&EPA
United States
Environmental Protection
Agency
Environmental Research
Laboratory
Athens GA 30613
EPA/600/9-86/024
August 1986
Research and Development
Problems of Aquatic
Toxicology,
Biotesting and
Water Quality
Management:
Proceedings of
USA-USSR
Symposium,
Borok, Jaroslavl Oblast,
July 30-August 1, 1984
-------
-------
DISCLAIMER
The information in this document has been funded in part by the United
States Environmental Protection Agency. Papers describing EPA-sponsored re-
search have been subject to the Agency's peer and administrative review, and
the proceedings have been approved for publication as an EPA document. Men-
tion of trade names or commercial products does not constitute endorsement
or recommendation for use by the U.S. Environmental Protection Agency.
11
-------
FOREWORD
For almost a decade and a half, cooperation and exchange ot scientific
information under the US-USSR Agreement on Cooperation in the Field of
Environmental Protection has helped both countries in their efforts to con-
trol environmental pollution. These efforts are pursued in recognition of
the international nature of the problem: pollution knows no boundaries.
Three projects are under the joint US-USSR Agreement's Working Group
on Cooperation in the Area of Water Pollution Prevention. These are:
Project 02.02-11 "Planning and Management ot Water Quality in River Basins,"
Project 02.02-12 "Protection and Management of Water Quality in Lakes ana
Estuaries," and Project 02.02-13 "Effect of Pollutants on Aquatic Organisms
and Ecosystems; Development of Water Quality Criteria."
Over the years, scientific delegations and individual scientists have
traveled to each other's countries to visit scientific institutions, pertorm
joint research, and exchange technical information. This Proceedings pre-
sents the papers that were delivered at the most recent formal symposium,
which was held in Borok, Jaroslavl Ob last in 1984.
Rosemarie C. Russo
Director
Environmental Research Laboratory
Athens, Georgia
USA
-------
PREFACE
The problem of water pollution and the control of water quality has
assumed international importance due to its urgency. The pooling of
efforts by scientists from different countries has helped to achieve suc-
cessful solutions. Within the context of the Soviet-American agreement in
the area of environmental conservation, provision has been made both tor the
carrying out of joint research and the holding of symposia at which special-
ists would be able to discuss results and outline prospects tor further work.
A regular Soviet-American symposium on the questions of water toxicology
was held in the USSR at the Institute for the Biology of Inland Waters Under
the USSR Academy of Sciences in Borok from 30 July through 1 August 1984. At
the symposium, eight Soviet and eight American papers were presented and the
materials of these are being published in the current collection. These
papers reflect the accomplishments of aquatic toxicology at the present stage
of its development and examines the following: water quality in natural
waters, question of utilizing indicators for the functional state of biota in
surface waters, approaches to biotesting of industrial effluents, and prin-
ciples of optimizing development programs for water conservation in the
system of water quality management considering point and non-point pollution
sources.
Of substantial interest are the results of studying the effect of com-
plex-composition effluents on hydrobionts, the resistance of aquatic animals
to organophosophorus pesticides and the possibilities for further transfer
of toxic organic substances. The spread of pollutants through the atmosphere
in a number of instances has substantial negative ecological consequences
and the study of this process is of important significance. In light of this
problem, within the context of Soviet-American collaboration, particular
attention is being paid to studying ammonia toxicity tor fish as well as to
the effect ot low pH values on aquatic animals.
The symposium, which was held in a spirit ot mutual understanding,
showed that the mutual interest of the United States and the USSR in study-
ing the problem of protecting bodies of water against pollution has success-
fully contributed in this area to the solving of problems confronting scien-
tists from both countries.
N. V. Butorin, Director
Institute for Biology of Inland Waters
Borok, Jaroslavl Oblast
USSR
IV
-------
ABSTRACT
Sixteen papers delivered by US and USSR scientists at a symposium en-
titled "Problems of Aquatic Toxicology, Biotesting, and Water Quality Man-
agement" are presented. The effect ot low pU on aquatic invertebrates and
the use of fish behavior in comparing toxic effects of chemicals are examined.
The production, and excretion of ammonia by fish, the relationship between
carbon dioxide excretion and ammonia toxicity, and the acute toxicity of iron
cyanides and thiocyanate to trout are discussed. Mechanisms of organophos-
phorus pesticide resistance are analyzed. Biological monitoring and testing
of surface waters and effluent is discussed and a complex effluents toxicity
information system is described. Processes in the formation of water quality
and trends in the development of water quality in the two countries are out-
lined. Pollution problems from toxic organic contaminants from point and
non-point sources in the North American Great Lakes are examined. Agricul-
tural water quality management, including the use of simulation models, is
discussed.
