrogen, total inorganic nitrogen, or
total Kjeldahl nitrogen attributable to salmon net-pens was observed. Concen-
trations of several of these parameters were elevated beneath a dock adjacent
to the culture operation, and it was suggested that fouling organisms on the

dock may be a source of some portion of the nitrogenous material.

Nitrogen and phosphorus are not retained equally in the solid wastes gen-
erated by net-pen culture. Most phosphorus consumed by a salmonid is lost as
solid waste (66-84%). In comparison only 11-22% of the consumed nitrogen is
lost in the solid phase, with the vast majority excreted in a dissolved form as
ammonia or urea (Ackefors and Sbdergren, 1985; Enell and Lbf, 1983b; Gowen, et
al., 1985; Merican and Phillips, 1985; Phillips, et al., 1985). Over time the
nutrients in the enriched sediments will be released to the water column, with
the release of nutrients accelerated under anaerobic conditions (Enell and Lbf,
1983b). In a freshwater environment where phosphorus concentration is typical-
ly the limiting nutrient controlling phytoplankton growth, the release of phos-
phorus from the sediments can contribute to eutrophication. However, in marine
environments where nitrogen is typically the limiting nutrient, the principal
nutrient effect of mariculture will be felt immediately with the excretion of
dissolved metabolites.

Conclusions

Floating mariculture generates large amounts of solid wastes in the form
of unutilized feed and feces (salmon culture) or feces, pseudofeces and shell
debris (mussel and oyster culture). These wastes may accumulate on the bottom
beneath the culture facility resulting in organic enrichment and associated
dramatic changes in sediment chemistry. The changes which have been observed
include decreased redox potential, increased sediment oxygen consumption and
increased concentrations of TVS, TOC, sulfides, nitrogenous compounds and phos-
phates.
The areal extent of effect on bottom sediments is dependent upon the sink-
ing speed of the waste, the current velocity and the water depth. Ideally, a
culture operation should be sited so that currents would disperse the wastes
over such a broad area that their deposition would not exceed the assimilative
capacity of the marine environment and there would be no measurable effect on
sediment quality. In practice, however, most mariculture facilites have been
located in relatively shallow water (< 20 m), and there has been accumulation
of solid wastes beneath the facility. From the literature reviewed, it appears
that if less than 15 m of water is maintained under the lowest point of the
culture structure there is a high probability that there will be visible accum-
ulation of wastes on the bottom. These observations have implications in site

selection and evaluation. For example, a culture operation with less than 15 m
of water beneath it should not be located over a habitat of special signifi-
cance or a habitat the loss of which would be considered unacceptable. Insuf-
ficient data were available for evaluation of sites in deeper water. Data on
current velocity and its relationship to solids accumulation are also extremely
limited. While most investigators have failed to provide quantitative data on
current velocity, this is obviously a critical parameter if we are ever to
obtain some predictive capabilities of the magnitude of effects.
However, even at sites in very shallow water, the areal extent of accumu-
lation of solid wastes appears to be very limited. Visible accumulation of
wastes typically are present only within about 30 m of the facility. This
areal extent of effect is inferred from visible accumulation of wastes. Less
information is available on the areal extent of alteration of sediment chemis-
try since most investigators on this topic established only distant reference
stations rather than sampling along a transect at frequent intervals. However,
it appears that the areal extent of effects on sediment chemistry is about the
same as the extent of visible accumulation. As discussed earlier, Pamatmat, et
al. (1973) found only a slight increase in benthic oxygen consumption 15 m from
net-pens. Mattson and Linden (1983) found elevation of TVS under mussel long-
lines was limited to within 15 m of the culture site. In addition, most inves-
tigators (Dahlbdck and Gunnarson, 1981; Ito and Imai, 1955; Kadowaki, et al,

1980; Merican and Phillips, 1985; Pease, 1977; Pease and Goodwin, unpub.) have
established reference stations within 100 m of the culture operation and freq-
uently within 35 m, thereby implying that the expected areal extent of effect
is very small.
There are several steps operators might take to minimize the accumulation
of solid wastes beneath facilities. Not only may these efforts reduce the
effect of the culture activities on the environment, but they may minimize
mortalities of the cultured animals or decreases in productivity due to the
generation of hydrogen sulfide by the waste deposits. Mitigative measures an
operator might take include:

1) Siting in areas with the greatest current velocity and water depth permitted
by anchoring and structural considerations and operational constraints.

2) Avoid siting the facility in areas where the bathymetry would promote the
accumulation of wastes (e.g. within silled embayments).

3) Arrangement of culture units so as to disperse wastes over as broad an area

as possible.

In other countries (e.g., Norway) operators have found a need to periodic-
ally rotate culture sites if the culture facilities are located in areas where
currents do not provide sufficient waste dispersal. Some operators have em-
ployed submersible mixers to disperse wastes over a wider area. Other tech-
niques to minimize solids accumulation which have been either proposed or
tested include the use of waste catchment devices beneath net-pens, dredging or
trawling of the sea bottom and the culture of shellfish around a net-pen in
order to use the solid wastes to produce a harvestable product (Braaten, et
al., 1983; Enell, et al., 1984; Pedersen, 1982; Rosenthal, 1985). None of
these techniques have been tested in Puget Sound.

3.5 BENTHIC MACROINVERTEBRATES

Concern

Given that wastes which settle to the bottom beneath floating mariculture
facilities can alter the physical and chemical properties of the substratum,
how do these changes affect the benthic invertebrate community? Are there
changes in community composition or abundance? Can there be complete mor-
tality?

Discussion

Benthos live in or on the sediments. Macrobenthos are operationally de-
fined as those invertebrates which pass through a 4 mm screen sieve but which
are retained on a 0.5 mm mesh sieve (some investigators have used a 1.0 mm as
the lower limit). The terms macrofauna, macroinvertebrates and macrobenthos
are used interchangeably. Macrobenthic communities are often numerically dom-
inated by polychaete worms, molluscs, and crustaceans. These groups serve as
prey organisms for many species of bottom-feeding fish. With the exception of
some bacteria (Section 3.8), invertebrates smaller than 0.5 mm are not discus-
sed in this report because of the paucity of data; they are rarely sampled
quantitatively. Invertebrates larger than 4 mm are considered as megafauna and
are addressed in Section 3.6.
The effects of mariculture facilities on macrobenthic communities are
largely attributable to organic enrichment of the substratum. With the excep-
tion of occasional use of antibiotics (Section 3.10), toxicants such as heavy
metals or synthetic organics, which are present in many other discharges to the
marine environment, are lacking in discharges from mariculture facilities.
Effects of organic enrichment in the marine environment are better documented
than effects of many other disturbances. An excellent review of the effects of
organic enrichment on the macrobenthic invertebrate community can be found in
Pearson and Rosenberg (1978), and much of the discussion in this section is
based on their work.
Figure 5 illustrates typical qualitative changes in species number, bio-
mass and species abundances along a gradient of organic enrichment. The abcis-
sa is shown as an axis of decreasing organic input. However, the axis could

4D
o o cO
oC~~~~~~~~~~~~~~ 00
-0A opportunistic 0
o species transition o
dominate zone I

B
A

increasing organic input

Figure 5 - Generalized trends in number of species (S),
biomass (B) and total macrofaunal abundance
(A) along a gradient of organic enrichment.
From Pearson and Rosenberg, 1978.

also represent the changes which occur with, increasing distance from the source
of organic input, or temporal changes that occur at a given point following
pollution abatement. The faunal zones shown an Figure 5 are not necessarily
observed in all instances of organic enrichment. Depending on local environ-
mental conditions certain zones may be lacking or others added. The presence
of discrete zones is an oversimplification for illustrative purposes. There is
actually a continuum of change, and boundaries between communities are gener-
ally indistinct. However, this model provides a useful summary of benthic
community response to organic enrichment.
Regardless of the nature of the source (e.g., mariculture, sewage treat-
ment, pulp mill, log handling, food processing, etc.), the response of the
benthic community to organic enrichment is generally predictable. At low
levels of organic input, a transition zone develops in which abundance, biomass
and species richness gradually decrease from levels typical of the unpolluted
environment. It may be noted on Figure 5 that an area within the transition
zone exists in which species richness and biomass reach a value somewhat great-

er than in undisturbed conditions. This phenomenon has been termed "biostimu-
lation" (Chen and Orlob, 1972). In this area the organic input provides a rich
food source for both deposit and suspension-feeding organisms, yet the rate of
input is not so great that it interferes with the mechanics of suspension feed-

ing (e.g., clogging of filtering structures) nor causes serious oxygen deple-
tion. The presence of a biostimulated zone is not universal, but the results
of several studies have suggested its presence (Chen and Orlob, 1972; Christie
and Moldan, 1977; Mackay, et al., 1972; Soule and Oguri, 1976; 1979).
At a somewhat higher rate of organic input, total macrofaunal abundance
attains a maximum value. Biomass may also be slightly elevated, but the number
of species is very low. The increased abundance and biomass results from the
proliferation of only a few species. Pearson and Rosenberg designated this
point as the "peak of opportunists". The polychaete Capitella capitata is the
epitome of an opportunistic species. It is found throughout the world in or-
ganically enriched environments where few other species are able to survive.

With still higher rates of organic input, there is a complete absence of
benthic macrofauna. The rate of organic input is so great and the rate of
water exchange so low that oxygen levels in bottom waters and sediments de-

crease (or sulfide levels increase) to such an extent that aerobic organisms
can not survive.

The extent of organic enrichment also affects the depth distribution of
benthic invertebrates in the sediment column (Figure 6). The redox potential
discontinuity layer (RPD) is the interface between aerobic and anaerobic zones
within the sediments. The fact that the RPD is generally deeper in the sedi-
ment at lower rates of organic input is attributable to both physical-chemical
and biological processes. The reduction in organic input reduces oxygen de-
mand, thereby allowing oxygen from the overlying water to passively diffuse to
greater depths in the sediment column. Animals accelerate this process. Bur-

rowing, burrow ventilation, tube-construction and feeding activities promote
movement of oxygen into the sediment and the transport of deeply buried anaer-
obic sediments to the sediment surface. At high rates of organic material
input, the RPD is close to the sediment-water interface. The aerobic zone is
limited to the top few millimeters or the entire sediment column is anaerobic.
The macrofauna is characterized by species which live and feed very near the
sediment-water interface. Deep-burrowing species are generally excluded.
Mattsson and Linden (1983) observed the progression of community changes
described above during a three year monitoring program beneath mussel longlines
in Sweden. Three months after the culture was initiated communities beneath

the longlines were similar in faunal composition to those at the reference
station. Within six months after the start of culture, brittle stars had dis-
appeared at the culture site. Species originally dominant in the community
decreased in number and finally disappeared after 15 months. Opportunistic
species became established in the culture area concurrently with the decline of
the original fauna. Within six months, large populations of Capitella
capitata were established, and the species later reached densities as high as
20,000 individuals m . Other opportunistic polychaetes (Scolelepis fulginosa
and Microphthalmus sczelkowii) appeared after about a year of culture. Total
abundance and biomass decreased initially but then fluctuated widely depending
on densities of the opportunistic polychaetes.
Mattsson and Linden (1983) also monitored the recovery of the benthos
after removal of a mussel culture facility that had been in production for
three years. Six months after removal the bottom was still covered by 20-40 cm
of mussel shells and sulfide-rich sediments. The benthos was numerically dom-
inated by C. capitata, S. fulginosa and the amphipod Corophium insidiosum.
Monitoring continued for a year and a half after mussel removal, during which
only very limited macrobenthic recovery was observed.

NORMAL TRANSITORY POLLUTED GROSSLY POLLUTED

Figure 6 - Diagrammatic representation of faunal and sedimentary changes
under increasing organic loading. From Pearson, 1976.

It is often difficult to distinguish the effects of mariculture related
organic enrichment from other sources of organic material. The observations of
Mattsson and Lind6n (1983) highlight this difficulty. The initial reference
station they established was 25 m from the culture site. When they found evi-
dence of organic enrichment at the reference station, they assumed it was too
close to the longlines, and established a second reference site 250 m away.
When this site also showed evidence of community changes, a third reference

site was established 1500 m away from the culture site. Mattsson and Linden

later determined that the effects on the benthic community attributable to the
culture operation were limited to within 20 m of the culture site. The changes
in the communities at the reference stations resulted from organic input relat-
ed to the decomposition of a regional phytoplankton bloom.
Table 9 presents the results of several studies which examined the effect
of floating mariculture on marine benthic macrofauna. With one exception (a
farm examined by Ervik, et al. in depths of 25-61 m) all studies demonstrated
some effect on the benthos. The effects noted were very similar to those sug-
gested by the Pearson and Rosenberg (1978) model. Directly beneath the culture
operation there was, at some sites, a complete absence of macrofauna. As the
rate of organic input decreased (or as one samples farther from the culture
site), a community dominated by opportunistic species was observed. This com-
munity was generally characterized by high total abundance and low numbers of
species. Capitella capitata was frequently a dominant community member.
With further reductions in the rate of organic input (or increasing dis-
tance from culture site), macrofaunal community composition gradually ap-
proached that characteristic of the reference sites. The area of effect of the

culture operation on the macrobenthos generally corresponded to the area of
altered sediment quality (Section 3.4), and typically was restricted to an area
within a 30 m radius of the culture facility.
Only two investigators (Ervik, et al., 1985; Mahnken, et al., unpub.) have
observed the presence of a biostimulated zone around mariculture facilities,
and in both cases data to support the claim were not presented. The presence
of a zone of biostimulation at some distance from a mariculture site does seem
plausible, but this potentially "beneficial" effect should not be construed to
mean that there is no net adverse effect of organic enrichment on the benthos
(i.e., that the zone of biostimulation compensates for zones of mortality or
zones of community degradation). As organic input increases, the extent of the

Table 9
Effects of floating mariculture on benthic macrofauna

Location (Reference) Animal cultured Observations

Weyerhaueser Co. Coho salmon High macrofaunal abundance and low number
Henderson Inlet, of species beneath net-pens. Fauna
Puget Sound (Pease, 1977) numerically dominated by Capitella capitata

Sweden Blue mussels Area under raft dominated by opportunistic
(Mattsson and Linden, 1983) species. Diversity decreased, biomass and
abundance fluctuated widely. Area of
effects extended 20 m from culture.

Ria de Arosa, Spain Blue mussels Macrofauna under raft dominated by
(Tenore, et al., 1982) opportunistic polychaetes. Biomass,
(L6pez-Jamar, 1985) abundance and diversity decreased in raft
area. Tube-building species predominated.

Kenepuru Sound, Green-lipped Biomass unaffected by culture although
New Zealand mussel diversity decreased. Only polychaetes
(Kaspar, et al., 1985) found beneath culture. Infauna from
reference area also included brittle stars,
molluscs and crustaceans.

Ireland Salmon Macrofauna absent beneath net-pens.
(Stewart, 1984) Zone around perimeter dominated
by Capitella capitata.

Norway Atlantic salmon Three farms were examined. Under one the
(Ervik, et al., 1985) number of species was low and the community
was dominated by opportunistic species. Under
the second the community was biostimulated.
Under the third there was only minimal effect.

Bellacragher Bay, Atlantic salmon Highly localized zone under net-pens with
Ireland no macrofauna. Unaffected community
Doyle, et al. (1984) established within 25-45 m.

Uchiura Bay, Yellowtail As organic material accumulated there was
Japan an increase in abundance of opportunistic
(Kitamori, 1977) polychaetes and a decrease in the relative
proportion of molluscs and crustaceans.

biostimulated zone increases as a function of the perimeter of the affected
area (i.e., scales as a function of length), but the extent of adverse effects
increases as a function of the entire affected area (i.e., scales as a function
of length squared). Therefore, as the level of organic enrichment increases,
the extent of potentially adverse effects increases much more rapidly than the

extent of potential biostimulation.
Following cessation of organic input, such as after harvest or removal of
the culture structure, a certain period of time will be required for the ben-
thic population to recover from the disturbance to a point at which the affect-

ed community becomes similar in species composition and abundance to reference
areas never affected by the organic enrichment. Although the Pearson and
Rosenberg model suggests the sequence of benthic community recovery following

the cessation of organic input, it does not predict the rate. As noted earlier
Mattsson and Linde'n (1983) found only slight community recovery 18 months after
removal of mussel longlines. Recovery of the benthos following closure of a
pulp mill has been found to require 3-8 years (Rosenberg, 1976). Vesco and

Gillard (1980) reviewed the results of many investigations along the Pacific
coast and found that recovery following disturbance typically requires 2 years
and often 10 years or more. The time required for recovery will depend on many
factors including the severity of the disturbance, the geographic location, the
amount of disturbance to which a community is normally exposed and the life
history of the species in the community (Vesco and Gillard, 1980).

Conclusions

The responses of benthic communities to organic enrichment fall along a
continuum. As the level of enrichment begins to increase, the macrobenthic
community begins a transition phase in which the background community is grad-
ually altered. There may be biostimulation as the community undergoes this
transition: a low rate of organic input may provide an enriched food source
for the benthos without resulting in environmental degradation. As enrichment
increases still further, sensitive species of the background community are
replaced by a few opportunistic species tolerant of low dissolved oxygen, high
sulfide concentrations and the other physical and chemical changes accompanying
organic enrichment. At high rates of organic input, complete community mortal-
ity occurs. This continuum of change from background community to total com-
munity mortality may exist spatially as a function of distance from a culture

site and temporally as a function of changes in organic input rates over time.
The extent of effects of floating mariculture on benthic communities is
dependent on the magnitude of changes in sediment quality, which is in turn
dependent upon the degree to which wastes are dispersed by water movements.
The relationships between these variables is shown in Figure 7. As the poten-
tial for waste dispersal (e.g., high current velocities and deep water) in-
creases, the likelihood that there will be significant effects on the benthos
decreases. Most floating mariculture facilities established to date have been
sited in relatively shallow water (< 20 m) and in areas of weak currents. Con-
sequently major alterations of benthic communities immediately beneath culture
facilities and for short (< 30 m) distances surrounding the sites have typic-
ally been observed. The effects of the mariculture operations on the benthos
can be expected to persist for the duration of culture activities and for at
least several years after their cessation.

Input

ORGANIC MATTER

Sedimentation

HIGH ORGANIC
CONTENT IN SEDIMENT
BACTERIAL
DECOMPOSITION

Well flushed moerate water Stagnation
re - newel
OXIDIZED SEDIMENT ANOXIC SEDIMENT
WITH PLENTIFUL REDUCED 02 EMERGING Eh AND OVERLYING
FOOD SUPPLY IN WATER IN SEDIMENT WATER

High Bioturbation Few niches

HIGH BIOMASS POOR MACROFAUNA NO MACROFAUNA

Figure 7 - Simplified diagram showing some pathways
of organic input to the marine environment
in relation to amount of water exchange.
From Pearson and Rosenberg, 1978.

3.6 FISH AND MEGAFAUNA

Question

How do changes in water quality or the accumulation of organic material on
the bottom affect natural fish and megafaunal populations in the vicinity of a
mariculture operation? Of particular concern are fish and megafauna of commer-
cial importance.

Discussion

Mariculture could potentially have adverse effects on nonmotile megafauna
living in the sediment (e.g., geoducks and other large bivalves). Low dis-
solved oxygen concentrations and high sulfide concentrations shown to be detri-
mental to infaunal macrobenthos (Section 3.5), are also likely to prove harmful
to megafaunal organisms which live in close contact with the enriched sedi-
ments. The severity of the effect will be somewhat reduced by the fact that
many of these megafauna maintain contact with the overlying water through
siphons (bivalves) or irrigated burrows (mud shrimp), but under conditions of
high organic enrichment some mortalities are probable. The areal extent of
impact would be expected to be limited to the immediate vicinity of the culture
as has previously been shown for effects on sediment chemistry (Section 3.4)
and macrofauna (Section 3.5).
Mariculture has been generally found to result in increased densities and
diversity of motile megafauna (e.g., crabs, starfish) and fish in the vicinity
of the culture operations. These organisms are able to exploit the enhanced
food resources of the culture environment, but are not so intimately associated
with the enriched sediments that toxic effects develop. Table 10 presents a
summary of observed effects of mussel and net-pen culture. With very few ex-
ceptions, culturing activities resulted in increased abundance, species rich-
ness and/or biomass in the immediate vicinity of the culture facilities.
Extreme organic enrichment of bottom sediments may reduce rather than enhance
numbers of motile megafauna (e.g., the loss of starfish from the Japanese
oyster grounds studied by Ito and Imai, 1955), but in the vast majority of
cases culture activities result in enhancement of fish and megafauna popula-
tions, including enhancement of some species of commercial importance.