v
-------
CONTENTS
FOREWORD iii
PREFACE „ iv
ABSTRACT v
FUNCTIONAL BASES FOR THE EFFECT OF LOW pH ON FISH
AND INVERTEBRATES 1
G. A. Vinogradov
AMMONIA PRODUCTION AND EXCRETION BY FISH 19
D. J. Randall and P. A. Wright
THE USE OF FISH BEHAVIOR IN COMPARING TOXIC EFFECTS
OF THREE CHEMICALS 31
M. G. Henry
RESISTANCE OF AQUATIC ANIMALS TO ORGANOPHOSPHORUS PESTICIDES
AND ITS MECHANISMS 37
V. I. Kozlovskaya, G. M. Chuyko, L. N. Lapkina, and
V. A. Nepomnyashchikh
ACUTE TOXICITY OF IRON CYANIDES AND THIOCYANATE TO TROUT 55
R. V. Thurston and T. A. Heming
PROCESS OF FORMATION OF NATURAL WATER QUALITY 72
M. I. Kuz'menko and A. I. Merezhko
C02 EXCRETION AND AMMONIA TOXICITY IN FISHES: IS THERE A
RELATIONSHIP? 83
T. A. Heming
USE OF INDICATORS OF FUNCTIONAL STATE OF BIOTA IN BIOLOGICAL
MONITORING OF SURFACE WATERS 95
V. A. Bryzgalo, L. S. Federova, T. A. Khoruzhaya,
L. S. Kosmenko, and L. P. Sokolova
LONG RANGE TRANSPORT OF TOXIC ORGANIC CONTAMINANTS TO THE
NORTH AMERICAN GREAT LAKES 107
W. R. Swain, M. D. Mullin, and J. C. Filkins
BIOLOGICAL TESTING OF INDUSTRIAL EFFLUENT 122
A. M. Beym
vii
-------
COMPLEX EFFLUENTS TOXICITY INFORMATION SYSTEM 136
R. C. Russo, A. Pilli, and J. Crane
OPTIMIZING A PROGRAM FOR THE DEVELOPMENT OF WATER CONSERVATION
IN A SYSTEM OF WATER QUALITY MANAGEMENT CONSIDERING POINT
AND NON-POINT POLLUTION SOURCES. 150
G. A. Sukhorukov
GREAT LAKES AGRICULTURAL POLLUTION CONTROL
IN PERSPECTIVE 168
V. J. Saulys
MODELING THE FORMATION OF WATER QUALITY IN WATER
CHANNELS RECEIVING SURFACE RUNOFF FROM FARM LAND 181
Ye. V- Yeremenko, V. Z. Kolpak and N. I. Selyu
CONTROL OF URBAN NONPOINT SOURCES IN GREAT LAKES BASIN 201
D. Athayde, P- Bubar, and J. Meek
TRENDS IN THE DEVELOPMENT OF WATER QUALITY IN THE
USSR AND THE US 219
L. A. Lesnikov
Vlll
-------
ACKNOWLEDGMENT
Preparation of a Proceedings is frequently a complex task, particularly
when the authors are from two countries with different languages. Of great
assistance in the preparation of this Proceedings was the contribution of
Dr. Richard A. Schoettger of the Fish and Wildlife Service's Columbia National
Fisheries Research Laboratory who arranged for the translation of the Soviet
manuscripts. Also essential to its preparation was the dedicated work of
Martha M, Brady of Computer Sciences Corporation who typed the final document.
Their contributions are gratefully acknowledged.
IX
-------
FUNCTIONAL BASES FOR THE EFFECT OF LOW pH
ON FISH AND INVERTEBRATES
by
G.A. Vinogradov-L
ABSTRACT
Low, acutely toxic pH levels increase the ion permeability ol gill epi-
thelium in lish ana invertebrates. As a result 01 the increased permeabil-
ity, the rate of salt diliusion Irom the organism increases by several told.
Electrolyte concentration in the blood drops rapidly, there is an intense
intake ot hydrogen ions according to tne concentration gradient into the
inner medium, ana acidosis develops in parallel with salt elimination. In
lish with low pH values, there are destructive changes in the gill epithel-
ium and these develop most intensely in a medium containing less than 20 mg/£
Ca"*""*". The pH values that cause substantial changes in gill epithelium
permeability lor Na+ correlate with low limits ol pH values that are en-
countered by adult specimens ol various fish and invertebrate species under
natural conaitions. Two adaptive processes occur simultaneously in lish in
acclimation to low pH values. On the one hand, as a consequence of. the re-
duced gill permeability, the elimination of Na+ and Ci~ from the organism
is reduced; on the other, after a significant suppression of the absorption
of IMa"*" and Ci~, a partial recovery ot this process is observed. As a result
of acclimation, alter 24-4b h, the losses and absorption of Na are equal-
ized. This makes it possible to maintain an ion balance between the external
and inner medium on a lower metabolic basis than in neutral ana slightly base
media. In certain species of crustaceans and mollusks, calcium metabolism
is more sensitive to a decline in the pH than is sodium metabolism.
INTRODUCTION AND DISCUSSION
Early experimental work dealing with the effect of active environmental
reaction on rish established that 0^ uptake Irom water diminishes with de-
cline 01 pH (.Wiebe et al. 1^34;. The toxic effect was attributed to coagu-
lation ol gill mucus and membranes ot gill epithelium causing "anoxia ot
coagulate lilm" (.Ellis iy37). The more precise physiological effect ot this
film on the process ot oxygen uptake from water was investigated relatively
llnstitute of Biology of Inland Waters, USSR Academy of Sciences, Borok, USSR
-------
recently (Ultsch and Gros 19—) . It was established thar. mucus secreted
under the ertect ot acid provides only partial decrease in permeability
ot fish gills for oxygen. As a result, asphyxia develops very slowly and
does not cause death.