Table 10
Observations of fish and megafauna in the vicinity of mussel and salmon culture facilities

Location Observations Reference

Penn Cove Mussels Sea cucumbers and hydroids which had fallen from the rafts were found Pease and Goodwin, unpub.
Penn Cove, on the bottom. The crab Cancer gracilis was found in the area of the
Puget Sound culture, presumably attracted by food in the form of mussels which
had fallen from the strings. Mud shrimp were reduced in density under
the culture, presumably because the shell debris inhibited burrowing.
Octopus were attracted to the area.

Weyerhaueser Co. The crab Cancer gracilis and three species of fish (rock sole, great Pease, 1977
Henderson Inlet, sculpin and shiner perch) were more abundant at sampling sites under
Puget Sound net-pens than in reference areas.

Scan Am Farms Starry flounder, tubesnouts, buffalo sculpin and sailfin sculpin foraged Kyte, unpub.
Hadlock, around net-pens. Baitfish (probably Pacific herring) congegrated around
Puget Sound pens. Fishermen have reported excellent Dungeness crab fishing in area.

Domsea Farms Surf perch, English sole and numerous sea anemones were present under Pease, unpub.
Clam Bay, net-pens. Geoducks were absent from area beneath pen, although they
Puget Sound would have been expected based on sediment type.

Domsea Farms Surf perch, flatfish and numerous sea anemones were present under net- Pease, unpub.
Fort Ward, pens. Kelp was absent from area under net-pens.
Puget Sound

Kenepuru Sound, Mussels which had fallen to the bottom provided substrates for the growth Kaspar, et al., 1985
New Zealand of epifaunal organisms and attracted predators such as fish and starfish.
In comparison, the reference area was largely devoid of megafauna.

Sweden The fish food lost in the form of infaunal mortalities was compensated Mattsson and Linden, 1983
compensated for by mussels and associated fouling organisms which fell
to the bottom.

Ria de Arosa, Biomass of crabs and starfish was greatly enhanced in raft areas because Tenore, et al., 1982
Spain of the rain of mussels and epifauna from rafts. Fish biomass also Romero, et al., 1982
was increased in raft areas but not to the same extent as the crabs Lopez-Jamar, et al., 1984
and starfish. The primary fish prey organisms outside the raft area
were infaunal invertebrates, whereas around the rafts fish fed primarily
on epifaunal species.

Maine Fish and shellfish densities were increased in the vicinity of net-pens K. Honey, Maine Dept.
and mussel rafts. Fishermen have reported that the sites are excellent of Marine Resources,
for lobster fishing. pers. comm.

Similar observations have generally been reported in freshwater culture.
Wild fish are frequently observed in high densities around cages containing

cultured fish (Beveridge, 1984; Eley, et al., 1972; Loyacano and Smith, 1976;
Hays, 1980). Culture activities have also been shown to increase the number
and biomass of native fish in lakes and reservoirs (Kilambi, et al., 1976;
Loyacano and Smith, 1976). This effect is due to both ingestion of excess

commercial feed by the wild fish and increased nutrient levels in the lakes
(Beveridge, 1974). However, adverse effects are possible in cases of intensive
aquaculture development in freshwater systems. Penczak, et al. (1982) reported
that net-pen culure of rainbow trout is resulting in disappearance of the
native Coregonidae in Lake Glebokie, Poland, presumably because of the in-
creased nutrient input. In Laguna de Bay, Phillipines the pen culture of
milkfish and Tilapia has altered the density and species composition of the
wild fish community (Ackefors, et al., 1984). That such an effect was observed
is not at all suprising since pens occupy about one-third of the surface area
of the lake.

Conclusions

Accumulation of organic material may result in the mortality or exclusion
of nonmotile megafauna which live in burrows or are otherwise in intimate con-
tact with the sediment. This effect should be limited to the zone of organic
enrichment in the immediate vicinity of the culture site. For megafauna living
on or above the sediment surface and for fish in the surrounding waters, the
effect of mariculture is more often positive than negative. Only in cases
where very limited flushing permits development of near-bottom anoxia or other
pronounced deterioration of water quality, would an adverse effect on fish or
epibenthic megafauna be expected. This deterioration has rarely been observed.
More often there is an increase in abundance, species number, diversity, and/or

biomass of fish and megafauna in the vicinity of culture operations. Commer-
cial and recreational fishing is typically enhanced in the vicinity of the
culture (e.g., Dungeness crabs - Kyte, unpub.; lobsters - K. Honey, Maine Dept.
of Marine Resources, pers. comm.; bluegills - Loyacano and Smith, 1976).
Wild fish or megafauna are attracted to the culture operations for several
reasons. In part, there is a behavioral tendency for fish to congregate around
floating objects (Gooding and Magnuson, 1967). A floating mariculture facility

also increases the availablity of food in the area. This food may take several
forms including excess feed unutilized by the cultured fish, epifaunal organ-
isms growing on the culture structure or which fall to the bottom, or the high
abundances of opportunistic macrofaunal species in the adjacent bottom sedi-
ments.

3.7 INTRODUCTION OF EXOTIC SPECIES

Questions

Could a species imported to Puget Sound for culture escape and establish a
reproducing population in the wild? Would this pose a threat to native species
or in any other way result in irreparable ecological harm? What are the dan-
gers of other organisms and diseases being imported with the cultured animal?

Discussion

For many centuries man has been responsible for the accelerated spread of
species from one area of the world to another. In some cases a species has

been intentionally introduced because it possessed certain characteristics
which were advantageous to man and which were not exhibited by indigenous
species. It is probable that an even greater number of introductions have been
unintentional. The transport of species on ship hulls and in ballast tanks
provides the best example of unintentional introductions. Some introductions
can be regarded as having largely positive results, at least from a commercial
perspective. The Pacific oyster (Crassostrea gigas) and the Manila clam
(Venerupsis japonica) were both introduced to Puget Sound from Japan in the
early part of this century. The species now support the bulk of the Washington
oyster and hardshell clam industries, respectively. Some introductions have
had largely negative impacts. The Japanese oyster drill (Ocenebra japonica),
which was probably accidentally introduced along with shipments of C. gigas, is
now a major oyster predator in Puget Sound. Perhaps the greatest number of
introductions involve species which have had no effect on commercial interests
and whose ecological effects have been largely unrecognized. An evaluation of
the role of mariculture in the introduction of exotic species requires that two
separate issues be addressed: 1) the intentional importation of species for
culture; and 2) the accidental introduction of organisms associated with the
cultured species. These two issues are addressed individually below.

Intentional introduction of cultured species-

Intentional introductions are often made for purposes of culture, to
establish a new fishery or to replace a declining native fishery. Intentional

introduction of species is a common occurrence in the world today. It has been
estimated that by 1978, intentional introductions of fish and shellfish in-
volved 1,500 species worldwide (Rosenthal, 1978; Villwock, 1984). Introduc-

tions are so commnonplace that the FAQ Fish Culture Bulletin provides monthly
updates on new introductions which have taken place around the world.
It is difficult to predict how a species will respond when it is placed in
a new environment without the ecological checks and balances of its native hab-
itat. In some cases, the species will be unable to survive and reproduce in
the new environment. The species may be unable to establish a self-sustaining
population, either dying if forced to live outside of the confines of the
culture environment, or requiring a continuous flux of imported juveniles to
maintain the population. For example, of 27 introductions of fish to the
Netherlands, most of the species, including many salmonids, have failed to
establish sustaining populations (DeGroot, 1985). At the other extreme, the
species may thrive in the new environment, displacing native species through

habitat alteration or competition for resources. Examples would include the
introduction of Tilapia spp. throughout Southeast Asia and rainbow trout (Salmo
gairdneri) into South Africa and South America (Rosenthal, 1980). In these

cases, the introductions have led to the establishment of important fisheries
or culture industries, but they have also led to the disappearance of native

fish species.
Of those species which are currently raised in floating mariculture in
Puget Sound, only the Pacific oyster (C. jgigs), red alga ("nori" - Porphyra
sp.), and the Atlantic salmon (Salmo salar) are not native to Puget Sound.
None of these species appears to represent a threat to indigenous Puget Sound
species. C. gigas was introduced to Puget Sound in 1902 to supplement dwin-
dling stocks of the native Olympia oyster (0. lurida). Although the species
grows well in the Sound and has established populations north to the Queen
Charlotte Islands, water temperatures are usually too low for successful
reproduction in most areas. Successful reproduction is limited to a few
isolated areas such as Quilcene Bay and Dabob Bay in Hood Canal and Pendrell
Sound in British Columbia. Good natural sets of oyster spat are not consistent
even at these sites; in Dabob Bay commercial quantities of seed are produced
6-7 years out of 10 (Chew, 1984).
Noni culture in Puget Sound requires the importation of pathogen and
parasite-free, unialgal cultures from Japan. Environmental conditions in Puget

Sound are not suitable to either the reproduction or dispersal of the species
(Hurlburt, 1984). The reproductive cycle of noni includes a stage which bores
into oyster shell. Given the long history of oyster importation to Puget Sound
from Japan, it is probable that there has been extensive and repeated inadver-
tent introduction of noni. As no wild growth of noni has occurred in Puget
Sound to date, it is unlikely that culture of the species would result in
future establishment of the species.
The potential for S. salar to establish a self-sustaining population in
Puget Sound has been thoroughly examined by Lindergh (unpub.). In Washington
State there have been deliberate attempts to stock the species in Chambers
Creek, Minter Creek and numerous eastern Washington lakes, 'some of which have
access to the Pacific Ocean. There have also been several escapes of the
species from the National Marine Fisheries Service facilities in Clam Bay,

Puget Sound. Despite these numerous releases, no returning adults have been
reported. Introductions have also been attempted in Oregon, California, Brit-
ish Columbia, New Zealand and Chile. In a few cases, self-sustaining, land-
locked populations have been established, but most require continued hatchery
support. No self-sustaining anadromous runs have been established anywhere in
the world outside of the species' native range. Lindbergh (unpub.) estimated
that about 5 million individuals of S. salar have been released in unsuccessful
stocking attempts in the Pacific Northwest. Given the failure of these ef-
forts , it would appear unlikely that future escapes from culture facilities in
Puget Sound would result in the establishment of a wild population.
While those species currently employed in floating mariculture in Puget

Sound seem unable to establish self-sustaining populations, future introduc-
tions of other species should be carefully evaluated on a case-by-case basis as
they are proposed. It is never possible to rule out adverse effects until the
introduction is an accomplished fact, and then, if successful, the introduction
may be irreversible. Thus, any introduction should be carefully planned and
thoroughly reviewed by the scientific community and appropriate government
agencies (see Courtenay and Robins, 1973 for a suggested review protocol). It
should be clearly demonstrated that the benefits to be gained by the introduc-
tion outweigh the potential ecological consequences.

Accidental introduction of associated organisms -

The transport of oysters throughout the world has been credited with con-
tributing to the spread of more species throughout the world than any other
human activity (Cronin, 1967). Oyster shells provide habitats for algae,
hydroids, bryozoans, amphipods and a great variety of other marine life.
Oysters may also serve as reservoirs for finfish pathogens (Meyers, 1984). The
accidental introduction of the oyster drill to Puget Sound along with imported
oysters from Japan was noted earlier. Other examples are provided by Quayle
(1964), Rosenthal (1980) and Rosenfield and Kern (1979). Many introduced
species currently in Puget Sound have come in with shipments of Crassostrea
gigas. However, the danger of continued new introductions is not as great as
it has been in the past. The escalating cost of obtaining oyster seed from
Japan has stimulated the development of oyster hatcheries along the west coast
of the United States. Since 1977, the importation of oyster seed from Japan
has virtually ceased, and the demand is now met by hatcheries along the west
coast and the collection of local natural seed from a few isolated areas (Chew,

1984). The west coast hatchery seed may still contain associated organisms
(Carlton, 1981), but it is likely that those species which may establish
themselves in Puget Sound have already done so by natural means of dispersal.
The importation of fish for culture has also frequently resulted in the
introduction of associated organisms. Disease agents (including viruses,
bacteria, fungi, and protozoan and metazoan parasites) are frequently of
greatest concern in fish imports. Hoffman (1970) documented at least 48
species of freshwater fish parasites which have been established on other
continents because of the importation of live or frozen fish. These species
include 5 protozoa, 36 trematodes, 3 nematodes, 1 acanthocephalan and 3 cope-
pods. Other examples of parasites and diseases introduced with fish include:

1) Whirling disease - A disease in freshwater rainbow trout caused by the
protozoan Myxosoma cerebralis. This organism has been spread throughout four
continents by the transfer of live and frozen trout. However, outbreaks of the
disease have been limited to cultured fish. The parasite is rarely detected in
wild fish (Hoffman, 1970). Federal law and Washington state regulations now
prohibit the importation of trout from areas where the disease is known or is
likely to occur.

2) Infectious hematopoetic necrosis - This viral disease first appeared in
sockeye salmon on the West Coast and later infected rainbow trout. Transfer of
diseased trout and infected eggs has spread the disease throughout the north-
western and north central states (Amend, et al., 1973). The disease was intro-
duced to Japan by eggs imported from the United States (Egidius, 1984).

3) Furunculosis - This disease, which is caused by the bacterium, Aeromonas
salmonicida, may have been introduced into Europe by the importation of rainbow
trout from North America (Rosenthal, 1980).

The risk of introduction of disease and other organisms associated with
cultured animals, both fish and molluscs, depends on the source of the animals
to be cultured. There is no risk of introduction of diseases or other unwanted
organisms in Puget Sound mussel culture, since growers use locally obtained,
wild seed. The risks involved in oyster culture and net-pen culture of salmon
depend on the source of seed, eggs or fish. However, if it is necessary to
import animals into Puget Sound from outside Washington, the procedures for
minimizing unintentional introduction of other organisms should be given very

close scrutiny. An accidental introduction could have serious consequences not
only to the cultured organisms but to native species as well.

Conclusions

The introduction of species, including pathogens, probably represents the
greatest environmental threat posed by mariculture. However, the probability
of an accidental introduction is considerably lower than it has been in the
past. Many of the past introductions of unwanted species and diseases occurred
in the past when scientists and environmental managers were unaware of the
potential ecological repercussions and the various pathogens. The risks we
face today are considerably reduced, in part, because the damage has already
been done, and also because we are aware of potential problems and implement
preventative measures.
For example, when C. gigas was first introduced to Puget Sound, the
animals were collected from Japanese waters and packed directly into crates for
shipment with little or no effort to inspect for associated organisms (Quayle,
1964). Only after many years of uncontrolled importation did high-pressure

washing of the shells and careful inspection of each shipment become standard

practice.
Many federal and state laws, rules and policies curently regulate the
importation of live organisms for aquaculture or other purposes. A full review

can not be presented here, but I have presented a few of the state requirements
most pertinent to floating mariculture in Puget Sound:

1) Importation of food fish or shellfish into the state requires that the
importer obtain a permit from the Department of Fisheries (WAG 220-20-038,
220-20-039).

2) Any request for the introduction of an exotic species is subject to the re-
quirements of the State Environmental Policy Act. A report must be prepared
outlining the potential benefits of the introduction as well as the potential

environmental risks.

3) Department of Fisheries personnel or an approved pathologist must inspect,
at the place of origin, any shipment of oysters, oyster shell or seed destined
for state waters. A certificate must be obtained stating that the shipment is
free of oyster drill and other pests (WAG 220-72-082).

4) Any fish to be transferred either within or from outside the state must be
examined for disease by a state-approved inspector unless, in the case of
intrastate transfers, the disease history of the stock alleviates the need for

inspection.

5) No live salmonids or their reproductive products may be imported from Europe
except eyed Atlantic salmon eggs. For the importation of these eggs, the par-
ent stock must be certified as disease-free, a health history of the stock and
hatchery must be submitted to the state, the eggs must be disinfected, and the
eggs must be held in a quarantine facility for 90 days following swim-up.

As a result of the recent passage of the Aquaculture Disease Control law
(RCW 75.58) the requirements pertaining to importation of live animals for
culture may change in the near future. Under this law, the Washington Depart-
ment of Fisheries now has disease prevention/control responsibility for all

cultured organisms, both plant and animal. The Department of Fisheries, in
cooperation with other state and federal authorities, is currently developing a
comprehensive program of disease inspection, prevention and control specific-
ally for aquaculture operators.

3.8 DISEASES

Questions

Do the confined and crowded conditions of the culture environment increase

the probability of a disease outbreak? Can the disease be transmitted from

cultured fish to wild fish? Can mariculture promote the growth of bacteria
capable of causing disease in humans?

Discussion

Transmission of disease between cultured and wild fish -

Mariculture could potentially introduce a disease to an area where it had
not previously occurred, and potentially lead to the infection of native popu-

lations (see Section 3.7). This discussion is concerned with those disease
organisms already present, but demonstrating no clinical symptoms in the wild
fish. It has been suggested that culture conditions may provide an environment
where these disease organisms could initiate an outbreak, become more virulent
and then reinfect the native fish populations (Mills, 1982; Odum, 1974).
It is generally recognized that many diseases in fish, whether cultured or
wild, are often associated with some form of stress (Sindermann, 1984). in a

culture environment fish may be subjected to stress by overcrowding, undernour-
ishment or poor water quality. Such factors have been directly correlated to
the frequency of disease outbreaks in cultured fish (Arizono, 1979). In addi-
tion, cultured fish are also subject to physical damage (e.g., handling, abra-
sion against the net). Thus, there may be a greater frequency of surface
lesions which provide a route for infection. Cultured fish may therefore be
more susceptible to diseases than wild fish, with the degree of susceptibility
determined, in large part, by the extent to which good husbandry practices are
followed by the culture operator.
Despite the potential for disease in a culture environment, there is lit-
tle evidence to suggest that this potential represents a threat to wild fish.
In fact, there are several examples of diseases which have had more than ade-
quate opportunity to infect wild fish, but have failed to do so. For example,
viral hemorrhagic septicemia (VHS) is a disease of cultured rainbow trout in

Europe. It is believed that the VHS virus has always been present in Europe,
although the wild fish have failed to show any symptoms of the disease. Rain-
bow trout imported to Europe and held in culture are very susceptible to the
disease and many mortalities have been reported. However, the disease does not
appear in free-living rainbow trout even if they harbor the virus, and the na-
tive brown and brook trout show no signs of the disease even though they have
been infected experimentally (Egidius, 1984).
A second similar example is provided by whirling disease, a disease of
cultured freshwater rainbow trout caused by the protozoan Myxosoma cerebralis.
M. cerebralis is a parasite which causes no apparent disease in native European
trout. Rainbow trout imported from North America are very susceptible to in-
fection and disease. Despite many outbreaks of whirling disease and devasta-
ting losses of rainbow trout in culture throughout Europe, the infectious para-
site is only rarely reported in wild fish (Hoffman, 1970).
A third example, and one of direct relevance to Puget Sound, involves bac-
teria of the genus Vibrio. Vibrio spp. are natural components of the marine
microbial community throughout much of the world, and are dominant components
of the normal microbial community of salmonids (Colwell and Grimes, 1984). In
a survey of 12 fish species in Puget Sound, 88% of the gill samples, 61% of the
skin samples and 32% of the gut samples contained Vibrio spp., including a com-
mon salmonid pathogen, V. anguillarum (Baross, 1973). Most Vibrio bacteria,
including potentially pathogenic species, are not normally virulent unless the
host animal is stressed. Vibrio disease outbreaks are frequently reported in
net-pen culture in Puget Sound, particularly during the summer months, but
there is no evidence that these outbreaks have had any effect on unstressed
fish beyond the confines of the culture facilities. While Vibrio disease does
occur in wild fish, outbreaks frequently have been associated with environ-
mental degradation (e.g., oil spills, municipal sewage) or some other form of
stress (Grimes, et al., 1984; Hada and Sizemore, 1981; Larsen, et al., 1978;
Minchew and Yarbrough, 1977; Robohm, et al., 1979). There is no evidence that
mariculture has contributed to an increased incidence of vibriosis in wild
fish.