A decline or medium ph leading to acidulation of blood elicits in
most aquatic animals a decrease in affinity ot respiratory pigments tor
oxygen. Ihe dissociation curve shifts in the direction ot higher P02
(.Bohr effect) with decline ot pH. This reflects the correlation between
aftinity of respiratory pigments for oxygen and their acid-base properties.
On the basis of this information, as well as data concerning signmcant
coagulation ot mucus on fish gill epithelium, it was believed tor a long
time that the toxic etfect of low pH is related to asphyxia. Subsequent
studies refuted the role ot asphyxia in acid poisoning as the primary
cause of fish death (Vinograaov 1*7*, hddy 1974). Acidulation ot medium
to a pH of 4.5 elicits a decrease in oxygen uptake in the carp, and it is
restored within 24 h. After acclimation ot fish to hypoxia, oxidation ot
water did not cause a decrease in oxygen uptake. The results ot experi-
ments showed that there is something in common in adaptation mechanisms
that react to medium oxidation and hypoxia (hogluna 19t>l, Vinogradov 1979).
Respiratory systems adapt to ph decline primarily by means of increase
in oxygen capacity or blood due to increase in hemoglobin, hematocrit,
and erythrocyte content (Vaala and Mitchell 1970).
At the present time, it can be considered established that the toxic
etfect of low medium pH is based on functional impairment ot systems ot
ionic and osmotic regulation localized in the gill epithelium of aquatic
animals (Maetz 1974, McWilliams 1980, Packer and Dunson 1970, Packer and
Dunson 1972, Spry et al. 19til, Vinogradov et al. 197b, Vinogradov et al.
197b). At the present time, it is assumed that specialized chloride cells
that maintain a specitic Na+ level in blood are possibly the location of
systems ot ion transport against a concentration gradient in fish (Laurent
and Dunel 19bu). The main element of ionic regulation in freshwater ani-
mals is active transport of Ma+, K+, Ca"1"*" and Cl~ by the gills from water
to the endogenous environment, as well as removal trom the body ot metabo-
lites such as NH4 H1", hC03 (Maetz 1973, Payan et ai. 19bl). Studies
conducted recently furnished more information about the mechanism of regu-
lation of acid-base equilibrium in fish and crustaceans. It was found that
Na /H , Na /NH4, HCOyCi turnover in gills of freshwater fish is of
substantial significance to maintaining blood ptt (Cameron and Randall 1972,
Cameron and Randall 1974, Eddy 1974, Randall and Cameron 11*73, Renzis and
Maetz 1973;. The correlation between ion uptake through the gills and
acid-base regulation was studied comprehensively by Haswell et al. (19t>0)
and Perry et ai. (,19»1). The rather low permeability of gill epithelium
tor water and salts plays an important part in adaptation to lite in fresh
water (.Packer and Dunson 1970, Packer and Dunson 1972, Vinograaov et al.
197D, Vinogradov et ai. 197b). Exogenous factors, which increase gill
permeability, should cause excessive discharge or electrolytes and cause
penetration into the body ot undesirable substances.
Another characteristic that is very important to processes ot ionic
regulation in freshwater animals is affinity of ion-absorbing systems ror
-------
concentration 01 substrate (Na"1", K+, Ca"1""1", Cl ) in the exogenous environ-
ment, which is usually referred to as the semi saturation constant (hams
19/2, Shaw 19b9, Sutclille 19bb, Sutclille 1971, Vinograciov 197b). In
typical treshwater species, soaium absorption begins with a concentration
ol about U.b mg/Jl, while saturation occurs at b-iU mg/i Na+. The values
are consiaerably lower lor potassium ions. The kinetics ot Ca"*"1" absorp-
tion are similar to those lor Na+ ions. Ca"^" absorption begins at concen-
trations ot 0.4-U.^ mg/£. This process reaches a maximum rate at 1U-4U
mg/J?,, and the values lor Na"*~ and Ca"1""*" ions are similar. The rate ol K
absorption Irom water is 1/bth-i/iUth the rate lor Na+ ana Ca"1""1" (Vinogra-
dov et al. 19«3, Vinogradov et al. 19b4).
It was established in several studies that a decline ot ph increases
significantly loss of Na"1", K+, Cl~ and Ca^"1" ions in lish, crustaceans and
mollusks (Packer and Dunson 197U, Packer and Dunson 1972, Vinograaov et
al. 197b, Vinograaov et al. 1979, Vinogradov et al. 1963;. Oxidation ol
the medium causes manifold increase in overall Na~*" loss in lish (Figure la)
and impairs Ca*~*~ metabolism. Similar findings have been made in crusta-
ceans ana mollusks (Figure Ib). Tnreshola pH levels impairing gill perme-
ability tor ions are dilferent in different animal species. In the perch
(Perca tluviatilis L.), which is one ol the most acid-resistant and common
fish species, in reservoirs with high concentration of hydrogen ions,
critical values ot pH are considerably lower than lor other species ot
fish. Investigation ot rate ot loss in perch inhabiting mildly alkaline
water (Rybinskiy Reservoir, pH 7.fc>-y.U) and acid lakes (Lake Motykino and
Lake Dubrovskoye in Volgograd Oblast, pH 3.b-4.2) revealed that permea-
bility ot gill epithelium tor Na+ is one-halt less in lake fish than those
in reservoirs (rivers).