Mariculture and the proliferation of human pathogens -

It has been suggested that mariculture may potentially lead to increased
numbers of bacteria that cause disease in humans (Baross, 1973). If environ-

mental alterations associated with a mariculture operation promoted growth of
human pathogens, these pathogens might then be accumulated in filter-feeding
bivalves (e.g., mussels, clams, oysters), and infect man upon ingestion of the
contaminated bivalves. These concerns have been directed specifically towards
the bacteria genus Vibrio, since the genus is common in marine systems and in-
cludes fish, shellfish and human pathogens.
The genus Vibrio includes approximately 20 species (Baumann, et al.,

1980). Five of the species, V. cholerae, V. parahaemolyticus, V. vulnificus,
V. alginolyticus and V. mimicus, are known human pathogens, and the pathogen-
icity of two other species, V. fluvialis and V. metschnikovii, is unclear

(Spira, 1984; West and Colwell, 1984). V. anguillarum and V. ordalii infect
salmonids in Puget Sound net-pen culture. There is no evidence that these
species are human pathogens.
In humans, Vibrio infections most frequently cause gastroenteritis. The

clinical symptoms include diarrhea, abdominal cramps, nausea, and vomiting.
The disease is normally mild to moderate in severity, and symptoms typically

persist for a few days (Blake, et al., 1980). Exposure to Vibrio spp. in

seawater can also cause infection of wounds. Primary sepsis, caused by
V. vulnificus, is the most serious of the Vibrio diseases. Infection occurs

most frequently in persons with chronic liver disease (Tacket, et al., 1984).
The V. vulnificus infection causes fever, chills and nausea, and results in
death in about half the cases.
Vibrio-related illnesses are reported infrequently in the Puget Sound
area. In all the counties bordering Puget Sound there were only three reported
cases of Vibrio parahaemolyticus gastroenteritis between 1982 and 1986 (exclud-
ing 4 cases contracted out-of-state) (G. Potter, Washington Dept. of Social and
Health Services, pers. comm.). Two of the three reported cases were acquired
after consumption of raw oysters from Hood Canal. The cause of the third case
was not established. It is possible that there may be additional undiagnosed
infections, but medical authorities are confident that the disease is rare in
the Puget Sound region (C. Nolan, Seattle-King County Health Department, pers.
comm.; G. Potter, pers. comm.). No Vibrio infections other than V. parahaemo-
lyticus have been reported in Washington state (excluding those contracted
elsewhere and brought into the state by travellers).
Vibrio bacteria, including both pathogenic and nonpathogenic species, are
common members of the microbial community in estuarine environments throughout

the world. They play significant roles in nutrient recycling, and are probably
the principle bacterial group responsible for the mineralization of refractory
organic material like chitin (Baross, 1973). Surveys of Puget Sound microbial
communities have indicated that Vibrio spp. are widely distributed in the
water, sediments and biota (Baross, 1973; Colwell and Liston, 1960). Vibrio
parahaemolyticus is commonly found in Puget Sound sediments and biota, partic-
ularly during the summer (Baross and Liston, 1970). A survey of seafoods mar-
keted in the Seattle area found V. alginolyticus in half of the seafood pro-
ducts tested (Baross, 1973). Such results are not unique to Puget Sound. A
survey of market-level seafood in Louisiana indicated that 10.7% of the pro-
ducts contained V. parahaemolyticus, 7.7% contained V. vulnificus and 5.4 %
contained V. fluvialis, all of which are known or potential human pathogens
(Barbay, et al,, 1984).
The fact that potentially pathogenic Vibrio species are widespread, but
the incidence of infection is relatively low is probably attributable to three
factors. First, not all strains of a pathogenic species are virulent. Vibrio
parahaemolyticus is widespread in Puget Sound, but most of the strains are in-
capable of causing infection in humans (Baross, 1973). The factors that cause
a small proportion of the strains to become virulent are unknown. Secondly,
while Vibrio parahaemolyticus may be isolated for environmental samples at
temperatures as low as 5-100C (Ayres and Barrow, 1978), rapid growth of the
species (and many other pathogenic vibrios) does not occur until water temp-
eratures reach 15 C (Baross, 1973; Kaneko and Colwell, 1973). The cool temp-
eratures which persist throughout the year in much of Puget Sound prevent the
species form reaching the densities necessary to cause infection. Finally, the
frequency of Vibrio infections is also minimized by cooking seafood and killing
the bacteria. Vibrio infections are contracted by eating raw seafood, inade-
quately cooked seafood, or cooked seafood which is subsequently left in contact
with raw seafood.
Suggestions that mariculture could lead to increased incidence of Vibrio
infections are based on two circumstances. First, Vibrio spp. have frequently
been found in greatest abundance in areas characterized by high inputs of or-
ganic matter and/or particulate material (Baross, 1973; Colwell, et al., 1984;
Joseph, et al., 1982). Such conditions may exist in the vicinity of maricul-
ture operations. Secondly, filter-feeding molluscs concentrate bacteria
through normal feeding activities (Greenberg, et al., 1984). Thus, there is a

potential route for human infection if a mariculture operation promotes in-
creased Vibrio spp. abundances in the vicinity of harvestable shellfish.
Our knowledge of marine microbial ecology is too limited to conclusively

establish how mariculture will affect the abundances of pathogenic or nonpath-
ogenic bacteria. In freshwater systems, bacterial abundance has been shown, in
some cases, to be increased in the vicinity of aquaculture operations (Eley, et
al., 1972). The elevated rates of biological oxygen consumption observed in
sediments under net-pens in Puget Sound (Pamatamat, et al., 1973) suggest that
bacterial abundances are high in those sediments. Since Vibrio spp. are facul-
tative anaerobes and prominent components of marine microbial communities, in-
creased abundances might occur in the organically enriched sediments associated
with mariculture operations, although data are inadequate to establish this.
There have been no indications to date that mariculture contributes to the
proliferation of bacteria species and strains pathogenic to humans. In fact,
the mariculture industry is moving towards the concept of polyculture (i.e.,
culturing several species in the same area). Specifically, the feasibility of
growing mussels adjacent to salmonid net-pens is being investigated since the
mussels are able to utilize the particulate material from the net-pen as a food
source (Wallace, 1980). The Norwegians are actively experimenting with such
techniques. There are no reports indicating that mariculture has increased the
occurrence of gastroenteritis or other Vibrio related diseases anywhere in the
United States (P. Blake, Center for Disease Control, pers. comm.) or elsewhere
in the world (based on this literature search). The Washington Department of
Fisheries prohibits siting of most floating mariculture facilities over a har-
vestable shellfish bed in order to avoid the adverse effects of sedimentation
on the shellfish. Should there be any risk that mariculture would promote
transfer of pathogenic bacteria to humans via shellfish, this siting restric-
tion should minimize this risk.

Conclusions

There are many examples of wild fish transmitting disease to cultured fish
(Harrell and Scott, 1985; Jarrams, et al., 1980; Johnstone, 1984; Munro and
Waddell, 1984; Wooten, 1979). While it is more difficult to document disease
transmittal to wild fish, there are no known examples of a culture operation
providing a site for disease organisms to multiply, become more virulent and

reinfect the wild populations. It can not be conclusively demonstrated that
such a scenario can not occur, but both this literature review and another

(Odum, 1974) failed to document any instance of disease transmittal of this
nature. Prevailing scientific opinion on the subject is best stated by Egidius

(1984): "Very little is known about diseases in the wild stocks of cultured
species, but it is not likely that culture creates new infectious diseases.
However, culture accentuates infectious diseases by crowding, environmental
stress etc., although most probably the disease potential as such originates
from the wild stocks".
The potential proliferation of human pathogens because of mariculture is
also an issue where conclusive scientific evidence is lacking, but experience
to date has failed to show cause for concern. There is no evidence that mari-
culture, either in this country or elsewhere, has resulted in increased inci-
dence of human disease. It is quite likely that mariculture would increase
total bacterial abundance in the organically enriched sediments below the cul-

ture structure. However, it is not known if these bacterial species are likely
to be human pathogens, if the particular strains would be virulent, or if con-
sumption of shellfish in the vicinity would pose a health risk. The current
policy of the Washington Department of Fisheries of prohibiting floating mani-

culture development over harvestable shellfish would minimize the risk, if any,
of disease transmission.

3.9 GENETIC EFFECTS

Question

For those operations which culture species native to Puget Sound, could
interbreeding of cultured and wild animals alter the genetic fitness of the
wild population?

Discussion

it should be stated at the outset that the potential for adverse effects
by the interbreeding of cultured and wild animals is largely speculative.
'Despite decades of hatchery-rearing and release, transplantations and other
fishery enhancement programs, the genetic consequences of these actions remain
largely unknown. While there is evidence that hatchery-reared fish interbreed
with wild populations (Allendorf, et al., 1980; Campton and Johnson, 1985), it
has yet to be shown that this interbreeding has threatened the survival of any
population.
The potential for genetic alteration of wild populations by cultured pop-
ulations does not exist for all floating mariculture currently practiced in
Puget Sound. It can only be an issue if species are native to Puget Sound and
the juveniles for the culture operation are not supplied by the wild popula-
tion, or if species not native to Puget Sound can hybridize with the native
species. The Pacific oyster Crassostrea gigas and the mussel Mytilus edulis
are the only molluscs commercially grown by suspended culture in Puget Sound.
C. gigas is an introduced species to Puget Sound. The species has shown no
evidence of hybridization with native oysters, and in fact, C. gigas rarely, if
ever, hybridizes with native oyster species wherever it is been introduced
(Newkirk, 1979). M. edulis is native to the Sound, but the seed for culture is
collected from wild populations so the difference between the genetic composi-
tion of the cultured organisms and the wild populations can be no greater than
the genetic variation between the individual wild populations.
Among finfish, only the Atlantic salmon (Salmo salar) and the coho salmon
(Onchorhynchus kisutch) are currently cultured in Puget Sound net-pens (exclud-
ing experimental culture of steelhead at Squaxin Island). The history of
Atlantic salmon introductions to the west coast was discussed in Section 3.7.

Despite numerous deliberate releases to attempt to establish wild populations,
there are no reports of hybridization of the Atlantic salmon with any of the
Pacific salmon species (Lindbergh, unpub.). The fact that the Atlantic and
Pacific salmon belong to different genera (i.e., Salmo and Onchorhynchus) would
suggest a low probability of successful hybridization. Three attempts have
been made to produce hybrids between Atlantic salmon and the rainbow trout or
steelhead (Salmo gairdneri), but all have proven unsuccessful (Lindbergh,
unpub.; Refstie and Gjedrem, 1975).
Thus, the potential for interbreeding with wild fish populations exists
only for the coho salmon. Operators of coho net-pens in Puget Sound maintain
their own brood stock (e.g., Domsea Farms) or obtain smolts from Puget Sound or
Columbia River hatchery stocks.
There are theoretical grounds for avoiding interbreeding of cultured and
wild populations. First, salmonids tend to evolve genetically discrete and
ecologically specialized subpopulations (see Stahl, 1983 for references). If a
population has remained in a particular habitat for a sufficient length of
time, natural selection may have led to the development of characteristics
optimally adapted for that habitat. These same characteristics are unlikely to
be equally adaptive if these individuals are placed in another habitat, as hap-
pens, for example, when individuals for culture are obtained from distant
sources.
Secondly, cultured individuals may have been selectively bred to promote
the development of one or more characteristics desirable in the culture envir-
onment. Selective breeding of salmonids for high fecundity, large egg size,
high hatching percentage, rapid growth, altered rates of maturity, high temp-
erature tolerance, disease resistance and other characteristics has been very
successful (Ackefors and Rosen, 1979). Domsea Farms in Clam Bay currently
raises coho salmon which have been selectively bred for improved growth effic-
iency. Selective breeding of molluscs is still in its infancy, although there
is great potential (Newkirk, 1980). Efforts are in progress to improve resis-
tance to summer mortality of oysters grown in Washington (Hershberger, et al.,
1984). While such traits may be desirable to improve the profitability of
culture operations, these individuals may be less able to survive the rigors of
the natural environment beyond the confines of the culture facility.
Thirdly, the maintenance of a high degree of genetic variability may be
critical to the survival of a wild population since this variability provides

the plasticity needed to respond to changing environmental conditions. For a
variety of reasons (founders' effect, genetic drift, genetic bottlenecking,
intentional and unintentional selection) genetic variability may be reduced in
cultured populations. Many studies have documented a lower genetic variability
in hatchery trout stocks than in wild populations (Allendorf and Phelps, 1980;
Ryman and Stahl, 1980; Stahl, 1983), and there are indications of the same
phenomenon in cultured oysters (Wilkins, 1976). This loss of variability,
often through inbreeding, has resulted in the development of deleterious traits
in some hatchery-reared rainbow trout and Atlantic salmon stocks (Aulstad, et

al., 1972; Kincaid, 1976a; 1976b; Ryman, 1970). However, there is no indica-
tion at present that this loss of variability has occurred in Pacific salmon
(specifically chinook). About 65 genetically distinct stocks have been iden-

tified along the west coast of North America, and there is no evidence that
stocks of hatchery origin have any less variability (as measured by relative
heterozygosity) than wild stocks (C. Mahnken, NMFS, pers. comm.).
This discussion has been derived primarily from the literature pertaining
to hatchery rearing and release programs. With floating mariculture opera-
tions , however, fish are held until they reach a marketable size and then har-
vested. Interbreeding with the wild populations becomes an issue only if
cultured fish escape. Since there are no data available on the extent of
escapes from Puget Sound facilities, it is impossible to evaluate the signif-
icance of the issue. Obviously a critical variable in evaluating potential

genetic effects is the number of cultured fish that escape relative to the size
of the wild population.
The greatest potential for genetic alteration of a wild population would
exist if escaped fish congregated to spawn in one particular area, and there-

fore comprised a larger proportion of the total breeding population in that
area. The likelihood of congregation depends upon how strongly cultured fish
imprint to the culture area. Cultured fish are typically placed in net-pens at
about the time of smolting when the tendency to imprint is the greatest. Even
if placement in salt water is delayed beyond smoltification, there appears to
be a long-term imprinting process if the fish are held in the net-pens for an
extended period. To evaluate homing ability of net-pen cultured fish, coho
salmon were intentionally released from net-pens in Clam Bay. Some of these
fish returned to Beaver Creek, the nearest freshwater stream to the net-pen
site (C. Mahnken, NMFS, pers. comm.).

There is also a question of potential for survival of cultured fish in the
wild. Cultured fish have become accustomed to a predator-free environment with
regularly-provided pelleted feed. It is unknown how well these fish adapt and
survive outside of the culture environment, but they would appear to be at a

competitive disadvantage, at least in the short-term. Hatchery trout have been
found to have a lower survival rate than wild fish (Chilcote, et al., 1985;
Reisenbichler and McIntyre, 1977). Of particular interest have been studies on

the growth and survival of the progeny of crosses between hatchery and wild

steelhead (Reisenbichler and McIntyre, 1977). In hatchery conditions, hatch-
ery/hatchery crosses showed the highest growth and survival. However, when

placed in natural streams the wild/wild fish showed the highest survival and
the hatchery/wild fish had the highest growth rate when significant differences

were found.
In general, cultured fish would appear to be at a competitive disadvan-

tage, but the relative extent to which selective pressures operate on the
escaped fish (thereby preventing interbreeding) or on their progeny is unclear.
In the worst case, there may be a temporary reduction of reproductive capacity
of the wild population since reproductive effort may be wasted in producing
less fit genotypes against which selection may occur. The potential for such
an impact depends upon the number of fish escaping, and may be significant only
when the proportion of escaped fish is large in comparison to the wild breeding
population.

Conclusions

Assessment of the potential for genetic alteration of wild populations by
escapes from mariculture facilities is difficult given the lack of reliable
field data and the speculative nature of the whole issue. However, to put the

potential problem into perspective, one must consider the history of fisheries
enhancement programs, particularly for salmonids. For nearly a century, fish-
eries management agencies in Washington have routinely been transferring
hatchery-reared fish between river systems to improve commercial and recrea-
tional catches. In addition, many net-pen facilities are used for delayed
release, and the number of fish released in these programs is substantial
(e.g., 3 million smolts per year at the Squaxin Island facility alone). Hatch-
ery and net-pen releases are not permitted in a few areas (e.g., Skagit Bay)

because of concern for potential effects on wild stocks. However, large parts
of the Puget Sound Basin are managed for one or more species on a hatchery

basis. The numbers of fish that could escape from mariculture facilities is

dwarfed by the number of fish intentionally released in fisheries enhancement
programs.
Cultured organisms may be at a competitive disadvantage in comparison to
wild organisms for several reasons: 1) they are not pre-adapted to the hab-
itat; 2) they may have been bred for characteristics desirable under culture
conditions, but maladaptive in the wild; and 3) they may have reduced genetic
variability, thus limiting their abilities to cope with environmental change.
Hatchery steelhead, for example, appear to have poorer survival than wild
steelhead (Reisenbichler and McIntyre, 1977), may be more susceptible to

disease (Allendorf and Phelps, 1980) and their spawning cycle may not be
attuned to the local environment resulting in release of offspring under sub-
optimal conditions (Chilcote, et al., 1985). Thus, there may be some selective

pressure against fish which escape from mariculture facilities. In the event
of escapes, the greatest effect, if there is any effect at all, may be a tempo-
rary loss of some reproductive capacity in the wild population. The magnitude
of this effect depends upon the relative size of the breeding population of
escaped fish in comparison to the wild breeding population.
The measures which a culture operator might take to minimize interbreeding
between cultured and wild populations are very limited. First, the number of
escapes should be minimized, obviously standard procedure for any successful
operator. Secondly, an operator could choose a cultured stock that is either:
1) genetically similar to the wild population so that interbreeding would en-
tail minimal risk of introduction of maladaptive traits; or 2) poorly adapted
to survival in the wild so that reproduction of the cultured individuals under
natural conditions would be unlikely. Finally, continued research should be
directed towards development of sterile stocks for cultivation through proce-
dures such as hybridization, hormonal sterilization or creation of triploid
animals (Chevassus, 1983; Purdom, 1983; Stanley, 1979).

3.10 TOXICANTS

Question

Does the use of chemicals for disease control in culture facilities pose a
threat to the surrounding environment?

Discussion

In the event of a disease outbreak among net-pen cultured salmon, it is
necessary to use an antibiotic to stop the spread of the disease and minimize
fish mortalities. There are presently three antibiotics licensed by the Food
and Drug Administration (FDA) for use on food fish: oxytetracycline (also known
by the trade name of Terramycin), sulfamerizine and Romet 30 (a potentiated
sulphonamide). Of these three, oxytetracycline is the most widely used in
Puget Sound, therefore the remainder of the discussion will be focused on this
chemical. No chemicals are used for disease treatment in suspended mollusc
culture.
Oxytetracycline (OTC) is a broad spectrum antibiotic useful in the treat-
ment of many fish diseases including general and hemorrhagic septicemia, fur-
unculosis, bacterial gill diseae, columnaris, vibriosis and enteric redmouth
(Austin, 1985). It is not used on a routine basis for disease prevention, but
only as needed for disease therapy. In Puget Sound its use is generally re-
quired in the summer months, typically 2-3 times a year (C. Mahnken, NMFS,
pers. comm.). It is provided to the fish in the feed at a concentration of

50-75 mg.kg-1 body weight.day-1 for a period of 10 days (Austin, 1985; Herman,
1969). The FDA does not allow food fish to be sold with detectable concentra-
tions of the antibiotic in the tissue, therefore a 21 day waiting period is
required after cessation of usage before the fish may be harvested.
OTC is widely used in human medicine because of its low toxicity, clinical
flexibility in dosage and administration, and its effectiveness against a wide
variety of bacteria, viruses and even some protozoa and metazoa. The antibiot-
ic is also widely used in agriculture as a feed supplement in the diet of ani-
mals such as cattle, chickens and pigs. However, unlike its use in mariculture
and human medicine, in which the antibiotic is used only in disease treatment,
it is given to livestock on a continuous basis to prevent diseases and improve
growth rate.

Assessment of the environmental consequences of OTC usage in net-pens is
severely hindered by a lack of available information on fate and effects of the
antibiotic in the marine environment. In addition to the literature reviewed

for this study, a computerized literature search on the drug (provided by J.
Forster, Sea Farm of Norway) has been reviewed, the Washington Department of

Agriculture has attempted to obtain information from the Food and Drug Admin-

istration and I have made inquiries with the manufacturer (J. Hilton and C.
Downing of Pfizer, Inc.). All efforts have met with little or no success to
date. Thus, all conclusions below must be regarded as tentative, having been
based on limited and in some cases conflicting data.
Despite extensive use of OTC in treatment of fish diseases both in the

United States and elsewhere, as of yet there has been no demonstration of sig-
nificant environmental effect. Many investigators have raised the issue of
potential effects of antibiotic use in mariculture, although almost without
exception the discussions have been entirely speculative, with no evidence of
observed effect and little or no data provided to argue against an effect
(Anonymous, 1983b; Beveridge, 1984; Midtlying, 1985; Pedersen, 1982; Solb6,
1982). I am aware of only one publication (Austin, 1985) which addressed the
issue through field studies at a fish farm, and in this case the potential
environment effects were left largely unresolved. There appears to be three
potential environmental effects of concern: 1) inhibition of microbial de-
composition of organic wastes; 2) development of OTC-resistant bacteria; and 3)
accumulation of antibiotic residues in shellfish. These three issues are ad-
dressed individually below.