1. Perch (Perco fluviolilis)
2. Crucian Carp [Carossius aurotus)
3. Roach (Rutilus rutilus)
4. Current Year Salmon (Solmo salon)
5. Ca**, Current Year Salmon (^ solar)
1000
c
o
- 500
in
v>
O
o
o
100
0
1. Freshwater Shrimp {Gommarus locustris)
2. Crayfish (Astocus leptodoctylus )
3. Snail ( Limnoeo stagnalis )
4, Branchipod ( Slreptocephalis joseflnae)
400
300
200
100
6 1
Water pH
B
Figure 1. hftect ot low pH on rate ot total loss ot Na"1" in tish(A) ana
invertebrates (ii).
-------
These aifrerences in permeability become insignificant atter perch
irom an alkaline environment unaergo b-aay acclimation to water with pH
3.J5 (.Table 1), The threshold values lor ambient pH eliciting arastic in-
crease in gill permeaoility lor ions, as well as resistance to oxidation,
are the same in both groups ol perch. These data indicate, in our opinion,(
the phenotypic nature of differences in fish from dilferent bodies of water.
TABLE i. EFFECT OF ACCLIMATION TO ph 3.35 OM Na+ LOSS IN PERCH FROM LAKE
AND RESERVOIR POPULATIONS
Acclimation,
days
u
3
b
Na"1"
lake
U. 25+0. 14
0.22+0. Ob
o.iy+o.oy
loss, pmol/(g'h)
reservoir
0.4«+0.12
0.25+0.13
0.24+0.11
As shown by field and laboratory studies, decrease in sodium concentra-
tion of blood plasma and its oxidation result from increased permeability of
gills for ions (.Leivestad and Muniz J.y7b, Maetz iy73, Packer and Dunson
iy7u). Data in the literature indicate that decline of Na+ content of blood
and hemolymph of crustaceans is the chief cause of rapid death of fish and
crustaceans in fresh water at low ph (.Vinogradov et al. iy7b, Vinogradov
lb*7y, Vinogradov iybi, Vinogradov et al. lybJU. Figure 2 and Table 2 provide
data on change in concentration of Na+ and ph in the endogenous environment.
Studies ol survival revealed that freshwater shrimp ^Gammaracanthus
lacustris; tolerates greater oxidation ol medium (.ph 4.1-4.2.) in diluted
salt water than in fresh water (pti 5.0-5.JJ. Death of the shrimps in
diluted salt and fresh water occurs at dilferent hemolymph pH and different
concentrations or ions.
In ireshwater shrimp adapted to salt water, Na+ transport through the
gills is inactivated. Ion content of hemoiymph is due to their levels in
the external environment, and the cavitary fluid is isotonic in relation to
the external environment, hemolymph is hypertonic in fresh water, in re-
lation to the external environment.
In diluted salt water, these animals die- at hemolymph pH b.3-6.4 and
in fresh water at higher ph. Ion content of hemolymph in shrimp from
fresh water diminishes with oxidation of water and reaches the same criti-
cal values before death as in desalinated distilled water. The toxic
eltect of low ph is drastically enhanced when Na+ concentration in water is
less than O.b-0.7 mmoi/£. Gammaracanthus adapted to water with 0.4 mmoi/
NaCl died at ph 5.5-5.7. This is probably related to diminished absorp-
tion or Na~*~ irom water containing less than U.b mmoJ-/£, as well as ae-
crease in sodium concentration and osmotic pressure ol hemoiymph due to
-------
o
CD
c
c
o
o
6 -
o
c
o
o
o 4
o
2
X
a.
Hn
^,
B
I 2 3
Condition
(l-distilled water, pH 3.7;
2-distilled water, pH 5.6-6-5;
3- fresh water, pH 7.5 )
T>
O
O
CD
ion
c 6
0)
o
o 5
0 4
O
o
Hh
I
•d
I
Condition
(l-distilled water, pH 3.8;
2- river water, pH 3.8;
3-distilled water plus 70mg/ICa++,
4- river water, pH 7.4)
figure 2. ph ana Na+ concentration in blood ot roach iRutilus rutilusj (A)
and crucian carp (.Carassius auratus) (B) as a runction ot di±±er-
ent ambient conditions, according to Vinogradov et al.,
Vinogradov et ai., I^b3, Vinogradov et ai., lb»7b.
TAULb 2. CONCENTkAllON OF ELECTROLYTES AND ph 01 HEMOLYMPH bEFORE DEATH
OF GAMMARACANTHUS LACUSTRiS IN ACID MEDIUM AND DISTILLED WATER
Number o±
Electrolyte and Na~*~ concentration expen-
Acclimation in hemolymph, ymol (scaled to NaCl) Hemoiymph mentai-
ph environment electrolytes sodium pH animals
3.1 Diluted salt
water (L27°)
3 .3 Fresh water
3.5 Diluted salt
water (12%)
4 .5 Fresh water
7.0 Distilled
water
310 + 15
105 + 10
305 + 15
105 + 15
255 + 5
5.35 +0.1
100 + 10 6.6 +0.2 4
255 + 10 6.4 + 0.1 5
yo + 15 b.y5 + 0.15 6
yo + 5 7.6 + o.i 6
-------
insult icient affinity o± the sodium-transporting system in this range or
concentrations. Among crustaceans, as in ±ish, acid-resistant species
encountered in acid waters do not present excessive loss of salts in ex-
periments over a wide range of changes in concentration of hydrogen ions.