Inhibition of microbial decomposition of wastes-

OTC retards bacterial growth not only of the target bacterial pathogens,
but of a wide spectrum of marine microbes. The natural bacterial community
involved in nutrient cycling can also be affected by the antibiotic, but the
limited evidence available suggests that these organisms can tolerate rela-
tively high concentrations of OTC with little loss of activity. Marine bac-
teria responsible for sulfur oxidation and the oxidation of ammonia to nitrite/
nitrate are unaffected by concentrations of OTC up to at least 10 mg-1
Almost complete loss of activity occurs at a concentration of 100 mg 11I
(Carlucci and Pramer, 1960). Bacteria responsible for production of ammonia

have shown about a 30% loss of activity at concentrations of OTC between I and

100 mg-l
The is sue of OTC effects on the natural microbial community becomes a
question of the concentration and persistence of the antibiotic. Given the
rapid dilution of dissolved substances that occurs in a net-pen environment, it
would seem that the only opportunity for OTC to remain at concentrations high
enough to elicit a bacteriostatic effect would be if the substance remained
associated with the particulate material (e.g., feces and excess feed). How-
ever, OTC is readily soluble in seawater (Musselman, 1956). At a pH and temp'-
erature typical of the marine environment, 4-20% of the OTC leaches from dry

feed pellets within 15 minutes; a leaching rate constrained by the surface area
of the pellet rather than the solubility of OTC (Fribourgh, et aLI., 1969).
Thus, it seems unlikely that significant quantities of OTC would persist around
a net-pen site. The effect of the antibiotic on the non-pathogenic bacteria
community would probably be, at most, a very short-term phenomenon.

Development of OTC-resistant bacteria strains -

Use of an antibiotic will encourage proliferation of those strains of
bacteria having a natural resistance to the substance. The presence of anti-
biotic-resistant bacteria has been demonstrated in the vicinity of effluent
discharge from antibiotic manufacturing plants (Cornelson, et al., 1958),
hospitals (Peou, et al., 1981) and domestic sewage treatment plants (Bell, et
al., 1981). Use of antibiotics in fish culture has resulted in the appearance
of resistant bacterial strains in the vicinity of the culture (Aoki, et al.,
1980; Austin, 1985; Bullock, et al., 1974).
Austin (1985) studied the development of antibiotic-resistance in bacteria
isolated form trout farm effluents, and monitored the duration of this resis-
tance. Before treatment with OTC there was no OTC-resistance among the bacter-
ial community. During OTC treatment 90% of the bacteria strains examined show-
ed resistance to the antibiotic. However, within 9 days after cessation of the
treatment, all OTC-resistance had been lost. Austin suggested this may indi-
cate that antibiotic-resistance is a short-term phenomenon, and that this
resistance is lost when antibiotic usage ceases. There is limited evidence of
this from human evidence (Forfar, et al., 1966), but there are insufficient
data at the present time to draw such a conclusion. Since Austin's studies

were conducted in a flow-through system, there is some question as to whether
he was monitoring anitbiotic resistance in the same population over time.

In agriculture, antibiotics are supplied to the animals on a routine basis
for disease prevention and growth enhancement. The limited, therapuetic usage
of antibiotics in net-pen culture would tend to lessen the likelihood of sig-
nificant, long-term effects resulting from antibiotic usage. Net-pen operators

typically use OTC only 20-30 days out of the year (2-3 treatments of 10 days
each), thus lessening the selective pressure for development of antibiotic

resistance in the microbial community.
The issue of microbial resistance has always provoked debate over antibi-
otic usage whether in human medicine or agriculture (see for example Solomons,
1978; Van Houweling and Gainer, 1978). On the one hand, there are immediate
and obvious benefits of antibiotic usage. However, the more they are used, the
less likely they are to be effective. Generally, the course which has been

followed in medicine is to use antibiotics as judiciously as possible. There
are still many unanswered questions regarding antibiotic usage in human medi-
cine: the unknowns are even greater in fish culture.

Ac-cumulation of OTC residue in shellfish -

Since the FDA prohibits sale of organisms having measurable residues of
OTC in their tissue, accumulation of the antibiotic in shellfish could prevent
their harvest. However, the probability of this presenting a problem appears
remote. First, dilution would probably reduce the concentration of OTC to im-

measurable levels within a very short distance of the culture facility. As an
example, one may make a rough approximation of the OTC concentration in the
water passing through the 250,000 kg net-pen operation discussed in Section

3.2. If the fish are treated with 75 mg OTC-kg body weight-day ,a daily
total of 19 kg OTC would be introduced into the net-pens. Assuming no reten-
tion by the fish, and all OTC being dissolving in the parcel of water which
passes directly through the pens during the day (72 m .sec' see Tables 4 and
-1
5), the OTC concentration in that parcel of water would be only 0.003 mg-l-
Secondly, OTC shows little potential to bioaccumulate so it is unlikely
that shellfish would concentrate the antibiotic in their tissues above its
concentration in the surrounding water. The potential for bioaccumulation of a
compound is directly correlated with its water solubility (Kenaga and Goring,

1980). Given the fact that OTC is readily water soluble, it is unlikely to be
concentrated in shellfish tissue. The fact that OTC-treated organisms lose
their OTC body burden within a matter of a few days to one month (Corliss,
1979; Herman, et al., 1979), provides further evidence of little bioaccumula-

tion potential.

Conclusions

Very little data were available with which to evaluate the environmental
consequences of OTC usage in net-pen culture, thus, all conclusions must be
considered tentative. However, on the basis of the available evidence OTC
usage does not appear to be of major environmental concern given the infre-
quency of its use, its high water solubility, the rapid dilution which could be
expected, and the limited mariculture development anticipated in Puget Sound.

100

4.0 MODELING OF ENVIRONMENTAL EFFECTS

4.1 INTRODUCTION

Models capable of predicting the environmental effects of mariculture
would be extremely valuable both to environmental managers and the industry.
With such models one might describe a priori the extent of changes in water
chemistry or phytoplankton productivity which may occur should a proposed cul-
ture operation be permitted. The carrying capacity of a region could be deter-
mined, and steps taken to insure that the number of culture units allowed in
this area does not exceed this capacity. The culture operator could use sim-
ilar models to determine the maximum stocking density given the limitations of
waste removal and oxygen supply.

Unfortunately, models capable of predicting the environmental effects of
mariculture are, at best, only in very preliminary stages of development, and,
in general, have not been tested. The development of new models or the refine-
ment and testing of existing models require major research efforts, far beyond
the scope of this study. It is not the intent of this report to provide defin-
itive models, for they do not exist. However, the types of models currently
available are reviewed and evaluated in order to provide a basis for further
work. Primary emphasis has been given to those models specifically developed
for, or applied to, mariculture, but several models developed for other situ-
ations are also evaluated with respect to their potential applicability to
mariculture. Two environmental effects of mariculture which may be amenable to
modeling are changes in water quality and the accumulation of organic-rich sed-
iments under the culture facility.

4.2 WATER QUALITY MODELING

Many models have been developed to predict effects of nutrient inputs on
water quality and productivity in the receiving water body (see Jorgensen,
1979; O'Connor, 1981; and Russell, 1975). Some investigators have applied
these models to aquaculture in lake systems (Beveridge, 1984; Beveridge and
Muir, 1982). However, the modeling of nutrient inputs to a lake is a far eas-
ier task than modeling inputs to Puget Sound. In a lake the system boundaries
can readily be established, and the surface area and the volume of the affected

101

water mass are easily defined. In Puget Sound, most suspended culture sites
are located in open, unconfined areas, and tidal exchange complicates quantifi-
cation of the affected water body.
Without the convenience of a relatively small, defined body of water like
a lake, the simplest approach to modeling water quality effects from net-pens
or suspended shellfish culture is to quantify effects on the parcel of water
which passes directly through the culture structure. Water flow is assumed to
be laminar, and no mixing of the water passing through the culture structure
with the surrounding water mass is assumed to occur. These two assumptions are
most assuredly always violated, and, therefore, severely limit the validity and
utility of the models as they are presently formulated. However, in the ab-
sence of any better approach, this approach was used in Section 3.2 and has
been used by many other investigators (Beveridge, 1984; Inoue, et al., 1966;
1970; Incze and Lutz, 1980; Kils, 1979) to obtain estimates of water quality
changes.
Most investigators modeling the water chemistry changes occurring in a
parcel of water as it passes through a culture structure have done so to max-
imize production. For example, the technique can be used to determine the
maximum culture intensity achievable given the oxygen or food content of the
incoming water (e.g., Beveridge, 1984; Incze and Lutz, 1980; Kils, 1979). The
same approach can be adapted for purposes of environmental protection.
A model to determine the number of mussel longlines which could be sup-
ported by a given seston concentration was developed by Incze and Lutz (1980)
and modified by Rosenberg and Loo (1983). The longline system modeled has a
depth (h), and a width (w) equal to the length of the longlines. The longlines
are oriented perpendicular to the current. The following terms are used:

A: (m2), area of the culture system normal to the flow, A=w~h
V: (m.hr-1), velocity of water mass entering normal to face A
N: (1.hr-1), volume of water entering face A per unit time, N=V.A.103

Sk: (mg.l-l), seston concentration flowing into longline k, k=1,2,...,n
M: (individuals), number of mussels suspended on each longline

F: (1.hr-l-individual-1), filtration rate of a mussel

The mussels will filter water at a rate of F-M 1-hr- with passage through

each longline, thus, S1-F-M mng-hr- is the rate at which the first longline

102

will filter seston from the water. The rate at which seston enters the first
-1
longline is S1 'N mg-hr . Therefore, the rate at which seston leaves the first
longline is the rate at which it enters minus the rate at which it is filtered
-1
by the mussels, or (S1. N)-(S1. F-M) mg.hr . This is equivalent to the rate at

which seston is available to the second longline, S2 N mg.hr . Thus:

S2-N = (SI'N)-(Si F M) (Equation 1)

Rearranging this expression to obtain the concentration of seston entering
longline 2 yields:

Si(tN-F.M)
S2 =(N-FM) (Equation 2)
N

Likewise, the concentration of seston entering longline 3 is:

S MS2(N - F- M) (Equation 3)
N

Substituting Equation 2 for S2 in Equation 3 yields:

S= SI [N - F M]2 (Equation 4)
N

The concentration of seston entering longline k is:

Sk- S [N -F-wM] k-I (Equation 5)
N

The same model can be used to approach the problem from a different
perspective. If the size of the culture operation and the minimum acceptable

seston concentration are known, one can determine the current velocity needed

to support such a system and suitable sites may be identified. Rearranging
Equation 5 yields:

k-I Sk N-F M
- - S. N

k-IS
k-I k N-FM
SI
N k: IV S -k

N i---V- =F.M
N ~[IFM

103

V-A- 103 N~ FM (Equation 6)
k-IS
I-~

Rosenberg and Loo (1983) used Equations 5 and 6 to develop the relation-
ships shown in Figure 8. Given the current velocity and the desired number of
longlines, the model can predict the seston concentration needed to support the
operation (Equation 5). Alternatively, if the seston concentration and number
of longlines are known, the required current velocity can be determined (Equa-
tion 6).
This approach represents the beat modeling tool currently available for
managing the water quality in the vicinity of floating mariculture operations.
Environmental managers and mariculture operators may find the approach useful
to model depletion of dissolved oxygen or accumulation of ammonia. If thresh-
old concentrations for these parameters can be established (either from the
perspective of protecting the aquatic environment or the health of the cultured
fish) and current velocities are known, one could predict the size of the cul-
ture operation which could be established without violating these standards.
It should be recognized that these standards are applied to the parcel of water
passing through the culture structure and not the receiving water body as a
whole, since the effects of dilution are not considered. Alternatively, for an
operation of known size, one could determine if a site was suitable, i.e., had
sufficient current velocity or adequate dissolved oxygen to support the opera-
tion. The approach was developed for mussel longlines, but could readily be
adapted for net-pen farming of salmon.
While the approach may eventually provide first-order approximations of
the water chemistry changes that occur in water bodies with passage through
culture structures, it should be recognized that it is untested. An evaluation
of the approach In the marine environment is recommended before the model is
used as a basis for decisions pertaining to siting or facility operation.

104

6- I -3
E o
I ,

EI~~ v~~E
-2B o
4- -2 a
z w
w w

z 9,{-I
w

0 0
I_ / ,/_

o~ I I I I I I I I 0
0 20 40 60 80
NUMBER of LONGLINES

Figure 8 - The seston concentration (A) and current speed (B)
needed to support a given number of mussel long-
lines. From Rosenberg and Loo, 1983.

Model parameters

A B

w: (m), length of longline 180 180

h: (m), depth of mussel strings 6 6

V: (m.hr- ), current velocity 36 Y

M: (individuals) mussels per longline 1,120,000 1,120,000

F: (1lhr -lindiv.-) filtration rate 3.2 3.2

S1: (mg.-l ) initial seston concentration Y 1.1

Sk+l: (mg.-l) minimum acceptable seston 0.05 0.05
concentration after passage through
all longlines.

k: number of longlines (k+l should be sub- X X
stituted for k in Equations 5 and 6)

105

4.3 SEDIMENTATION MODELS

The general goal of modeling sedimentation associated with mariculture Is
to determine the depth, rate and areal extent of accumulation of feces, pseudo-
feces or excess feed on the sea floor. Such models might be used in siting
culture facilities to minimize effects on the benthos or to establish adequate
distances from certain critical habitats. While the objective of sedimentation
modeling is relatively straightforward, the accomplishment of this goal is not
an easy task. The development of a sedimentation model requires the inclusion
of many parameters that are very difficult to quantify at present. The param-
eters necessary may be grouped into the general categories of particle charac-~
teristics and fluid flow characteristics.

Particle characteristics

When a particle is placed in a non-moving fluid it accelerates toward the
lower boundary of that fluid in response to gravity. However, as the velocity
of the particle increases, so do the drag forces opposing the downward move-
ment. When the gravitational and drag forces are in balance the particle falls
at its terminal velocity, known as the settling velocity. Quantification of
particle settling velocity is critical to any sedimentation model.
Settling velocity may be determined theoretically by Stoke's or Newton's
equations, or empirically in the field or laboratory. Each approach has inher-
ent advantages and disadvantages. The theoretical approaches (i.e., analytical
equations) based on physical properties are preferred, but at present are well-
developed only for solid, spherical particles. The empirical approaches are
costly and the results do not generalize to conditions or sites other than
those for which the data were collected. However, given the complex nature of
the particles produced by mariculture operations, empirical determination of
settling velocities is the only feasible approach at present.
Estimates of the sinking speed of particulate material from trout and
yellowtail culture have ranged from 2-12 cm-sec- (Collins, 1983; Hagino, 1977;
ligura, 1974). However, these estimates of settling velocity were empirically
derived and are not necessarily applicable to other sites wherre flow condi-
tions may be very different. The use of a single number to quantify the set-
tling velocity of the particles from a mariculture facility is misleading and

106

inaccurate, since these particles are likely to encompass a wide range in
settling velocities. In still water the standard deviation of the settling
velocity may be half as great as the mean, and may exceed the mean value when
currents are present (liagino, 1977).
The determination of particle settling velocity by an analytical model

will require quantification of the many individual parameters affecting set-
tling velocity. The settling velocity of an individual particle is a function
of particle shape, size and density. The concurrent settling of multiple par-
ticles is also affected by cohesive or adhesive forces between the particles
and by the concentration of the particles in suspension.
Shape - Settling velocity is dependent upon particle shape, and most an-
alytical sedimentation models assume the settling particles to be perfect
spheres. Feed and fecal material are particle aggregates whose shapes are not
accurately described by spheres. Nevertheless, most existing analytical models
require that their shape be approximated as such.
Density - The densities (i.e., mass to volume ratio) of individual partic-
les also affects their settling velocities. Aggregate densities are extremely
difficult to determine, and one mean value is not likely to be an adequate
representation for the complex mix of particles from mariculture. These aggre-
gates are also porous. The porosity of fecal material affects determinations
of both particle density and settling velocity. Both effects are difficult to
quantify. In order to apply analytical sedimentation models it will probably
be necessary to model the particles as solid objects, but the potential arti-
facts introduced must be recognized.
Size - Given constant particle density, settling velocity increases with

particle size. Modeling the transport of particulate material from mariculture
is complicated by the fact that the size of the particles encompasses a very
broad range. Furthermore, the size of a given particle may change throughout
the descent as aggregates disintegrate or reform.
Cohesion/Adhesion - Particle aggregates may be formed by either cohesion
(electrostatic interactions) or adhesion ("gluing" of particles by a binding
agent). The organic aggregates from mariculture are both cohesive and adhes-
ive, although most sedimentation models are incapable of addressing these par-
ticle interactions.
Concentration - The rate at which particles settle in a fluid is dependent
upon their concentration. Therefore, it is necessary to determine the concen-

107

tration of particles beneath a culture operation in order to chose the appro-
priate model. At low concentrations the individual particles are sufficiently
far from one another that no interactions occur. As concentrations increase a
phenomenon known as "hindered settling" occurs in which interaction between
particles reduces the settling velocity from that of the particles settling
independently. At still higher concentrations "group settling" causes the par-
ticles to fall much faster than they would independently. Group settling is
not likely to occur given the particle concentrations anticipated in mari-
culture.
Input rate - Models may be developed either for a pulsed input of partic-
ulate material or a continuous input. The later type would probably be most
appropriate for application to mariculture, although there is a pulse component
(i.e., feeding intervals) in the particulate input.

Fluid flow characteristics

Flow regime - Flow may be described in one, two or three dimensions, de-
pending on the desired complexity of the model. At a minimum the model should
account for hi-directional tidal flow in one dimension. Two dimensional models
would permit the contouring of sediment accumulation around a culture facility.
Water movement in the third (vertical) dimension may be important in sediment
transport under some conditions, but is ignored in most models.
The forces on a particle and its movement in a fluid are highly dependent
on the viscosity and density of the fluid. Both parameters are dependent upon
temperature and salinity, and are two of the few necessary model parameters
that are easily quantified. The water column may be stratified by density and
a sedimentation model should be capable of quantifying sediment transport in a
vertically stratified density gradient if it is to have broad application.
The simpler sedimentation models assume that water depth is constant with-
in the area of potential deposition. However, it would be preferable to use a
more complex model capable of predicting particle accumulation on a sloping or
uneven bottom.
Benthic boundary conditions - The probabilities of deposition and resus-
pension are determined by boundary layer structure and bed shear stress, both
of which are functions of the near-bottom flow regime. The bed shear stress is
in turn dependent upon bottom topography, bottom roughness and other factors.

108

In order to demonstrate how particle and fluid characteristics are incor-

porated in sedimentation models, and to assess the types of models currently
available, various sedimentation models are evaluated with respect to their
applicability to modeling of sediment input and transport from Mariculture. A

review of all available models has not been attempted. Instead, several model
types representing alternative applications are discussed. The models reviewed
have been developed for ocean dumping, manganese nodule mining, estuarine sedi-
ment transport and aquaculture.

Ocean dumping

Many models have been developed for ocean disposal of wastes including

dredged material (Brandsma and Divoky, 1976; Stoddard, et al.,1985), mining
wastes, (Bigham, 1986), drilling muds (Brandsma, et al., 1983) and sewage
sludge (Koh, 1971; Koh and Chang, 1973). Ocean dumping typically incorporate
two phases of particle descent. The first phase is convective descent. When
the material is released to the ocean, the particle concentration is high, and
the density of the plume is likely to be much greater than that of the sur-
rounding seawater. Because of "group settling" the plume tends to settle as a
single, large mass. The settling velocity of the plume may be several orders
of magnitude greater than the settling velocities of individual particles
acting independently. As the plume descends it becomes less dense through
entrainment of the ambient ocean water. If the water column is density-
stratified, the convective descent of the plume may terminate at some point of
neutral bouyancy at which the density of the diluted plume is equal to that of

the surrounding seawater.
The second phase of particle transport is passive dispersion. At this
point the plume is assumed to have reached a point of neutral bouyancy. Fur-
ther particle dispersion occurs as a result of currents and turbulence. Typi-
cally, most models will consider only two dimensions in this phase where, other
than particle settling, vertical water movement is neglected.
The phenomenon of convective descent is entirely inapplicable to the mod-
eling of particle transport from suspended mariculture. If ocean dumping mod-
els are to be applied to mariculture, this phase of particle settling can be
ignored.

109

An additional problem with most of the ocean dumping models is that they
are based on a single input of particles over a very short time period (e.g.,
the release of dredged material from a barge). These models would be inappro-
priate to the long-term, essentially continuous release of particulates from
mariculture. Trajectory models have been used to predict the dispersal of
drilling muds when the discharge has negligible initial momentum or bouyancy
and the discharge continues for long periods with variable currents (Brandsma,
1984; Runchal, 1978), and might be applicable.