According to our data, threshold pH increasing discharge of Na"1" in the
hog slater Usellus aquaticus) is 3.1-3.2 and for the freshwater shrimp
iCammarus lacustris) 5.5-b.b, which is consistent with the results of
field studies (Malley
Thus, the viability of some species of freshwater fish and crusta-
ceans in acid waters is largely determined by the tolerance o± integument,
primarily gill epithelium, to low ambient pH without change in permeabil-
ity of these tissues for monovalent Ma"1", Cl and HT" ions.
General tissue permeability to substances is determined by its cellu-
lar (.membrane) and intercellular elements. The correlation between rates
of excessive loss of Na, K+ and CX~ ions through fish gills in an acid
medium corresponds to the concentrations of these ions in blood and their
mobility (Vinogradov et al. iS>79). This is indicative primarily of in-
crease in intercellular permeability in an acid medium, hlectron micro-
scope studies of reaction of fish gill epithelium to an acidulated medium
revealed that the increase in passive exit of Na+ under the effect of low
pH is attributable to impairment of intercellular interactions in gill
epithelium. Thus, in the crucian carp placed in water at ph 3.t>, there
is impairment of integrity of external cell membranes and lysis of some
parts of them (Vinogradov et al. 19b3). There is widening of intercellu-
lar spaces in the region of simple connections. In some cases, there is
partial impairment of solid connections of the apical segments of gill
epithelial cells.
On the basis of data concerning excessive loss of Na"1", K+ and Cl~
in an acid medium, as well as the capacity of calcium ions to normalize
these losses, the hypothesis was expounded that the protective effect
of Ca"1""1" on fish at low pH is based on the properties of Ca"*"1" to cement
intercellular contacts with exposure to factors that disintegrate the
gill epithelium (Vinogradov et al. i97y). With pH 3.6, Ca"1"1" normalizes
not only Na+ metabolism in Carassius carassius L. , but maintains acid-base
homeostasis. In the absence of Ca"1"1" ions, there is a drastic drop of
blood Na"1" level. It is oxidized (Figure 2) and fish death occurs within
7-8 h. Ga"1"1" capacity to stabilize in an acid medium both sodium and acid-
base homeostasis indicates that Ca"1""1" ions limit gill permeability not
only to Na+, but to h+.
It was established in several works that Ca"1"1" and Kg"*""1" ions affect
permeability of gill epithelium for Na"1" and Cl~ in freshwater and euryha-
line fish (Evans iy7i>, Maetz iy/4, Potts and Fleming iy — ). In our exper-
iments, we studied the effect of Ca"1""1" on permeability of gill epithelium
of the crucian carp (Carassius auratusj tor Ma+ in a neutral (ph 6.&-7.U)
and an acid medium (ph 3.h). It was shown that, with a neutral medium reac-
tion, the increase in Ga"1""1" concentration in water lowers the discharge of
Na"1" from the body. An increase in Ca"1"*" content beyond 30 mg/ £ has virtu-
ally no eitect on loss ot Ma"1" through the gills, whereas a decrease in
-------
Ca"1"1" concentration to less than ID mg/ a elicits drastic increase in per-
meability ot gills in the crucian carp. A drop of pH ±rom b.b to 3.tt in
distilled water increases by many times the rate of loss of Na+ in fish
(.Figure 3a). These data indicate that an increase in Ca"1^" concentration
in water has an efrect on gill epithelium that is the opposite of that of
decline of ptt of the medium.
The main changes in gill permeability for Na+ are observed within the
rirst li> mm after addition of Ca"^ to both an acid (ph 3.6) and neutral
(.pH b.b) medium (.Figure 3b>. However, Na+ loss decreases to less than
1/bth in acidulated distilled water with added Ca"^1", as compared to a
neutral medium. The dynamics ot the process of increase in permeability
in crucian carp placed in distilled water at pH 3.b are inversely propor-
tionate to tne change under the eftect ot Ca"^". The most significant
increase in permeability is observed within the first few minutes of ex-
posure or the fisti to an acid medium (.Figure 3b;.
Investigation of total ha"1" loss by fish as related to change in ion
composition of water established that the decline in yield ol Na"1" with
decline ot water pH is not a specilic reaction. A decrease in total Na
loss is observed in the perch (.Perca fluviatilis) and salmon fry (.Salmo
salar) when kept tor a long time in distilled water and CaCl^ solution
(.Figure 4). Analysis of the results enables us to conclude that insuf-
ncient intake of Ma"1", regardless ol causes, leads to decrease in Na+
I. Permeability with Change in Medium pH
2. Permeability with Change in Ca+* Content, pH 6 8-7.0
3. Effect of pH 3.8 on Permeability in Distilled Water
4. Effect of Ca*+ (70mg/l) on Permeability, pH 3.8
5. Effect of Ca**(70mg/l) on Permeability, pH 6 85
A.Gill Permeability as Function of Ca++
and H+ Concentration in Medium
200
+
D
400
JD
o
a>
E
k_
-------
o»
0)
0.2
O.I
I, Perch, river water, pH 4,0
2, Perch, distilled water (DW)
3. Perch, DW plus Ca++
4, Salmon, DW
Acclimation Time, days
ligure 4. Eitect oi aesalinadon ana low pH on Ma loss in the perch
(.Perca tluviatilis) and current year salmon (Salmo salar).
loss and, probably, decrease in permeability ot gill epithelium. Evi-
dently, Ca^"4" ions are not ot basic signi±icance to regulation ot gill
permeability tor Na~*~ under natural conditions at optimum pH values in
treshwater tish capable ot living in waters with low mineral content.