Manganese nodule mining

Lavelle et al. (1981) developed a model to predict dispersion of the sed-
iment plume generated during manganese nodule mining in the deep sea. As the
nodule collector moves across the bottom, it resuspends silts and clays. The
model predicts the depth of redeposition as a function of lateral distance from
the collector track.
One advantage of this model to those discussed previously is that a con-
vective descent phase is not included in the model. The mining process does
create a high density plume which goes through a stage of gravitational col-
lapse immediately behind the collector. However, the model only attempts to
predict the passive dispersion which occurs following the convective descent
stage.
The model has a number of assumptions and limitations which make it un-
suitable for application to mariculture without substantial modification.
First, it assumes a constant, unidirectional current and a uniform settling
velocity for all particles. Both assumptions may be reasonable in the deep sea
but inappropriate in coastal environments. Secondly, the model utilizes an
empirically-determined settling velocity. While this does not preclude its
applicability to mariculture, it should be recognized that an appropriate
empirically determined settling velocity must be obtained whenever the model is
used. Thirdly, there is no provision for continuous sediment input. It is
assumed that any given point on the bottom is subject to accumulation from only
a single sediment pulse. Fourthly, sediment input to the model is in the form

of a concentration gradient of particles near the bottom, with no particles
more than 5 m above the sea floor. Such a sediment distribution would not be
expected in the case of floating mariculture, and the input conditions must be

110

modified as appropriate for mariculture. Finally, there is no provision for
resuspension of particles once deposited. Such a simplifying assumption may be
necessary in a first generation model, but the limitations should be recog-
nized, and the possibility of resuspension and redistribution of sediments from
mariculture should be considered.

Estuarine sediment transport

Complex numerical (i.e., empirical) models have been developed to describe
sediment transport in estuarine environments (e.g., Ariathurai, et al., 1977;
Sheng, 1983). The modeling of estuarine sediment transport is considerably
more difficult than modeling of sediment dispersion from floating mariculture.
Given the input of fresh water, fluid density differences are of particular
concern in estuarine modeling. Unlike mariculture, sediments are introduced
into an estuary within a fluid (the river) having a very different density than
the receiving (marine) waters. This density difference has profound conse-
quences on the time and path of particle descent. In addition, as salinity

increases the fine particles tend to aggregate. The effective particle size,
and consequently the settling velocity, may be increased. Finally, estuarine
sediment transport models must take into account the influence of wind forcing.
This factor will probably be neglected in initial mariculture models.
Despite the added complexities of estuarine sediment transport models,
they can serve as patterns for development of numerical models for mariculture:
1) sediment input is continuous rather than single or pulsed events; 2) sedi-
ment concentration is low in comparison to other inputs such as dredged mater-
ial disposal, thus avoiding the modeling problems associated with group set-
tling; and 3) the settling behavior of cohesive sediments must be considered.
No single estuarine sediment transport model currently in use is readily avail-
able for application to mariculture, and the modification of existing models
would require substantial effort.

Aquaculture

Hagino (1977) developed a simple model to predict the dispersion of feces
and unutilized feed from net-pen culture of yellowtail in Japan. Rather than
modeling the complex physical processes involved in determining settling vel-

111

ocity, this parameter was determined empirically by the use of sediment traps
placed around the net-pens. The settling velocity of particles from the net-
pens was found to encompass a very broad range. The range of settling veloc-
ities was quantified as a normal distribution with a mean and standard devia-
tion of 7.5 cm-secI and 10-13 cm-sec- I, respectively.
The extent of sediment accumulation as a function of distance was
determined based on the formula:
u-h
X. =
Iw.

where xI equals the horizontal distance from the net-pen traversed by the par-

ticle, u equals the current speed, h equals the water depth and w.i equals the
particle sinking speed. Using a probability function of settling velocity in
place of w, -and assuming a constant, unidirectional current, an estimate of
sediment accumulation as a function of distance was obtained (Figure 9a). The
extent of sediment accumulation in two directions from the net-pen (Figure 9b)
was estimated by modeling tidal currents as a sine function. With field data
on the current regime in the vicinity of the net-pens, contours were developed
to predict the accumulation of feces in all directions. The distribution of
accumulated sediment predicted by the model was found to resemble the actual
depth of accumulations measured in the field.
The model of Hagino (1977) is a very simplified approach to a complex
problem. The major limitation of the approach is that it is empirical, and
therefore, site-specific. No attempt was made to determine the many variables
affecting settling velocity. Values were derived empirically with no knowledge
of how settling may have been influenced by site-specific conditions of current
and turbulence. Therefore, the estimated settling velocities are not necessar-
ily applicable to other sites and conditions. The process of sediment resus-
pension also was not considered by the model.

It is clear that the development of analytical models is a monumental
task. Certain simplifying assumptions (e.g., no vertical water movement, no
particle interactions during settling, particles modeled as solid spheres) may
approximate actual sediment transport processes. This development of analy-
tical models is an active field of research, but has not yet reached the point
where sediment transport related to mariculture can be modeled realistically.

112

(A)
8- DEPTH 15m
z -- - 'THEORETICAL VALUES
o
O-- THEORETICAL VALUES -CURRENT SPEED = 10 cm-sec-'
w r 6- o-----o CALCULATED VALUES SETTLING VELOCITY-

ctI - 3s MEAN = 7cm sec-
w 0:. 4 ,STD. DEV lOcm. sec

. -J 2-

~-0
5 10 15 20 25 30
DISTANCE FROM SOURCE (m)

(B)
NORMAL DISTRIBUTION
4-k .CALCULATED DEPTH i5m
~z ~~A 1 / VALUES SETTLING VELOCITY-
0K ,-'---re/ MEAN =75 cm'sec"
. *u �eSTD. DEV = I0 cm sec1
CURRENT VELOCITY (U)
O0 2- U UD Sjii
L .o 2-'oN- U Uo sin-t

! Il -
w I < / < T= 12 HOUR

I.-
I I I I I I II
-20 -15 -10 -5 0 5 10 15 20
DISTANCE FROM SOURCE (m)

Figure 9 - Bottom accumulation of feces and excess feed from a
net-pen facility predicted by a one-dimensional
empirical model. (A): Constant unidirectional
current. Theoretical values determined by a
probability density function of sinking speed.
Calculated values based on empirical settling
velocities. (B): Simulated bidirectional tidal
current. From Hagino, 1977.

113

The presently available empirical models like that of Hagino (1977) may
provide reasonable approximations of actual deposition. Given the present
state of our knowledge of sediment transport processes, the simple, empirical
Hagino model is probably as valid and as useful as any of the more mathemat-
ically complex models, whether analytical or empirical. Empirical models are
limited by site-specificity, but they represent the only approach of immediate
applicability. The Hagino model, or any other empirical model should be used
cautiously, and efforts should continue towards development of models with
better predictive capabilities.
Concurrent with the development of models should be further efforts to
relate the rate of sedimentation with the consequent biological effects. It is
not yet possible to determine what biological changes will occur with any given
rate of sedimentation or organic enrichment. The magnitude of the effect will
be determined largely by the rate of oxygen supply to the sediments and the
tolerances of the specific benthic community to burial, oxygen depletion, etc.
Further studies of benthic communities exposed to varying rates of sedimenta-
tion and organic enrichment are necessary to quantify the relationship. if
".acceptable" rates of sedimentation/enrichment can be established, these rates
can then be integrated into the appropriate sedimentation model.

114

5.0 ENVIRONMENTAL REGULATION OF MARICULTURE

5.1 INTRODUCTION

Mariculture is practiced much more extensively in other countries than it

is in the United States. Therefore, the laws and regulations developed else-
where to minimize environmental effects of mariculture may provide valuable
insight into management of mariculture development in Puget Sound.
A survey of mariculture regulations in other states and countries was con-
ducted . No attempt was made to identify mariculture regulations pertaining to
issues such as navigational rights, property access, recreational or commercial

fishing interests, or sale and public consumption of products. Emphasis was
placed on identifying those laws and regulations intended to minimize altera-
tions to the physical, chemical and biological environment resulting from
establishment of the culture operation. Such regulations can be grouped into
two general categories: regulations which attempt to minimize environmental

degradation through siting and operational restrictions, and regulations which
are intended to prevent the spread of disease or introduction of exotic spe-
cies. Siting and operational restrictions may include requirements regarding
water depth, distance between culture operations, the exclusion of specific
habitats, stocking density, production limits and feeding restrictions. Regu-
lations minimizing opportunity for the spread of disease or the introduction of
species may include inspection of shipments, quarantining, pathological examin-
ations, restrictions on the country of origin or bans on the importation of
certain species.
The states and countries in which marine net-pen culture of salmon and/or
suspended culture of shellfish is practiced were surveyed, since the regulation
of these two industries is most relevant to mariculture development in Puget
Sound. The states surveyed were Washington, California and Maine. Extensive
culture of shellfish, particularly oysters and scallops, is practiced in Cal-
ifornia. Maine is the only other state with significant suspended mussel
culture. There are also several operations in Maine for net-pen culture of
Atlantic salmon. The countries surveyed were Canada (specifically British
Columbia), Chile, Faroe Islands, Japan, New Zealand, Norway, Scotland, Spain
and Sweden. These countries are either major producers of mussels (by suspend-
ed culture) and/or net-pen reared salmon, or which have adopted particularly
stringent requirements to minimize the environmental effects of mariculture.

115

5.2 STATE REGULATIONS

Washington
Sources: Eric Huriburt, Washington Department of Fisheries, pers. comm.

Proposed mariculture facilities are subject to permit reviews by a variety
of government agencies. These reviews include, among many other considera-
tions, the potential effect of the operation on nearby habitats (e.g., kelp and
eelgrass beds, spawning areas). The counties have taken an active role in this
review process, and each has adopted its own approach to project siting. There
are currently no state regulations on permissible depths for mariculture opera-
tions, stocking densities, or proximity to other culture operations. There
are, however, some state agency policies which would affect project siting.
The Department of Social and Health Services prohibits the sale of shellfish
from decertified areas (as determined by fecal coliform counts) or from within

0.5 miles of a sewage outfall. The Department of Fisheries has a policy of
prohibiting net-pens over harvestable shellfish beds.
Introductions of exotic species must comply with the State Environmental
Policy Act (SEPA), through which they are reviewed for possible impacts on
native species and habitats. The importation of fish, shellfish and aquatic
plants, or the transfer of fish within the state is subject to the issuance of
a permit, for which the potential accidental introduction of diseases and
undesirable organisms is assessed. Department of Fisheries policy prohibits
the importation of all live salmonids or their eggs from Europe, except for
eyed Atlantic salmon eggs. The importation of these eggs requires that they
originate from a disease-free facility, they must be surface disinfected, and
they must be held in quarantine for 90 days following swim-up.

California
Sources: Bowden, 1979; Smith, 1985; Earl Smith, Marine Resources Supervisor,
Marine Resources Division, California Department of Fish and Game, pers. comm.

There is presently no net-pen salmon culture in California. There are two
salmon growers, one which holds the fish until they attain marketable size and
the other which is involved in ocean ranching, but both hold the fish in
land-based containment structures (e.g., tanks, raceways). There is, however,

116

extensive suspended mollusc culture in the state including oysters, scallops
and mussels.
Applicants for an aquaculture permit are required to go through a review

process in which the California Department of Fish and Game evaluates the en-
vironmental acceptability of the proposed operation. There are, however, no
formal siting or operational criteria intended to minimize environmental
effects. The Marine Resources Division of the Department of Fish and Game is
unaware of any adverse environmental effects resulting from operation of sus-
pended shellfish culture in California (E. Smith, pers. comm.).
Written permission from the California Department of Fish and Game is
required to import live fish and shellfish into the state. Salmon eggs must
originate from an approved source and be accompanied by a disease-free certifi-
cation. Imported oysters are inspected for drills and other pests. A cul-
turist wishing to introduce an exotic species must demonstrate that a sound
reason exists for the introduction and that the species will not cause harm to
native species. The state has recently adopted guidelines for the importation
of exotic species proposed by the International Council for the Exploration of
the Sea (ICES, 1982; Rosenthal, 1985). Complete reviews of California policy

regarding importation of live animals for aquaculture can be found in Bowden
(1979) and Smith (1985).

Maine
Sources: DMR, 1985; Ken Honey, Maine Department of Marine Resources, pers.
COMM.

Both net-pen culture of salmon and suspended culture of mussels are prac-
ti ced in the state of Maine. An applicant for either type of operation must
submit an aquaculture lease application to the Department of Marine Resources.
While there are no specific criteria designed to minimize environmental ef-
fects, the Department does consider potential environmental consequences during
application review. However, while permit requests have been denied by the
state for a variety of reasons, no mariculture permit has ever been denied
because of concerns of environmental degradation. The Maine Department of
Marine Resouces is unaware of any adverse environmental changes in the vicinity
of mariculture operations within the state (K. Honey, pers. comm.).

117

5.4 REGULATIONS OF OTHER COUNTRIES

Canada (specifically British Columbia)
Sources: Frances Dickson, Aquaculture Co-orditiator, Field Services Branch,
Fisheries and Oceans, pers. comm.; Edward Black, Ministry of the Environment,
pers. comm.; Bruce Kay, Environmental Protection Service, pers. comm.

Salmon net-pen culture is rapidly expanding in British Columbia. For
example, in Sechelt Inlet alone there are currently four net-pen facilities
currently in operation, and applications in review for an additional 25 farms.
Such dramatic growth is typical for much of the marine waters within the prov-
ince. About 200 net-pen farms are expected to be in operation in the province
by 1987.
Federal and provincial authorities have begun developing siting criteria,
although because the industry is relatively new, the criteria are frequently
refined and updated. Some of the major siting criteria currently under consid-
eration for adoption include:

1) A new net-pen facility will not be permitted within 0.5 nautical miles of a
major salmon spawning stream or within 0.5 nautical miles of an existing net-
pen operation. This minimum distance is intended to minimize deterioration of
water quality and the potential for disease transmission. The exact distance
is a somewhat arbitrary "best guess", and is not based on firm scientific data.
Provincial authorities are allowing a new farm to locate within 0.5 n.m. of an
existing farm if approval of the existing farm's owner can be obtained, but
federal authorities are not generally advocating this practice.

2) A new facility may not be located closer than 125 m from a shellfish bed,
and there is some consideration being given to increasing this distance.

3) New net-pens may not be located in less than 10 m water.

4) Net-pens may not be located near eelgrass beds, herring spawning areas or
other habitats of special significance.

118

British Columbia also has relatively restrictive policies regarding the
importation of salmon into the province. These regulations include:

1) Pacific salmon may not be imported. Only Atlantic salmon and non-anadro-
mous rainbow trout may be brought into the province.

2) Smolts may not be imported. Eggs brought into the province must be quaran-
tined for a minimum of 12 months, during which the young fish must be inspected
at least four times by a pathologist. The last inspection must occur after
smolting.

3) Imported eggs must originate from approved, disease-free hatcheries. No
importation of eggs is permitted from Europe, the southern hemisphere, or other
countries where viral hemorrhagic septicemiae has been reported or is likely to
be present. All eggs must be surface-disinfected prior to importation.

4) Consideration is being given to halting all Atlantic salmon importation

after March 31, 1989.

Chile
Sources: Ron Zebal, WDF, pers. comm.; Jon Lindbergh, pers. comm.

There are no regulatory restrictions imposed on salmon growers with re-
spect to number of net-pens, stocking density, spacing between operations or
similar siting and operational considerations. The government does not require
any formal review of the potential for environmental alterations in the vicin-
ity of a proposed culture site.
The only regulations intended to minimize environmental effects are those
pertaining to the introduction of exotic species and diseases. The importation
of live fish is prohibited, although eggs may be imported if certified to be
disease-free. Applications for the introduction of exotic species are reviewed
on a case-by-case basis. For example, review of a proposal to introduce the
oyster Crassostrea gigas took five years. The introduction was finally allowed
with the restriction that they not be placed near natural or cultured beds of
the native oyster.

119

Faroe Islands
Sources: Tommy Petersen, Secretary of the Faroese Sea Breeding Commission,
pers. comm.

The Faroe Islands are not major producers of net-pen grown fish; 1985
production was only 1,100 metric tons of Atlantic salmon and sea trout. The
environmental policies of this country pertaining to mariculture were chosen
for review because environmental concerns have were reported to have resulted
in strict regulation of the industry (NOAA, 1985). However, the information I
obtained from the Faroese Sea Breeding Commission, the government agency re-
sponsible for licensing fish farms, does not indicate that environmental poli-
cies are any more restrictive than in many other countries.
The most restrictive government policies pertain to the importation of
live material and the potential introduction of disease organisms. No live
fish can be imported for culture purposes, nor has the importation of eggs been
permitted since 1984.
Applications for a net-pen facilities are carefully scrutinized with re-
spect to the education, experience and economic status of the operator. Envi-
ronmental concerns have played a minor role in permit reviews to date because
the industry is relatively new and few farms are currently in operation. How-
ever, as the industry expands the Faroese government is becoming increasingly
aware of environmental issues, and siting criteria may be developed in the near
future.

Japan
Sources: Kunihiko Fukusho, National Aquaculture Research Institute, pers.
COMM.

There are no national standards regarding siting or operation. Instead,
standards are established by each individual prefecture. The prefectures set
standards as necessary for environmental protection, but primarily from the
standpoint of preventing deterioration of the culture grounds to the point
where the health of the cultured animals is threatened. Guidance of the cul-
turists is followed in establishing the maximum culture intensity achievable in
a given area, optimal spacing between operations, maximum stocking densities,
etc.

120

Since each prefecture has its own siting and operational standards, it is
not possible to provide a thorough review of these requirements here. However,
a few examples are provided.
Kagoshima Prefecture - The culture grounds have been divided into several
subareas as determined by the minimum annual dissolved oxygen concentration.
In Zone A, where oxygen concentrations outside the net-pens reach as low as
5.5 mg.1- (4.5 mg.1-l inside the pens), stocking densities can not exceed
-3 -1
11 kg.m and culture intensity can not exceed 40 metric tons -hectare . Zones
B and C experience lower dissolved oxygen concentrations, and consequently,
permissible stocking densities and culture intensities are reduced. In Zone C
-3
(oxygen concentration undetermined) stocking densities must be 7 kg.m or less
and the culture intensity can not exceed 20 tons.hectare-.
-3
Mie Prefecture - Stocking density in net-pens is limited to 10 kg.m .
Only one 7x7x7 m net-pen is permitted per 700 m2. The use of moist feed is
encouraged over the use of wet feed because of the proportionately lower amount
of wasteage.
Kagawa Prefecture - Phytoplankton blooms are a frequent occurrence in this
area, thus net-pens must be at least 20 m. deep to provide the fish a refuge
from dense phytoplankton populations in the surface waters. Stocking density
-3
is limited to 7 kg.m-.

New Zealand
Sources: John Galat, New Haven Salmon Ranch, pers. comm.; Jon Lindbergh, pers.
Comm.

An applicant desiring to establish a salmon net-pen operation in New
Zealand must submit a permit request and operations plan to the Ministry of
Agriculture and Fisheries for review and approval. While potential environ-
mental effects are one consideration in this review process, there are no
specific siting or operational requirements. There are no water depth require-
ments and, in fact, some operators have net-pens which rest on the sea bottom.
Siting and operational practices which can affect environmental quality are
generally left to the discretion of the culture operator, since deterioration
in environmental quality would be reflected in decreased productivity of the
culture operation itself.

121

There is also minimal regulation of suspended mussel culture with respect
to potential environmental effects. Marlborough Sound, a major culture center
for the green-lipped mussel (Perna canaliculus), is divided into contiguous
lease tracts. There are no restrictions on the number, size or spacing of
longlines or rafts within each tract.
New Zealand has one of the most restrictive policies regarding importation
of fish and shellfish into the country. Concerned about the introduction of
diseases, the government allows no importation of live, fresh or frozen fish.
only processed seafoods (e.g., canned goods) are allowed into the country.
While there are regulatory pathways through which one may apply for special
exemptions, the stringent requirements (e.g., two years holding time in a quar-
antined facility) render the importation of live animals unfeasible in
practice.

Norway

Sources: Anonymous, 1984; Edwards, 1978; Hurlburt, unpub.; NOAA, 1985; NOU,
1985; Knut Senstad, Sea Farm of Norway at Bergen, pers. comm.

Norway is the world's largest producer of net-pen reared salmon. In 1985
the country produced 35,500 metric tons of Atlantic salmon. As of 1984, 575
licences to operate salmon and trout fish farms had been granted and 1,700 more
applications had been submitted.
Many of the early net-pen operations established in Norway were sited in
silled fjords with little water exchange. This siting has resulted in accumu-
lation of so much organic matter beneath the pens that the health of the cul-
tured fish is threatened. Some operators have moved their pens to sites with
better water exchange, while others have begun periodic rotation of net-pen
sites or adopted the use of submersible mixers to disperse the accumulated
solids.
The Norwegians have since become more cognizant of protecting environ-
mental quality when siting new net-pen operations. The government will not
license a proposed facility in an area where flushing is considered inadequate
because of the presence of a sill or other physical factors. A water depth of
10-15 m beneath the pens is considered desirable, although this criterion has
not adopted as a formal requirement.

122

The Norwegian government also sets standards for minimum distance between
culture operations. New fish farms in Norway were previously required to main-
tain a minimum distance of 1000 m from established farms, although this dis-

tance has recently been reduced to 500 m. This minimum distance has been es-
tablished with the intent of minimizing the possibility of disease transmission
between facilities.
Norway imposes a maximum size limit of 4,000 m 3on new net-pen operations

(K. Senstad, pers. comm.; elsewhere quoted as 3,000 m 3 and 5,000 m 3). if,
after several years, the grower is able to demonstrate a successful operation,
he may be granted a license to add additional net-pens and expand holding ca-
3
pacity to 8,000 m . With the exception of operations in place prior to passage
of the law in the mid-1970s, no net-pen operation in Norway is allowed to

exceed the 8,000 m 3limit. While the imposition of this limit may have the
effect of lessening environmental disturbance, it was adopted for socioeconomic
rather than environmental reasons. It was the intent of the Norwegian govern-
ment to exclude the large companies and promote development of the industry by
small growers, thereby alleviating the unemployment in small, isolated fishing
villages along the coast.
The importation of eggs and live fish into the country is allowed after

receipt of the necessary permits. All shipments must be accompanied by a
disease-free certification.

Scotland
Sources: Anonymous, 1984; Marvyn Eddie, Marine Harvest Limited, Edinburgh,
pers. comm.

As of 1983 there were 41 freshwater and 62 marine salmon farms in Scotland
producing 2500 metric tons of Atlantic salmon. By 1985 the production in-
creased to 7,000 tons, and a production of 10,000 tons is projected for 1986.
Applications for new net-pen operations in Scotland must be approved by
the Department of Agriculture, Fisheries and Food. This agency consults other
regulatory bodies, fishermen, yachtsmen and other interested parties prior to
granting an operating license. The potential environmental effects of the op-
eration are considered in this review, although there are no established cri-
teria for siting or operation. Pollution-related issues have not been of major
concern to date, in part because many of the farms are small in size.

123

The Scottish government prohibits the importation of live fish. A license

for the importation of eggs may be obtained if steps have been taken to mini-

mize the potential for introduction of disease with the shipment (e.g., eggs

may not originate from restricted areas, the disease history of the hatchery is

known).

Sweden

Sources: Anonymous, 1982.

Mussel culture operations with annual production in excess of 50 metric

tons and salmon culture operations are regulated under the Swedish Environ-

mental Protection Act. The County Administration examines applications for new

aquaculture facilities or modification of existing facilities in accordance
with the provisions of this Act. While there are no standardized siting or

operational criteria, the County Administration does assess the potential for

environmental alteration, and may require the operator to take actions to mini-

mize environmental disturbance.

Application for an aquaculture permit must also be made to the Swedish

National Board of Fisheries. This agency reviews the potential for spread of

fish diseases and the introduction of exotic species. In a few cases it has

also been necessary to submit the application for review by a third agency, the

Water Rights Court. This body reviews the proposed facility if it will trans-

gress public or private rights.

In summary, this review of environmental regulations pertaining to mari-

culture indicates that the policies of Washington are generally comparable to

those of many of the other states and countries surveyed. There are usually

few, if any, formalized siting or operational criteria designed to minimize

environmental effects. Instead, each application is handled on a case-by-case

basis through a review process which includes an opportunity for the respon-

sible government agencies to evaluate potential environmental effects. Such is

generally the case in Washington as well as in California, Maine, Chile, Faroe

Islands, New Zealand, Scotland and Sweden. Of the countries surveyed, only

Norway and Canada (specifically British Columbia) have developed specific nu-

merical criteria regarding water depth, distance between operations and similar

siting requirements.

124

Importation of eggs or live animals typically requires a permit from the
responsible government agency. The permit will carry with it certain condi-
tions which, depending on the particular state or country, may include restric-
tions on the country of origin, visual and/or pathological inspectios, periods

of quarantine, or disinfection requirements. There appears to be a growing
tendency to completly ban importation of eggs or live animals for culture.
Chile and Scotland ban the importation of live fish for mariculture, but allow
the importation of eggs if certain steps have been taken to minimize the intro-
duction of disease. The Faroe Islands and New Zealand prohibit the importation
of both eggs and live animals.

125

6.0 RECOMMENDATIONS

This report has evaluated the likely environmental consequences of float-

ing mariculture development in Puget Sound based on the best information cur-

rently available. However, some unanswered questions remain which complicate

assessment of environmental effects and frustrate attempts by regulatory agen-

cies to develop defensible siting and operational criteria. Work in progress

will provide some of the needed information. For example, the National Marine

Fisheries Service is currently analyzing benthic samples collected over a 2.5

year period after removal of a net-pen. These samples will provide information

on the time required for recovery of the benthic community following cessation

of culture operations. The Washington Department of Fisheries and other agen-

cies are currently reviewing state policies pertaining to the importation of

plants and animals for aquaculture in order to insure that adequate safeguards

are in place to prevent disease transfer.

Three additional avenues of research appear to have the greatest priority:

Factors affecting the accumulation of organic-rich sediments beneath floating

mariculture facilities

This review has indicated that most floating mariculture operations have
resulted in substantial alteration of sediment chemistry and the benthic inver-

tebrate community in the immediate vicinity of the culture. Environmental man-

agers have recognized this fact, and have endeavored to site new operations

distant from benthic habitats that are considered critical or sensitive.

Rather than accepting accumulation of organic-rich sediments as an un-

avoidable consequence of mariculture, it would be preferable to find the means

to avoid this accumulation. It may be possible to establish conditions of cur-

rent velocity and water depth at which the rate of organic input to the bottom

would not exceed the assimilative capacity of the benthos. Determination of

the maximum allowable rate of organic input, or the conditions of water depth

and currents necessary to achieve it, will require the following steps be

taken:

127

I1) Field studies at numerous net-pen and mussel culture sites in Puget Sound
are needed to better quantify the chemical and biological gradients in sur-
rounding sediments. These field surveys should include measurements of the
current regimes at the sites. This information is critical, but is generally
lacking from studies to date.

2) Based on the field studies and a literature review some estimate of an
11acceptable" rate of organic input may be obtained. It is not anticipated,
however, that one estimate of loading rate will be applicable to all conditions
since it will depend on sediment porosity, bottom current velocity, BOD of the
natural sediments, oxygen content of the overlying water, biological community
structure and other factors. Site by site evaluations will be necessary until
our understanding of benthic processes increases to the point that predictive
capabilities are developed.

3) Sediment transport models are needed to describe the behavior of particu-
late material falling from the culture structure. Modeling sediment transport
Is a very active area of research, but we are still far from realizing the
types of predictive models needed for mariculture applications. We can begin
with development of empirical models that attempt to quantify the interdepen-
dence of water depth and current velocity in determining the rate of sediment
accumulation on the bottom. Data collected in the field studies can be used to
evaluate and refine these empirical models.

The environmental effects of antibiotic usage in salmon net-pen culture

The use of antibiotics in mariculture does not appear at present to be an
environmental threat given the characteristics of the antibiotic most commonly
used , the infrequency of its use and the limited development of mariculture in
Puget Sound, but the data available on potential effects are extremely limited.
As mariculture continues to develop in Puget Sound, the need for more informa-
tion on environmental release of antibiotics grows in importance. The follow-
ing information is needed:

128

I1) Additional data are needed on environmental fates of antibiotics currently
in use or of potential use in Puget Sound. Literature pertaining to metabolism

of the compounds in fish (e.g., retention time and metabolic breakdown prod-
ucts) and the persistence of the compounds in the marine environment (e.g.,

half-life, light sensitivity, susceptibility to chemical or biological degrada-
tion) should be reviewed. Laboratory studies may be needed to supplement
existing information.

2) Antibiotics are currently in use in mariculture operations in Puget Sound.
This on-going use provides an excellent opportunity to monitor environmental
fates. Monitoring of antibiotic concentrations in water, sediments and biota
should be performed by sampling before antibiotic treatment, during and immed-

iately after treatment, and continuing for at least several weeks thereafter.
Monitoring for the development of antibiotic resistance in microbial popula-
tions should also be included.

Determination of environmental carrying capacity

This review suggests that toxic effects attributable to accumulation of

metabolites a-ad depletion of dissolved oxygen are not likely to occur if mani-
culture operations are sited in well-flushed areas, and if the intensity of
mariculture (i.e., number of operations in an area) remains low. Unfortunate-
ly, the quantification of adequate flushing and determination of a maximum
allowable culture intensity are very difficult.
The determination of the environmental carrying capacity for mariculture
has typically been done by trial and error. A preferable approach would be the
development of mathematical models to predict changes in water chemistry param-
eters as a function of mariculture development. No such a predictive capabil-
ity currently exists, although some investigators have begun to explore the
concept of carrying capacity (e.g., Beveridge, 1984; Sakamoto, 1977). The
objective of these past studies has typically been the maintenance of the
health of the cultured animals rather than environmental protection (not neces-
sarily mutually exclusive objectives). Most work on the subject has been done
by the Japanese, where intensive culture in shallow, enclosed embayments has,
in some cases, exceeded the carrying capacity of the culture grounds.

129

Given the many competing priorities for use of Puget Sound, it is diffi-
cult to envision mariculture development to the extent practiced in Japan.
However, should the industry show continued expansion, and development occur in
areas of marginal suitability, a need may arise to determine how many opera-
tions in a given area are "too many"' and how little water exchange is "too
little". At present the best we can do is to learn from past mistakes and
evaluate each case on a site-by-site basis. However, the development of carry-
ing capacity models is recommended as a long-term goal.

130

Literature Cited

Ackefors, H. and K. Grip. 1985. Musslor och Ostron. Swedish Council for
Planning and Coordination of Research (FRN). 84 pp.

Ackefors, H., K. Murray and H. Rosenthal. 1984. A European view on the Asian
approach to modern aquaculture development. International Council for the
Exploration of the Sea C.M. 1984/F:3. 38 pp.

Ackefors, H. and A. S6dergren. 1985. Swedish experiences of the impact of
aquaculture on the environment. International Council for the Exploration of the
Seas C.M. 1985/E:40. 7 pp.

Alabaster, J.S. 1982. Survey of fish-farm effluents in some EIFAC countries.
In J.S. Alabaster (ed.), Report of the EIFAC Workshop on Fish-farm Effluents.
Silkeborg, Denmark, 26-28 May 1981. pp. 5-20. EIFAC Tech. Pap. 41.

Allendorf, F.W., D.M. Espeland, D.T. Scow and S. Phelps. 1980. Coexistence of
native and introduced rainbow trout in the Kootenai River drainage. Proc.
Montana Acad. Sci. 39:28-36.

Allendorf, F.W. and S.R. Phelps. 1980. Loss of genetic variation in a hatchery
stock of cutthroat trout. Trans. Am. Fish. Soc. 109:537-543.

Amend, D.F., W.T. Yasutake, J.L. Fryer, K.S. Pilcher and W.H Wingfield. 1973.
Infectious hematopoietic necrosis (IHN). In W.A. Dill (ed.), Symposium on the
Major Communicable Fish Diseases in Europe and Their Control. pp. 80-87. EIFAC
Tech. Pap. 17, Suppl. 2.

Anonymous. 1982. Far Jag Lov? Vattenbrukets juridik. [May I have permission?
Legal concerns in aquaculture]. Swedish Council for Planning and Coordination of
Research (FRN). 73 pp.

Anonymous. 1983a. Water quality surveyed in Clam Bay. Quarterly Report,
Jan.-Mar. 1983, pp. 11-12, Northwest and Alaska Fisheries Center, National Marine
Fisheries Service, National and Oceanic Atmospheric Administration, Seattle,
Washington.

Anonymous. 1983b. The Environmental Impact of Aquaculture. Report no. 83:5,
Swedish Council for Planning and Coordination of Research (FRN). 74 pp.

Anonymous. 1984. Review of Salmon Aquaculture. Prepared by the Senate Advisory
Council, Alaska State Legislature.

Aoki, T., Y. Jo and S. Egusa. 1980. Frequent occurrence of drug resistant
bacteria in Ayu (Plecoglossus altivelis) culture. Fish Path. 15(1):1-6.

APHA. 1981. Standard Methods for the Examination of Water and Wastewater.
Fifteenth edition, 1980. American Public Health Association, Washington, D.C.

Arakawa, K.Y. 1973. Yoshoku kaki seisan no suii narabi ni kaisan seibutsu
gunshu no seni kara mita [Aspects of eutrophication in Hiroshima Bay viewed from
transition of cultured oyster production and succession of marine biotic
communities]. Nihon Kaiyo Gakkai-Shi 11(2):43-48. (Translated by the
Multilingual Services Division, Dept. of the Secretary of State of Canada).

131

Arakawa, K.Y., Y. Kusuki and M. Kamigaki. 1971. Kaki yoshokujo ni okeru
seibutsu gentaiseki gensho no kenkyu (I) yoshoku tekisei mitsudo ni tsuite
[Studies on biodeposition in oyster beds (I) economic density for oyster
culture]. Venus 30(3):113-128.

Ariathurai, R., R.C. MacArthur and R.B. Krone. 1977. Mathematical model of
estuarial sediment transport. Technical Report D-77-12. Waterways Experiment
Station, Vicksburg, Mississippi.

Arizono, M. 1979. Disease control in mariculture, with special reference to
yellowtail culture. In G. Yamamoto (ed.), Proc. Seventh Japan-Soviet Joint
Symposium on Aquaculture, September 1978, Tokyo and Tsuruga, Japan., Tokai
University. pp. 79-88.

Aulstad, D., T. Gjerdrem and H. Skjerold. 1972. Genetic and environmental
sources of variation in length and weight of rainbow trout (Salmo gairdneri). J.
Fish. Res. Bd. Canada 29:237-241.

Austin, B. 1985. Antibiotic pollution from fish farms: effects on aquatic
microflora. Microbiol. Sci. 2(4):113-117.

Ayres, P.A. and G.I. Barrow. 1978. The distribution of Vibrio parahaemolyticus
in British coastal waters: report of a collaborative study 1975-6. J. Hyg.
Camb. 80:281-294.

Barbay, J.R., H.B. Bradford, Jr. and N.C. Roberts. 1984. The occurrence of
halophilic vibrios in Louisiana coastal waters. In R.R. Colwell (ed.), Vibrios
in the Environment. pp. 511-520. John Wiley and Sons, New York.

Baross, J.A. 1973. Some influences of temperature, bacteriophage, and other
ecological parameters on the distribution and taxonomy of marine vibrios. Ph.D.
thesis, University of Washington, Seattle. 451 pp.

Baross, J.A. and J. Liston. 1970. Occurrence of Vibrio parahaemolyticus and
related hemolytic vibrios in the marine environments of Washington state. Appl.
Microbiol. 20(2):179-186.

Baumann, P., L. Baumann, S.S. Bang and M.J. Woolkalis. 1980. Reevaluation of
the taxononmy of Vibrio, Beneckea and Photobacterium: abolition of the genus
Beneckea. Curr. Microbiol. 4:127-132.

Bayne, B.L. and C. Scullard. 1977. Rates of nitrogen excretion by species of
Mytilus (Bivalvia: Mollusca). J. mar. biol. Ass. U.K. 57:355-369.

Bayne, B.L. and J. Widdows. 1978. The physiological ecology of two populations
of Mytilus edulis L. Oecologia (Berlin) 37:137-162.

Bayne, B.L., J. Widdows and R.J. Thompson. 1976. Physiology II. In B.L. Bayne
(ed.), Marine Mussels: Their Ecology and Physiology. pp. 207-260. Cambridge
Univ. Press, Cambridge.

Bergheim, A. and A.R. Selmer-Olsen. 1978. River pollution from a large trout
farm in Norway. Aquaculture 14:267-270.

132

Bergheim, A., A. Sivertsen and A.R. Selmer-Olsen. 1982. Estimated pollution
loadings from Norwegian fish farms. I. Investigations 1978-1979. Aquaculture
28:347-361.

Beveridge, M. 1984. Cage and pen fish farming. FAO Fisheries Technical Paper
255. 131 pp.

Beveridge, M. and J.F. Muir. 1982. An evaluation of proposed cage fish culture
on Loch Lomond, an important reservoir in central Scotland. Canadian Wat. Res.
J. 7(2):181-196.

Bigham, G.N. 1986. Chapter 23: Oceanic disposal of wastes from manganese
nodule processing. In M.A. Champ and P.K. Park (eds.), Oceanic Processes in
Marine Pollution. Vol. 3: Marine Waste Management: Science and Policy. R.E.
Kreiger Publ. Co., Malabar, Florida. (in press).

Blake, P.A., R.E. Weaver and D.G. Hollis. 1980. Diseases of humans (other than
cholera) caused by vibrios. Ann. Rev. Microbiol. 34:341-367.

Bowden, G. 1979. Law, politics and biology: aquaculture in the primordial
ooze. In R. Mann (ed.), Exotic Species in Mariculture. pp. 306-330. MIT Press,
Cambridge, Massachusetts.

Braaten, B., J. Aure, A. Ervik and E. Boge. 1983. Pollution problems in
Norwegian fish farming. International Council for the Exploration of the Sea
C.M. 1983/F:26. 11 pp.

Brandsma, M.G. 1984. Simulation of drilling mud and cutting discharges in
Gainesville area block 707 and Apalachicola area block 1003. Prepared for
Continental Shelf Associates, Tequesta, Florida under contract to Sohio
Petroleum, Inc. 23 pp.

Brandsma, M.G. and D.J. Divoky. 1976. Development of models for prediction of
short-term fate of dredged material discharged in the estuarine environment.
Prepared by Tetra Tech, Inc., Pasadena, California for the U.S. Army Corps of
Engineers Waterways Experiment Station, Vicksburg, Mississippi.

Brandsma, M.G., T.C. Sauer and R.C. Ayers. 1983. Mud discharge model. Report
and user's guide model version 1.0. A model for predicting the fate of drilling
fluid discharges in the marine environment. Exxon Production Research Company,
Houston, Texas.

Bullock, G.L., H.M. Stuckey, D. Collis, R.L. Herman and G. Maestrone. 1974. In
vitro and in vivo efficacy of a potentiated sulfonamide in control of furunculos-
is in salmonids. J. Fish. Res. Bd. Canada 31:75-82.

Butman, C.A. in press. Sediment trap biases in turbulent flows: results from a
laboratory flume study. J. Mar. Res.

Butman, C.A., W.D. Grant and K.D. Stolzenbach. in press. Predictions of
sediment trap biases in turbulent flows: a theoretical analysis based on
observations from the literature. J. Mar. Res.

133

Butz, I. and B. Vens-Cappell. 1982. Organic load from the metabolite products
of rainbow trout fed with dry food. In J.S. Alabaster (ed.), Report of the EIFAC
Workshop on Fish-farm Effluents. Silkeborg, Denmark, 26-28 May 1981. pp. 73-82.
EIFAC Tech. Pap. 41.

Campbell, S.A., W.K. Peterson and J.R. Postel. 1977. Phytoplankton production
and standing stock in the Main Basin of Puget Sound. Prepared by the Universtiy
of Washington, Dept. of Oceanography for the Municicpality of Metropolitan
Seattle, Seattle, Washington.

Campos, M.J. and J. Marino. 1982. Comparison of phytoplankton species
composition and abundance in the Rias Arosa and Muros (NW Spain). Mar. Ecol.
3(1):1-12.

Campton, D.E. and J.M. Johnston. 1985. Electrophoretic evidence for a genetic
admixture of native and nonnative rainbow trout in the Yakima River, Washington.
Trans. Am. Fish. Soc. 114:782-793.

Carlton, J.T. 1981. Caveat indigenae: The unintentional introduction of exotic
species with target species in mariculture operations, and the potential
ecological impact on the adjacent environment. International Council for the
Exploration of the Sea C.M. 1981/F:17. 5 pp.

Carlucci, A.F. and D. Pramer. 1960. An evaluation of factors affecting the
survival of Escherichia coli in sea water. Appl. Microbiol. 8(4):251-254.

Chen, C.W. and G.T. Orlob. 1972. The accumulation and significance of sludge
near San Diego outfall. J. Wat. Pollut. Control Fed. 44:1362-1371.

Chevassus, B. 1983. Hybridization in fish. Aquaculture 33:245-262.

Chew, K.K. 1984. Recent advances in the cultivation of molluscs in the Pacific
United States and Canada. Aquaculture 39:69-81.

Chilcote, M.W., S.A. Leider and J.J. Loch. 1985. Hatchery summer-run steelhead
natural reproductive success in the Kalama River. Washington Dept. of Game,
Report No. 85-24, Olympia, Washington.

Christie, N.D. and A. Moldan. 1977. Effect of fish factory effluent on the
benthic macrofauna of Saldanha Bay. Mar. Poll. Bull. 8:41-45.