Unlike the phenomenon ot increased gill permeability, which is dis-
tinctly manitested only with lethal or close to lethal ph values, in-
hibition ot Ma"1" transport through the gills ot treshwater tish is observed
in the tolerated range ot pH values (.Maetz 1S*73, Maetz iy~/4 . Packer and
Dunson 197U, Packer and Dunson iy?2). The extent ot inhibition ol Na+
absorption ot iish depends on ambient ph (.Figure b). However, the data
obtained in short-term experiments reliect only the initial stage ol
reactions ot sodium- and chlorine-transport systems ol treshwater tish to
oxidation ot medium. Studies dealing with acclimation or tish to decline
oi medium ph revealed that there is partial or complete restoration ot
Ma+ absorption tunction during acclimation, depending on intensity ot
-------
exposure and tish species (Sokolov and Vinogradov 1977;. The capacity ot
the soaium-transport system o± iish to adapt to acidulation ot medium was
continued in subsequent works (McWilliams 19bU, Thurston J.y7ya, Vinograaov
et al. iy?y,). For exampie, with acclimation oi the crucian carp to a ph ot
5.5, the rate ot Na+ absorption was signiticantiy slower already within 3~b
h than in control tish (52-407.,,). Na+ sorption increased 12 h atter the
start ot the experiment, and it constituted 707, and 7'bX ot initial level
atter 24 and 4b h, respectively (Figure fa). A decline ol pH to 4.5 atter
4b-h acclimation ot the crucian carp to a pH ot 5.5 again elicited a decrease
in Na transport. Na+ absorption atter 7 days ot acclimation constituted
only 5U% ot the Na+ sorption level in control tish (Thurston et al. 'iy7ya).
I. Perch (Perca fluviatilis)
2. Three-spined Stickleback (Gasterosteus aculeathus)
3. Crucian Carp (Carassius ouratus)
4.Bolti (Tilapia mossambica)
5.Current Year Salmon (Salmo solar)
100
90
80
70
60
50
40
30
20
10
0
c
o —
o §
4 5
Water pH
7
Figure 5. Na"1" absorption in tish as a iunction ot ambient pH, according
to Vinogradov et al., 197y.
-------
Decrease in activity o± diiferent enzymes could be a probable cause
oi depression ot Na+ and Cl~ transport in ±ish giils with change in ambient
ph. This applies, first o± ail, to Na+, K+, Mg "^-activated Al'Pases and
succinate dehyarogenase (.SDH), which are enzymes directly involved in trans-
membrane ion transport. At the present time, it is assumed that active Na+
transport through ceil membranes is attected by the tunction oi JNa+, K+-
ATPase, which is a good explanation lor the presence oi ionic gradients
between intracelluiar and extracellular media.
Na , K -ATPase, however , constitutes an insignilicant part of overall
ATPase activity 01 gill tissue to maintain ionic assymetry between cells
oi the gill epithelium in iresh water, which has very low levels of ions
(.Thurston et al . 1979 b). Strophanthin-resistant ATPase plays the leading
role (.over 9u% ot all ATPase activity). The optimum ph lor Na+, K+-
ATPase and strophanthin-resistant ATPase is in the range of 7.5 to b.5,
and it^does not change when fish are acclimated to a more acid medium.
ATPase activity remains stable during 10 days of acclimation of
arp t0 a ph ot 5>i' The strophanthin-resistant component ot overall
activity increases during acclimation to low ph. Reliable ditter-
ences were noted one day after the start ot the experiment (figure 7).
the
\.
2.
3.
4.
5.
6.
Rate of Total Na+ Loss in Crucian Carp, pH 5.5
Na* Absorption in Carp, pH 5.5
Rate of Na* Loss in Stickleback, pH 5.0
No* Absorption in Stickleback, pH 5.0
Rate of Total Na* Loss in Crucian Carp, pH 4.5
Na+ Absorption in Crucian Carp, pH 4.5
2345
Time, days
7
Figure 6.
Effect ol low pil on Ma"1" exchange in crucian carp (Carassiu.s
auratus., according to Thurston et ai., iy79, ana stickelback
(Gasterosteus aculeathus), according to Matey et al..,
10
-------
I. Control 2. 7 Days 3. 10 Days
a. Strophanthin-resistant Element of ATPase Activity
b. Na+ K+ -ATPase Activity
7 8
pH of Incubation Medium
9
figure /- ATPase activity in crucian carp (.Carassius auratus; gilis with
acclimation to pH ^.5, accoraing to Thurston et ai., 1^/y.
These results warrant the belief that, in freshwater fish, Strophanthin
(,ouabain)-insensitive ATPase plays an important part in ion regulation.
The results indicate that there are two concurrent adaptive processes
in tish during acclimation to low pH. On the one hanci, there is less loss
of Na"t from the body and, on the other hand, atter -an initial significant
inhibition o± Na+ absorption, there is partial restoration of rate ot sorp-
tion ot this ion, which makes it possible to maintain a balance between
external and internaT Ma"1".