Clark, E.R., J.P. Harman and J.R.M. Forster. 1985. Production of metabolic and
waste products by intensively farmed rainbow trout, Salmo gairdneri Richardson.
J. Fish. Biol. 27:381-393.

Collias, E.E. and J.H. Lincoln. 1977. A Study of the Nutrients in the Main
Basin of Puget Sound. Prepared by the University of Washington, Department of
Oceanography for the Municipality of Metropolitan Seattle, Seattle, Washington.

Collias, E.E., N. McGary and C.A. Barnes. 1974. Atlas of Physical and Chemical
Properties of Puget Sound and its Approaches. Washington Sea Grant, Univ. of
Washington Press, Seattle. 235 pp.

134

Collins, I. 1983. A study on the environmental impact of particulate matter,
derived from a salmonid cage culture system on L. Fad, Isle of Bute, Scotland.
B.Sc. Thesis, University of Stirling. (not seen - cited in Gowen, 1985).

Colwell, R.R. and D.J. Grimes. 1984. Vibrio diseases of marine fish
populations. Helgolander Meeresunters. 37:265-287.

Colwell, R.R. and J. Liston. 1960. Microbiology of shellfish -- Bacteriological
study of the natural flora of Pacific oysters (Crassostrea gigas). Appl.
Microbiol. 8:104-109.

Colwell, R.R., P.A. West, D. Maneval, E.F. Remmers, E.L. Elliot, and N.E.
Carlson. 1984. Ecology of pathogenic vibrios in Chesapeake Bay. In R.R.
Colwell (ed.), Vibrios in the Environment. pp. 367-387. John Wiley and Sons,
New York.

Corliss, J.P. 1979. Accumulation and depletion of oxytetracycline in juvenile
white shrimp (Penaeus setiferus). Aquaculture 16:1-6.

Cornelson, D.A., I. Sechter, E. Balteanu and V. Avram. 1958. Specific bacteri-
ological pollution of streams by waste waters from penicillin plants. Arch.
roum. Path. exp. 17:271-278.

Courtenay, W.R., Jr. and C.R. Robins. 1973. Exotic aquatic organisms in Florida
with emphasis on fishes: a review and recommendations. Trans. Am. Fish. Soc.
102(1):1-12.

Cronin, L.E. 1967. The role of man in estuarine processes. In G. Lauff (ed.),
Estuaries. pp. 667-689. Amer. Assoc. Adv. Sci., Washington, D.C.

Dahlback, B. and L.A.H. Gunnarson. 1981. Sedimentation and sulfate reduction
under a mussel culture. Mar. Biol. 63:269-275.

De Groot, S.J. 1985. Introductions of non-indigenous fish species for release
and culture in the Netherlands. Aquaculture 46:237-257.

DMR. 1985. Maine Marine Resouces Laws. Prepared by the Maine Department of
Marine Resources, Augusta, Maine.

Doyle, J., M. Parker, T. Dunne, D. Baird and J. McArdle. 1984. The impact of
blooms on mariculture in Ireland. International Council for the Exploration of
the Sea, Special Meeting on the Causes Dynamics and Effects of Exceptional Marine
Blooms and Related Events, Copenhagen, 4-5 October 1984.

Eckman, J.E. and A.R.M. Nowell. 1984. Boundary skin friction and sediment
transport about an animal-tube mimic. Sedimentology 31:851-862.

Edwards, D.J. 1978. Salmon and Trout Farming in Norway. Fishing News Books
Limited, Farnham.

Egidius, E. 1984. Diseases of salmonids in aquaculture. Helgolander
Meeresunters. 37:547-569.

135

Eley, R.L., J.H. Carroll and D. DeWoody. 1972. Effects of caged catfish culture
on water quality and community metabolism of a lake. Proc. Okla. Acad. Sci.
52:10-15.

Ellis, M.M., B.A. Westfall and M.D. Ellis. 1948. Determination of water
quality. U.S. Fish and Wildlife Service, Res. Rept. 9. 122 pp.

Enell, M. and J. L6f. 1983a. Milj6konsekvenser av akvakultur. Sedimentation
fran fiskkasseodlingar. Report for the National Swedish Environment Protection
Board. 46 pp. (not seen - cited in Anonymous, 1983b)

Enell, M. and J. L6f. 1983b. Milj6effekter av vattenbruk - sedimentation och
narsaltbelastning fran fiskkasseodlingar [Environmental impact of aquaculture -
sedimentation and nutrient loading from fish cage culture farming]. Vatten
39:364-375.

Enell, M., J. L6f and T.-L. Bjorklund. 1984. Fiskkasseodling med Rening,
Teknisk Beskrivning och Reningseffekt. Institute of Limnology, Lund. (not seen
- cited in Ackefors and Sodergren, 1985)

Environment Canada. 1979. Water quality sourcebook, a guide to water quality
parameters. Inland Waters Directorate, Water Quality Branch, Ottawa, Canada.

Ervik, A., P. Johannessen and J. Aure. 1985. Environmental effects of marine
Norwegian fish farms. International Council for the Exploration of the Sea C.M.
1985/F:37. 13 pp.

Fenchel, T.M. and R.J. Reidl. 1970. The sulfide system: a new biotic community
underneath the oxidized layer of marine sand bottoms. Mar. Biol. 7:255-268.

Forfar, J.O., A.J. Keay, A.F. Maccabe, J.C. Gould and A.D. Bain. 1966. Liberal
use of antibiotics and its effect in neonatal staphylococcal infection, with
particular reference to erythromycin. Lancet ii:295-300.

Fribourgh, J.H., F.P. Meyer and J.A. Robinson. 1969. Oxytetracycline leaching
from medicated fish feeds. U.S. Bur. Sport Fish. Wildlife, Tech. Pap. 40. 7 pp.

Friebertshauser, M., K. Krogslund, V. Wong, J. McCulloch and P. Stoops. 1971.
Puget Sound and Approaches Seasonal Variations of Oceanographic Parameters in
Near-surface Waters. Washington Sea Grant, Univ. of Washington, Seattle.
235 pp.

Gooding, R.M. and J.J. Magnuson. 1967. Ecological significance of a drifting
object to pelagic fishes. Pac. Sci. 21:486-497.

Gowen, R.J., N.B. Bradbury and J.R. Brown. 1985. The ecological impact of
salmon farming in Scottish coastal waters: a preliminary appraisal.
International Council for the Exploration of the Sea C.M. 1985/F:35. 13 pp.

Greenberg, E.P., H.B. Kaplan, M. DuBoise and B. Palhof. 1984. In R.R. Colwell
(ed.), Vibrios in the Environment. pp. 479-493. John Wiley and Sons, New York.

Grimes, D.J., S.H. Gruber and E.B. May. 1984. Experimental infection of lemon
sharks, Negaprion brevirostris (Poey), with Vibrio species. J. Fish. Dis.
8(2):173-180.

136

Hada, H.S. and R.K. Sizemore. 1981. Incidence of plasmids in marine Vibrio spp.
isolated from an oil field in the Northwestern Gulf of Mexico. Appl. Environ.
Microbiol. 41:199-202.

Hagino, S. 1977. [Physical properties of the pollutants]. In Japanese Society
of Scientific Fisheries (ed.), Senkai Yoshoku to Jika Osen [Shallow-Sea
Aquaculture and Self-Pollution], Suisan-gaku Shirizu 21 [Fisheries Series 21].
pp. 31-41. Koseisha Koseikaku, Tokyo. (In Japanese - translated by Madelon and
Louis Mottet).

Hagstrom, A. and A-M. Larsson. 1982. Pelagiska bakterier och musselodling
[Pelagic bacteria and mussel culture]. Meddelande Havsfiskelaboratoriet Lysekil,
No. 286. 21 pp. (In Swedish - partial translation by Daniel Olsen)

Harrell, L.W. and T.M. Scott. 1985. Kudoa thyrsitis (Gilchrist) (Myxosporea:
Multivalvulida) in Atlantic salmon Salmo salar L. J. Fish. Dis. 8(3):329-332.

Hata, Y. and H. Katayama. 1977. [Chemical processes of the pollutants]. In
Japanese Society of Scientific Fisheries (ed.), Senkai Yoshoku to Jika Osen
[Shallow-Sea Aquaculture and Self-Pollution], Suisan-gaku Shirizu 21 [Fisheries
Series 21]. pp. 52-66. Koseisha Koseikaku, Tokyo. (In Japanese - translated by
Madelon and Louis Mottet)

Hays, T. 1980. Impact of net pen culture on water quality and fish populations
on Bull Shoals Reservoir. Completion Rep. Arkansas Game Fish. Comm. 10 pp.
(not seen - cited in Beveridge, 1984)

Herman, R.L. 1969. Oxytetracycline in fish culture -- a review. U.S. Bur.
Sport Fish. Wildlife, Tech Pap. 31. 9 pp.

Herman, R.L., D. Collins and G.L. Bullock. 1969. Oxytetracycline residues in
different tissues of trout. U.S. Bur. Sport Fish. Wildlife, Tech. Pap. 37.
6 pp.

Hershberger, W.K., J.A. Perdue and J.H. Beattie. 1984. Genetic selection and
systematic breeding in Pacific oyster culture. Aquaculture 39:237-245.

Hinshaw, R.N. 1973. Pollution as a result of fish cultural activities.
EPA-R3-73-009. U.S. Environmental Protection Agency, Washington, D.C.

Hisaoka, M., K. Nogami, O. Takeuchi, M. Suzuki and H. Sugimoto. 1966. Studies
on sea water exchange in fish farm. II. Exchange of sea water in floating net.
Bull. Naikai Reg. Fish. Res. Lab., Contr. No. 115, pp. 21-43.

Hoffman, G.L. 1970. Intercontinental and transcontinental dissemination and
transfaunation of fish parasites with emphasis on whirling disease (Myxosoma
cerebralis). In S.F. Snieszko (ed.), A Symposium on Diseases of Fishes and
Shellfishes. pp. 69-81. American Fish. Soc., Spec. Publ. No. 5.

Hurlburt, E.F. 1984. Aquaculture in Puget Sound - Its potential and its
possible environmental impact. State of Washington, Department of Fisheries,
Olympia, Washington.

137

ICES. 1982. Proposed guidelines for implementing the ICES Code of Practice
concerning introductions and transfers of marine species. International Council
for the Exploration of the Sea. CM 1982/F:33.

Iigura, T. 1974. [Sinking speed of feces and uneaten food in hamachi culture].
Suisan zoshoku 22:34-39.

Imai, T. 1971. Aquaculture in Shallow Seas: Progress in Shallow Sea Culture.
Koseisha Koseiku, Tokyo. Translated by the National Marine Fisheries Service and
republished in 1977 by Amerind Publishing Co. Pvt. Ltd., New Dehli.

Incze, L.S. and R.A. Lutz. 1980. Mussel culture: an east coast perspective.
In R.A. Lutz (ed.), Mussel Culture and Harvest: A North American Perspective.
pp. 99-140. Developments in Aquaculture and Fisheries Science, 7. Elsevier
Scientific Publ. Co., Amsterdam.

Inoue, H. 1972. On water exchange in a net cage stocked with the fish, hamachi.
Bull. Japanese Soc. Sc. Fish. 38(2):167-176. (In Japanese - translated by Madelon
and Louis Mottet)

Inoue, H., Y. Tanaka and K. Fukuda. 1970. On water exchange in a shallow marine
fish farm - II. Hamachi fish farm at Tanoura. Bull. Jap. Soc. Sci. Fish.
36(8):776-782.

Inoue, H., Y. Tanaka and M. Saito. 1966. On the water exchange in the shallow
marine fish farm - I. Hitsuishi fish farm of hamachi. Bull. Jap. Soc. Sci.
Fish. 32(5):384-392.

Ito, S. and T. Imai. 1955. Ecology of oyster bed. I. On the decline of
productivity due to repeated cultures. Tohoku J. Agric. Res. V(4):251-268.

Iwasaki, H. 1976. [Red tides - Questions concerning their outbreaks]. Ocean
Publications Co., Ltd., Tokyo. (not seen - cited in Takagi, et al., 1980).

James, W. unpub. Internal memorandum dated November 8, 1985 from Bill James to
Eric Hurlburt, Washington Department of Fisheries.

Jarrams, P., A. Starkie, K. Easton and R.G. Templeton. 1980. Salmonid rearing
in floating net cages in freshwater reservoirs owned by the Severn Trent Water
Authority. Fish. Mgmt. 3:101-117.

Johnstone, A.K. 1984. Pathogenesis and life cycle of the myxozoan Parvicapsula
sp. infecting marine cultured coho salmon. Ph.D. thesis. University of
Washington, Seattle.

Jorgensen, S.E. 1979. State-of-the-art in Ecological Modelling. Pergamon
Press, Oxford. 891 pp.

Joseph, S.W., R.R. Colwell and J.B. Kaper. 1982. Vibrio parahaemolyticus and
related halophilic vibrios. CRC Crit. Rev. Microbiol. 10(1):77-124.

Kadowaki, S. and H. Hirata. 1984. Oxygen distribution in the coastal fish farm.
I. Effects of feeding on the distribution patterns. Suisan zoshoku
32(3):142-147. (In Japanese - translated by Madelon and Louis Mottet)

138

Kadowaki, S., T. Kasedo and T. Nakazono. 1978a. Continous records of DO
contents by cruising in the coastal culture farm. I. Relation between DO
content and fish density in cages. Mem. Fac. Fish., Kagoshima Univ.
27(1):273-280.

Kadowaki, S., T. Kasedo, T. Nakazono and H. Hirata. 1978b. Continuous records
of DO contents by cruising in the coastal culture farms. II. Diffusion of
suspended particles by feeding. Mem. Fac. Fish., Kagoshima Univ. 27(1):281-288.

Kadowaki, S., T. Kasedo, T. Nakazono, Y. Yamashita and H. Hirata. 1980. The
relation between sediment flux and fish feeding in coastal culture farms. Mem.
Fac. Fish., Kagoshima Univ. 29:217-224.

Kaneko, T. and R.R. Colwell. 1973. Ecology of Vibrio parahaemolyticus in
Chesapeake Bay. J. Bacteriol. 113:24-32.

Kaspar, H.F., P.A. Gillespie, I.C. Boyer and A.L. MacKenzie. 1985. Effects of
mussel aquaculture on the nitrogen cycle and benthic communities in Kenepuru
Sound, Marlborough Sounds, New Zealand. Mar. Biol. 85:127-136.

Kautsky, N. and I. Wallentinus. 1980. Nutrient release from a Baltic
Mytilus-red algal community and its role in benthic and pelagic productivity.
Ophelia, Suppl. 1:17-30.

Kenaga, E.E. and C.A.I. Goring. 1980. Relationship between water solubility,
soil sorption, octanol-water partitioning, and concentration of chemicals in
biota. In J.G. Eaton, P.R. Parrish and A.C. Hendricks (ed.), Aquatic Toxicology.
pp. 78-115. ASTM STP 707, American Society for Testing and Materials.

Kennedy, W.A. 1978. A handbook on rearing pan-size Pacific salmon using
floating seapens. Fisheries and Marine Service Industry Report No. 107, Pacific
Biological Station, Nanaimo, British Columbia. 111 pp.

Kilambi, R.V., et al. 1976. Effects of cage culture fish production upon the
biotic and abiotic environment of Crystal Lake, Arkansas. Prepared by the
Department of Zoology, University of Arkansas for the Arkansas Game and Fish
Commission and U.S. Dept. of Commerce. 127 pp. (not seen - cited in Beveridge,
1984)

Kils, U. 1979. Oxygen-regime and artificial aeration of net-cages in
mariculture. Meeresforsch. 27(4):236-243.

Kincaid, H.L. 1976a. Inbreeding in rainbow trout (Salmo gairdneri). J. Fish.
Res. Bd. Canada 33:2420-2426.

Kincaid, H.L. 1976b. Inbreeding depression in rainbow trout. Trans. Am. Fish.
Soc. 105:273-280.

Kitamori, R. 1977. [Changes in the species composition (principally benthic
organisms)]. In Japanese Soc. Scientific Fisheries (ed.), Senkai Yoshoku to
Jika Osen [Shallow-Sea Aquaculture and Self-Pollution], Suisan-gaku Shirizu 21
[Fisheries Series 21]. pp. 67-76. Koseisha Koseikaku, Tokyo. (In Japanese -
translated by Madelon and Louis Mottet)

139

;� B

Koh, R.C.Y. 1971. Ocean sludge disposal by barges. Water Resources Res.
7(6):1647-1651.

Koh, R.C.Y. and Y.C. Chang. 1973. Mathematical model for barged ocean disposal
of wastes. EPA 660-2-73-029. U.S. Environmental Protection Agency, Washington,
D.C.

Korzeniewski K., J. Trojanowski and B. Mrozek. 1982a. Effect of intensive trout
culture on contents of nutrients in water. Pol. Arch. Hydrobiol.
29(3-4):625-632.

Korzeniewski, K., Z. Banat and A. Moczulska. 1982b. Changes in water of the
Uniesc and Skotawa Rivers, caused by intensive trout culture. Pol. Arch.
Hydrobiol. 29(3-4):683-691.

Kramer, Chin and Mayo. undated. An environmental impact assessment for the
NOAA/UW Regional Fisheries Laboratory. Prepared for the National Oceanic and
Atmospheric Administration and University of Washington, Seattle, Washington.

Kuenzler, E.J. 1961. Phosphorus budget of a mussel population. Limnol.
Oceanogr. 6:400-415.

Kusuki, Y. 1977a. Fundamental studies on the deterioration of oyster growing
grounds - I. Production of faecal materials by the Japanese oyster. Bull.
Japanese Soc. Sci. Fish. 43(2):163-166.

Kusuki, Y. 1977b. Fundamental studies on the deterioration of oyster growing
grounds - II. Organic content of faecal materials. Bull. Japanese Soc. Sci.
Fish. 43(2):167-171.

Kyte, M.A. unpub. Report on a geoduck survey performed for the Scan Am fish
farm off Crane Point, Indian Island, Washington.

Lappalainen, A. and P. Kangas. 1975. Littoral benthos of the Northern Baltic
Sea II. Interrelationships of wet, dry and ash-free dry weights of macrofauna in
the TvArminne area. Int Revue ges. Hydrobiol. 60(3):297-312.

Larsen, J.L., N.J. Jensen and N.D. Christensen. 1978. Water pollution and the
ulcer syndrome in the cod Gadus morhua. Vet. Sci. Commun. 2:207-216.

Larsson, A-M. 1985. Blue mussel sea farming - effects on water quality. Vatten
41:218-224.

Lavelle, J.W., E. Ozturgut, S.A. Swift and B.H. Erickson. 1981. Dispersal and
resedimentation of the benthic plume from deep-sea mining operations: a model
with calibration. Mar. Min. 3(1-2):59-93.

Leavens, K. 1983. An overview of netpen rearing as an enhancement technique.
Prepared for the Department of Fisheries and Oceans, Vancouver, British Columbia.

Liao, P.B. and R.D. Mayo. 1972. Salmonid hatchery water reuse systems.
Aquaculture 1:317-335.

Liao, P.B. and R.D. Mayo. 1974. Intensified fish culture combining water
reconditioning with pollution abatement. Aquaculture 3:61-85.

140

Lieffrig, D.V.M. 1985. The effects of hydrogen sulfide (H2S) on aquaculture
production. MSc thesis, University of Stirling. (not seen - cited in Phillips,
et al., 1985)

Lindbergh, J.M. 1972. Pacific salmon aquaculture program - incubation and
cultivation phases. Univ. of Washington Sea Grant Final Report, Prepared by
Ocean Systems, Inc., Domsea Farms, Inc., Reston, Virginia. 52 pp.

Lindbergh, J.M. unpub. Potential interaction between net pen farmed Atlantic
salmon and native salmonid species in the Pacific Northwest. Prepared for Sea
Farm of Norway, Inc., Greenwich, Connecticut.

Loo, L-O. and R. Rosenberg. 1983. Mytilus edulis culture: growth and
production in western Sweden. Aquaculture 35:137-150.

Lopez-Jamar, E. 1985. Distribucion espacial del poliqueto Spiochaetopterus
costarum en las Rias Bajas de Galacia y su posible utilizaci6n como indicador de
contaminaci6n organica en el sedimento. Bol. Inst. Esp. Oceanogr. 2(1):68-76.

Lopez-Jamar, E., J. Iglesias and J.J. Otero. 1984. Contribution of infauna and
mussel-raft epifauna to demersal fish diets. Mar. Ecol. Prog. Ser. 15:13-18.

Loyacano, H.A., Jr. and D.C. Smith. 1976. Attraction of native fish to catfish
culture cages in reservoirs. Proc. Ann. Conf. Southeast. Assoc. Game Fish. Comm.
29:63-73.