In the crayfish, Astacus leptodactylus, Na+ exchange in gills, like
in the hog slater, Asellus aquaticus, is not sensitive to pH drop in the
range or t>.0-4.0. however, acidulation ot the environment elicits a
drastic increase in loss ot Ca4"*". In craynsh placed in water with pti ot
4..}, the rate or general loss ot Ca"1"1" increases by b-y times, as compared
to the control, and remains unchanged tor 7 days. Threshold pti at which
Ca"*"1" absorption does not compensate tor its loss is 4.t>-4.7 C^'igure ba).
At neutral pH, Ca"1"1" sorption in crayfish is 7-b times greater than loss
(.Vinogradov et al. iyb3) . This creates conditions tor accumulation and
deposition in the body ol Ca"*"1" required tor craylish growth and ecdysis.
11
-------
Apparently, impairment ot Ca"*""1" exchange, along with changes in regulation
ol Na"1" ana Ca"*""*" uptake when the environment was aciduiatea (.Figure tta ana
b) . Investigation ot Ca"1"1" ana Na"1" metabolism revealed that there is in-
crease in total loss ol Ca"1"1" and Na+ in moliusks starting at a ph ol less
than b.b. In our opinion, impairment ol Ca"1""1" metabolism is ol primary
significance to survival ol mollusks, since the intensity ol their uptake
ol Ca""" is substantially greater than its uptake by lish and crustaceans
(.Figure 4 to dillerent pH values alter placing ireshwater shrimp,
Gammaracanthus lacustris, in it revealed that total electrolyte content
of water at pH ol less than 4.3 diminishes drastically, in spite ol the
significant migration into water ol Na"1" and Cl~ ions from the body (.Figure
iU,). The results of this experiment are indicative ol considerably great-
er change in gill permeability lor hyarogen ions than lor salts at low
environmental pH. Analogous data, which revealed that permeability o±
the integument ol crustaceans lor h+ in an aciduiatea environment is much
greater than lor other ions and that intake of H+ exceeds total output of
salts, were also obtained tor other crustacean species.
On the basis of many years of our own studies and analysis ol aata
in the literature, it was aetermined that lish and invertebrate death with
decline 01 ph. was aue essentially to impairment ot processes or ionic reg-
ulation. In an acid environment, exchange oi Na"1", Ci~ ana Ca4"*" Between
the organism ana water shifts in the airection of excessive escape ot
these ions into the environment, ana there is unbalanced intake ot H+.
Low, acutely toxic pH levels increase permeability of fish gill epi-
thelium tor ions. In this case, the rate ot ailtusion ot salts trom the
body increases by several times. Blooa electrolyte concentration ae-
I.
2.
3.
4.
05
Ca++ Uptake in Mollusk (Sphoerium suecicum)
Ca++ Loss in Mo I lusk (S. suecicum)
CQ++ Uptake in Crayfish (AJ^qtodactyjus)
Ca+ + Loss in Crayfish (A. leptodactylus)
(Sphaerium suecicum)
I. Ca++ Uptake 3. Ca++Loss
r3. Na+Uptake 4. Na+ Loss
B
4.0 4.5 5.0 5.5 6.0 6.5 7.0
Water pH
4.0 45 5.0 5.5 6.0 6.5 7.0
Figure a. Uptake ana loss ot ions at different environmental ph in
invertebrates.
12
-------
creases rapidly, there is intensive passage ot hydrogen ions over concen-
tration gradient to the endogenous environment, and concurrently with
desalination there is development of acidosis. At such pH values, there
are serious destructive changes in the gill epithelium, which are the most
intensive in a medium with less than 20-40 mg/£ Ca++.
Low sensitivity of transepithelial Na+ transport and giii permeability
tor ions with drop ol ambient pH are typical features o± one of the most
acid-resistant fish species, the perch. In an acid medium, Ca"*"1" ions
normalize both sodium ana acid-base homeostasis in fish thanks to the
capacity of this cation to limit giii permeability tor Na+ and H+ by means
ot "cementing" intercellular contacts and stabilizing external cytopiasmic
membranes.
I. Ca++ Uptake by Sphaerium suecicum
2. Na+ Uptake by S. suecicum
3. Co"*"1" Uptake by Limnaea peregra
4. Na* Uptake by L. peregra
a>
j*.
o
0.75
o \
0 o
"= H
o ••*.
0.25-
5 10 15 20
Ion Concentration, mg/l
Figure y. Ca"1"1" and Na+ uptake in treshwater mollusfcs Limnaea peregra
and Sphaerium suecicicum as a function ot concentration ot
these ions in water.
13
-------
PH values tnat elicit substantial changes «
epithelium tor Na+ correspona to a low range or pH £
mens ot dilterent species are encountered under natural
»*
in the tolerated range ol low PH, lisn
chan.es in activity ot metabolic process. « H iniially
organization ol gil± epithelial cells. A J^1^ ° o± cellSj anQ SDH
depresses Na+ uptake, stimulation ot synthetic activity or '
activity in gill epithelium. Suosequently , development ol these processes
slows down and at the next stage there is partial restoration ot Na
transport, decrease in permeability ot gill epithelium "abiization ot
metabolism, but usually at a lower level tnan in a neutral and mildly
alkaline environment.