Mackay, D.W., W. Halcrow and I. Thornton. 1972. Sludge dumping in the Firth of
Clyde. Mar. Poll. Bull. 3:7-10.

Mahnken, C.V.M. 1975. Status of commercial net-pen farming of Pacific salmon in
Puget Sound. In Proceedings of the Sixth Annual Meeting World Mariculture
Society, Seattle, Washington, January 27-31, 1975. pp. 285-298. Louisiana State
Univ., Baton Rouge, Louisiana.

Mahnken, C.V.W., D.M. Damkaer and J. Vidal. unpub. Impact of intensive marine
net-pen culture on water quality and benthic community structure. Unpublished
research proposal, August 1982.

Mantle, G.J. 1982. Biological and chemical changes associated with the
discharge of fish-farm effluent. In J.S. Alabaster (ed.), Report of the EIFAC
Workshop on Fish-farm Effluents, Silkeborg, Denmark, 26-28 May 1981. pp.
103-112. Food and Agricultural Organization of the United Nations, EIFAC Tech.
Pap. No. 41.

Mason, J. 1976. Cultivation. In B.L. Bayne (ed.), Marine Mussels: Their
Ecology and Physiology. pp. 385-410. Cambridge Univ. Press, Cambridge.

Mattsson, J. and 0. Linden. 1983. Benthic macrofauna succession under mussels,
Mytilus edulis L. (Bivalvia), cultured on hanging long lines. Sarsia 68:97-102.

Merican, Z.O. and M.J. Phillips. 1985. Solid waste production from rainbow
trout, Salmo gairdneri Richardson, cage culture. Aqua. Fish. Mgmt. 1:55-69.

141

Meyers, T.R. 1984. Marine bivalve mollusks as reservoirs of viral pathogens:
significance to marine and anadromous finfish aquaculture. Mar. Fish. Rev.
46(3):14-17.

Midtlyng, P.V. 1985. Bruk av medikamenter og desinfeksjonsmidler i norske
fiskoppdrettsanlegg. LUse of drugs and disinfectants in Norwegian fish farms].
Vann 3:177-180.

Mills, S. 1982. Britain's native trout is floundering. New Scientist
25:498-501.

Milne, P.H. 1970. Fish farming: a guide to the design and construction of net
enclosures. Dept. Agric. and Fish. for Scotland, Marine Research, No. 1.

Milner-Rensel Assoc. 1986. Aquatic conditions at the Seafarm of Norway net-pen
site in Port Angeles harbor in April, 1986. Prepared for Sea Farm of Norway,
Port Angeles, Washington.

Minchew, C.D. and J.D. Yarbrough. 1977. The occurrence of fin rot in mullet
(Mugil cephalus) associated with crude oil contamination of an estuarine pond
ecosystem. J. Fish. Biol. 10:319-323.

Morrison, J.A. unpub. Memorandum dated July 16, 1985 from J.A. Morrison to
other staff of the Canada Department of Fisheries and Oceans.

Muir, J.F. 1985. The role of aquaculture in the industrial regions of the
temperate zone. GeoJournal 10(3):277-298.

Munro, A.L.S. and I.F. Waddell. 1984. Furunculosis: experience of its control
in the sea water cage culture of Atlantic salmon in Scotland. International
Council for the Exploration of the Sea C.M. 1984/F:32. 9 pp.

Murakami, A. 1973. [Eutrophication of the hydrosphere and the increase of
fisheries culture]. In [Water Pollution and Eutrophication]. pp. 37-48.
Japanese Fisheries Society, Koseisha Kosei Kakuhan. (not seen - cited in
Arakawa, 1973)

Musselman, M.M. 1956. Terramycin (Oxytetracycline). Medical Encyclopedia,
Inc., New York. 144 pp.

Newkirk, G.F. 1979. Genetic aspects of the introduction and culture of
nonindigenous species for aquaculture. In R. Mann (ed.), Exotic Species in
Mariculture. pp. 192-211. MIT Press, Cambridge, Massachusetts.

Newkirk, G.F. 1980. Review of the genetics and the potential for selective
breeding of commercially important bivalves. Aquaculture 19:209-228.

Nishimura, A. 1982. Effects of organic matters produced in fish farms on the
growth of redtide algae Gymnodinium type-'65 and Chattonella antiqua. Bull.
Plankton Soc. Japan 29(1):1-7.

NOAA. 1985. The outlook for salmon and shrimp aquaculture products in world
markets. Prepared for the NOAA Assistant Administrator for Fisheries by the
Aquaculture Project Group, National Oceanic and Atmospheric Administration.
109 pp.

142

Nose, T. 1985. Recent advances in aquaculture in Japan. GeoJournal
10(3):261-276.

NOU. 1985. Akvakultur i Norge. Norges Offentlige Utredninger 1985: 22.
153 pp. (In Norwegian - partial translation by Leslie Grove).

Nowell, A.R.M. and P.A. Jumars. 1984. Flow environments of aquatic benthos.
Ann. Rev. Ecol. Syst. 15:303-328.

Nyegaard, L. 1973. Coho salmon farming in Puget Sound. Washington State
University, Cooperative Extension Service, College of Agriculture. Extension
Bull. 647. 14 pp.

O'Connor, D.J. 1981. Modeling of eutrophication in estuaries. In B.J. Neilson
and L.E. Cronin (eds.), Estuaries and Nutrients. pp. 183-223. Humana Press,
Clifton, New Jersey.

Odum, W.E. 1974. Potential effects of aquaculture on inshore coastal waters.
Envir. Cons. 1(3):225-230.

Oviatt, C.A., A.L. Gall and S.W. Nixon. 1972. Environmental effects of Atlantic
menhaden on surrounding waters. Ches. Sci. 13(4):321-322.

Pamatmat, M.M., R.S. Jones, H. Sanborn and A. Bhagwat. 1973. Oxidation of
organic matter in sediments. EPA-660/3-73-005. U.S. Environmental Protection
Agency, Washington, D.C.

Parsons, T.R., M. Takahashi and B. Hargrave. 1977. Biological Oceanographic
Processes. Second edition. Pergamon Press, New York. 332 pp.

Pearson, T.H. 1976. A comparative study of the effects on the marine environ-
ment of wastes from cellulose industries in Scotland and Sweden. Ambio 5:77-79.

Pearson, T.H. and R. Rosenberg. 1978. Macrobenthic succession in relation to
organic enrichment and pollution of the marine environment. Oceanogr. Mar. Biol.
Ann. Rev. 16:229-311.

Pearson, T.H. and S.O. Stanley. 1979. Comparative measurement of the redox
potential of marine sediments as a rapid means of assessing the effect of organic
pollution. Mar. Biol. 53:371-379.

Pease, B.C. 1977. The effect of organic enrichment from a salmon mariculture
facility on the water quality and benthic community of Henderson Inlet,
Washington. Ph.D. Dissertation, Univ. of Washington, Seattle, Washington.

Pease, B. unpub. Memorandum dated May 7, 1984 from B. Pease to R. Westley,
Washington Department of Fisheries.

Pease, B. and L. Goodwin. unpub. Memorandum dated February 27, 1986 to Dick
Burge, Washington Department of Fisheries.

Pedersen, A. 1982. Milj0pAvirkning fra fiskeoppdrett [Environmental effects
from fish farming]. Prosjektrapport FP 80802 (F.430) NIVA:153 pp. (In Norwegian
- partial translation by A. Bohn)

143

Penczak, T., W. Galicka, M. Molinski, E. Kusto, and M. Zalewski. 1982. The
enrichment of a mesotrophic lake by carbon, phosphorus and nitrogen from the cage
aquaculture of rainbow trout, Salmo gairdneri. J. Appl. Ecol. 19:371-393.

Peou, S., M.F. Blech and P. Hartemann. 1981. Study of the frequency of
antibiotic resistant microbial indicators in hospital and urban sewage. Envir.
Tech. Letters 2(8):347-356.

Phillips, M.J., M.C.M. Beveridge, and J.F. Muir. 1985. Waste output and
environmental effects of rainbow trout cage culture. International Council for
the Exploration of the Sea C.M. 1985/F:21. 17 pp.

Purdom, C.E. 1983. Genetic engineering by the manipulation of chromosomes.
Aquaculture 33:287-300.

Quayle, D.B. 1964. Distribution of introduced marine Mollusca in British
Columbia waters. J. Fish. Res. Bd. Canada 21(5):1155-1181.

Querellou, J., A. Faure and C. Faure. 1982. Pollution loads from rainbow trout
farms in Brittany, France. In J.S. Alabaster (ed.), Report of the EIFAC Workshop
on Fish-farm Effluents. Silkeborg, Denmark, 26-28 May 1981. pp. 87-97. EIFAC
Tech. Pap. 41. 166 pp.

Refstie, T. and T. Gjedrem. 1975. Hybrids between Salmonidae species.
Hatchability and growth rate in the freshwater period. Aquaculture 6:333-342.

Reisenbichler, R.R. and J.D. McIntyre. 1977. Genetic differences in growth and
survival of juvenile hatchery and wild steelhead trout, Salmo gairdneri. J.
Fish. Res. Bd. Canada 34:123-128.

Rensel, J. unpub. Marine farms in Washington State - Existing farms: March
1986.

Roberts, R.J. 1978. Fish Pathology. Balliere and Tindall, London.

Robohm, R.A., C. Brown and R.A. Murchelano. 1979. Comparison of antibodies in
marine fish from clean and polluted waters of the New York Bight: relative
levels against 36 bacteria. Appl. environ. Microbiol. 38:248-257.

Romero, P., E. Gonzalez-Gurriaran and E. Penas. 1982. Influence of mussel rafts
on spatial and seasonal abundance of crabs in the Ria de Arousa, North-west
Spain. Mar. Biol. 72:201-210.

Rosenberg, R. 1976. Benthic faunal dynamics during succession following
pollution abatement in a Swedish estuary. Oikos 27:414-427.

Rosenberg, R. and L-O. Loo. 1983. Energy-flow in a Mytilus edulis culture in
western Sweden. Aquaculture 35:151-161.

Rosenfield, A. and F.G. Kern. 1979. Molluscan imports and the potential for
introduction of disease organisms. In R. Mann (ed.), Exotic Species in
Mariculture. pp. 165-189. MIT Press, Cambridge, Massachusetts

144

Rosenthal, H. 1978. Bibliography on transplantation of aquatic organisms and
its consequences on aquaculture and ecosystems. Spec. Publ. European Mariculture
Soc. 3:1-146.

Rosenthal, H. 1980. Implications of transplantations to aquaculture and
ecosystems. Mar. Fish. Rev. 42(5):1-14.

Rosenthal, H. 1985. Constraints and perspectives in aquaculture development.
GeoJournal 10(3):305-324.

Runchal, A.K. 1978. Drilling fluid dispersion and biological effects study for
the lower Cook Inlet C.O.S.T. well. Dames and Moore, Anchorage, Alaska. 309 pp.

Russel, C.S. (ed.). 1975. Ecological Modeling in a Resources Management
Framework. John Hopkins Univ. Press, Baltimore, Maryland.

Ryman, N. 1970. A genetic analysis of recapture frequencies of released young
of salmon (Salmo salar L.). Hereditas 65:159-160.

Ryman, N. and G. Stahl. 1980. Genetic changes in hatchery stocks of brown trout
(Salmo trutta). Can J. Fish. Aquat. Sci. 37:82-87.

Ryther, J.H. 1969. The potential of the estuary for shellfish production.
Proc. Nat. Shellfisheries Assoc. 59:18-22.

Sakamoto, I. 1977. [Discussion. Estimating the carrying capacity for culturing
fish]. In Japanese Society of Scientific Fisheries (ed.), Senkai Yoshoku to Jika
Osen [Shallow-Sea Aquaculture and Self-Pollution]. Suisan-gaku Shirizu 21
[Fisheries Series 21]. pp. 119-129. Koseisha Koseikaku, Tokyo. (In Japanese -
translated by Madelon and Louis Mottet)

Sheng, Y.P. 1983. Mathematical modeling of three-dimensional coastal currents
and sediment dispersion: model development and application. U.S. Army Corps of
Engineers, Washington, D.C. Tech. Rep. CERC-83-2.

Sindermann, C.J. 1984. Disease in marine aquaculture. Helgolander
Meeresunters. 37:505-532.

Smith, E.J. 1985. Marine resources - information leaflet: regulations
governing marine aquaculture. Prepared by the California Department of Fish and
Game, Sacramento, California.

Solbe, J.F. de L.G. 1982. Fish-farm effluents; a United Kingdom Survey. In
J.S. Alabaster (ed.), Report of the EIFAC Workshop on Fish-farm Effluents.
Silkeborg, Denmark, 26-28 May 1981. pp. 29-55. EIFAC Tech. Pap. 41.

Solomons, I.A. 1978. Antibiotics in animal feeds - human and animal safety
issues. J. Anim. Sci. 46(5):1360-1368.

Soule, D.F. and M. Oguri (eds.). 1976. Bioenhancement studies of the receiving
waters in outer Los Angeles harbor. Marine Studies of San Pedro Bay, California,
Part 12. Allan Hancock Foundation, Los Angeles.

145

Soule, D.F. and M. Oguri (eds.). 1979. Ecological changes in outer Los
Angeles-Long Beach harbors following initiation of secondary waste treatment and
cessation of fish cannery waste effluent. Marine Studies of San Pedro Bay,
California, Part 16. Allan Hancock Foundation, Los Angeles.

Speece, R.E. 1973. Trout metabolism characteristics and the rational design of
nitrification facilities for water reuse in hatcheries. Trans. Amer. Fish. Soc.
102(2):323-334.

Spira, W.M. 1984. Tactics for detecting pathogenic vibrios in the environment.
In R.R. Colwell (ed.), Vibrios in the Environment. pp. 251-268. John Wiley and
Sons, New York.

Stahl, G. 1983. Differences in the amount and distribution of genetic variation
between natural populations and hatchery stocks of Atlantic salmon. Aquaculture
33:23-32.

Stanley, J.G. 1979. Monosex and sterile fish for transplantation and aqua-
culture. International Council for the Exploration of the Seas C.M. 1979/F:32.

Steidinger, K.A. and K. Haddad. 1981. Biologic and hydrographic aspects of red
tides. Bioscience 31(11):814-819.

Stewart, K.I. 1984. A study on the environmental impact of fish cage culture on
an enclosed sea loch. M.Sc. thesis, University of Stirling. (not seen - cited
in Gowen, et al., 1985)

Stoddard, A., R. Wells and K. Devonald. 1985. Development and application of a
deepwater ocean waste disposal model for dredged material: Yabucoa Harbor,
Puerto Rico. Mar. Tech. Soc. J. 19(3):26-39.

STOWW. 1974. Salmon pen culture at Squaxin Island Indian Reservation. Prepared
by the Small Tribes Organization of Western Washington, Federal Way, Washington
for the Economic Development Administration, U.S. Office of Economic Opportunity
and the Bureau of Indian Affairs. Washington, D.C. 68 pp.

Strickland, R.M. 1983. The Fertile Fjord. Washington Sea Grant Publication,
Univ. of Washington Press, Seattle. 145 pp.

Sutterlin, A.M. and S.P. Merrill. 1978. Norwegian salmonid farming. Department
of Fisheries and Environment, St. Andrews, New Brunswick. Fisheries and Marine
Service, Tech. Rep. 779.

Tacket, C.O., F. Bremer and P.A. Blake. 1984. Clinical features and an
epidemiological study of Vibrio vulnificus infections. J. Infec. Dis.
149(4):558-561.

Takagi, M., H. Kamada and A. Tazaki. 1980. A consideration on the occurrence of
Protogonyaulax sp. at the scallop ground in Funka Bay. Bull. Fac. Fish. Hokkaido
Univ. 31(2):215-221. (In Japanese - translated by Madelon and Louis Mottet)

146

Tenore, K.R., L.F. Boyer, R.M. Cal, J. Corral, C. Garcia-Fernandez, N. Gonzalez,
E. Gonzalez-Gurriaran, R.B. Hanson, J. Iglesias, M. Krom, E. L6pez-Jamar, J.
McClain, M.M. Pamatmat, A. Perez, D.C. Rhoads, G. deSantiago, J. Tietjen, J.
Westrich and H.L. Windom. 1982. Coastal upwelling in the Rias Bajas, NW Spain:
contrasting the benthic regimes of the Rias de Arosa and de Muros. J. Mar. Res.
43:701-772.

Tenore, K.R. and N. Gonzalez. 1975. Food chain patterns in the Ria de Arosa,
Spain: an area of intense mussel aquaculture. Tenth European Symposium on
Marine Biology, Ostend, Belgium, Sept. 17-23, 1975, Vol. 2:601-619.

Tett, P. 1981. Modelling phytoplankton production at shelf-sea fronts. Phil.
Trans. R. Soc. London A 302:605-615.

Trojanowski, C.Trojanowska and H. Ratajczyk. 1982. Effect of intensive trout
culture in Lake Letowo on its bottom sediments. Pol. Arch. Hydrobiol.
29(3-4):659-670.

Trussel, R.P. 1972. The percent un-ionized ammonia in aqueous ammonia solutions
at different pH levels and temperatures. J. Fish. Res. Bd. Canada 29:1505-1507.

Tynan, T. unpub. Letter dated April 11, 1985 from T. Tynan, Squaxin Island
Tribe to G. Cloud, Washington Department of Ecology.

U.S. EPA. 1976. Quality Criteria for Water. U.S. Environmental Protection
Agency, Washington, D.C.

Van Houweling, C.D. and J.H. Gainer. 1978. Public health concerns relative to
the use of subtherapeutic levels of antibiotics in animal feeds. J. Anim. Sci.
46(5):1413-1424.

Vesco, L.L. and R.M. Gillard. 1980. Recovery of benthic marine populations
along the Pacific coast of the United States following natural and man-made
disturbances including pertinent life history information. POCS Ref. Pap. No.
53-4. Bureau of Land Management, Los Angeles, California.

Villwock, W. 1984. Some remarks on genetic influences of mariculture practices,
realized or potential ones, on the natural resources of populations. In Report
of the Working Group on Genetics, 1984, St. Andrews, Canada, 13-15 June 1984.
Appendix 1. International Council for the Exploration of the Sea C.M. 1984/F:4.

VKI. 1976. Vandkvalitetsinst. & Jydsk Teknologisk Inst.: Forskellige
driftsparameters indflytelse pa forureningen fra dambrug. [The influence of
different operational parameters on the pollution from fish farms]. Rapport til
Teknologiradet. Horsholm, Denmark. (not seen - cited in Ackefors and Sodergren,
1985; Warrer-Hansen, 1982)

Wallace, J.C. 1980. Growth rates of different populations of the edible mussel,
Mytilus edulis, in north Norway. Aquaculture 19:303-311.

Warrer-Hansen, I. 1982. Evaluation of matter discharged from trout farming in
Denmark. In J.S. Alabaster (ed.), Report of the EIFAC Workshop on Fish-farm
Effluents. Silkeborg, Denmark, 26-28 May 1981. pp. 57-63. EIFAC Tech. Pap. 41.

147

Washington Department of Ecology. 1973. Water quality standards, June 19, 1973.
Washington State Department of Ecology, Olympia. 17 pp.

Welch, E.B., R.M. Emery, R.I. Matsuda and W.A. Dawson. 1972. The relation of
periphytic and planktonic algal growth in an estuary to hydrographic factors.
Limnol. Oceanogr. 17(5):731-737.

West, P.A. and R.R. Colwell. 1984. Identification and classification of
Vibrionaceae -- an overview. In R.R. Collwell (ed.), Vibrios in the Environment.
pp. 285-363. John Wiley and Sons, New York.

Wilkins, N.P. 1976. Genic variability in marine Bivalvia; implications and
applications in molluscan mariculture. In G. Persoone and E. Jaspers (eds.),
Tenth European Symposium on Marine Biology, Ostend, Belgium, Sept 17-23. Vol.
1:549-563. Universal Press, Welteren.

Willoughby, H., H.N. Larsen and J.T. Bowen. 1972. The pollutional effects of
fish hatcheries. Amer. Fish. U.S. Trout News 17(3):6-7, 20-21.

Wooten, R. 1979. Tapeworm threat to trout in floating freshwater cages. Fish
Farmer 2(3):5.

Winter, D.F., K. Banse and G.C. Anderson. 1975. The dynamics of phytoplankton
blooms in Puget Sound, a fjord in the northwestern United States. Mar. Biol.
29:139-176.

Yake, B. unpub. Internal memorandum dated December 27, 1981 from B. Yake to D.
Cunningham, Washington Department of Ecology.

Zook, W.J., J. Rensel, J.M. Gearheard, J.C. Grover, B. Hopley, S. Rainey,
L. Rutter, T. Scribner, H. Senn and M. Stratton. unpub. Recommendations for
proposal and evaluation of salmonid production facilities. May 20, 1984.

148