Strophanthin-resistant ATPase is apparently very important to osmotic
regulation in treshwater fish. Activity ot Na+ , K^-ATPase constitutes an
insignificant part ot overall ATPase activity of gills. Ihe role ot Na ,
K+-ATPase in gills of treshwater fish is apparently limited to regulation
ot intracelluiar proportions of Na"1" and K .
.c
a>
I. Rate of Total Na* Loss (Gammaracanthus iacustris)
2.Total Concentration of Electrolytes (G_. lacustris)
3.Rate of Total Na* Loss (Gommarus lacustris)
4.Total Concentration of Electrolytes (G_. lacustris)
0.7
(O U»
o E 0.5
+ cf
Z o
H- ^ 0.3
O T3
-------
There is impairment ol processes or ion regulation in the case ot
acidulation, in crustaceans ana moiiusks, as well as iish. One observes
increased loss ol Na+ and Ca"1""*", and inhibition ol Na+ uptake Irom the
environment. Calcium metabolism is more sensitive to a decline of pH than
sodium metabolism in some species ol crustaceans and moiiusks. This war-
rants the beliel that the changes in regulation ol Ca"1"1" that occur in an
acid medium could restrict the existence ol some species in waters with
low environmental ph.
BIBLIOGRAPHY
Cameron, J.N. and U.K. Randall. 1972. The eflect oi increased ambient
CO^ on arterial CO^ tension, C02 content and ph in rainbow trout.
J. Exp. Biol. 5/ (3) . o73-ottU .
Cameron, J.N. and i .A; Poihemus. 1974. Theory ol C02 exchange in trout
gills. J. Exp. Biol. bO(l) ; Ib3-194.
Eddy, t.b. 1974. Acid-base balance in rainbow trout (Salmo Gairdneri)
subjected to acid stresses. J. Exp. Biol. b4(l) : 159-171.
Ehrenlela, J. 1974. Aspects ol ionic transport mechanisms in craylish
Astacus Leptodactiius. J. Exp. Biol. 51 (1) :57-7U.
Ellis, M.M. 1W3. Detection and measurement ol stream pollution. Bull.
Bur. fisheries. 4b(.22) :3b5-437 .
Evans, D.H. 1975. Ionic exchange mechanisms in lish gills. Comp. Bio-
chem. Physioi. 511,491-495).
Harris, R.R. 1972. Aspects or sodium regulation in a brackish water
and a marine species ol the isopod genus sphaeroma. l2(l):b-27.
Haswell, h.S., O.J. Randall, ana b.F. Perry. 190U. Fish gill carbonic
anhydrase. acid-base regulation or salt transport? Am. J. Physioi.,
23b.24U-2btt.
Hogiund, L.B. 1901. The reaction ol lish in concentration gradients.
Rep. Inst. Freshwater Res., 43:1-147-
Kerstetter, T.H., ana L.B. Kirschner. 1972. Active chloride transport by
the gills ol rainbow trout (.Salmo Gairdneri). J. Exp. Biol. 5b(.l).
2b3-272.
Laurent, P. and S. Dunel. 19bU. Morphology ol gill epithelia in lish.
Amer. J. Physiology. 23«(3):147-159.
Leivestad, H. and 1. Muniz. 197o. Fish kill at low pH in a Norwegian
river. Nature (London). 259:391-392.
Maetz, J. 197/. Branchial sodium exchange and ammonia excretion in the
goldlish Carassius Auratus. Ellects oi ammonia-loading and temper-
ature changes. J. Exp. Biol. 5b.6Ul-blU.
15
-------
Maetz, J. 1*73. Na+/<, Na+/H+ exchanges, NH, movement across gill of
Carassius Auratus. J. Exp. Biol. bb(l):25b 2/5.
Maetz, J. 1974. Origin or dirierence in trans branchial electric potential
in Carassius Auratus goldfish. Importance ot Ca Ion. C.R. Acad.
Sci. (Pans). 279:1277-12bU.
Maetz, J. and J?. Garcia-Romeu. I9b4. The mechanism of sodium and chloride
uptake by the gills ol a treshwater tish Garassius Auratus. L. tvi
dence for Nb£/Na+ and HCO'/Cl" exchanges. J. General Physiology.
47(7):1U29-1229.
Malley, D. i9b(J. Decreased survival and calcium uptake oy the crayfish
Orconectes Viridis in low ph. Can. J. Fish. Aquatic Sciences.
37.364-372.
Matey, V.Ye., A.D. Kharazova, and G.A. Vinogradov. 1961. Reaction ot
chloride cells o± gill epithelium or stickleback, Gasterosteus Acul-
eatus L., to change in pK and salinity 01 environment. Tsitologiya.
23(2) .159-lb5. pp. 159-lb5.
McWilliams, P- 19bU. Eliect ol ph on soaium uptake in Norwegian brown
trout (Salmo Trutta) trom an acid river. J. Exp. Biol. bb:259~2b7.
Packer, R.K. and W.A. bunson. 19 /(_). Effects of low environmental ph on
blood ph and sodium balance ot brook trout. J. hxp. Zool. 74(1).
fai-72.
Packer, K.K. and W.H. Dunson. 1972. Anoxia and sodium loss associated
with the death of brook trout at low ph. Coinp. Biochem. Physiol.
41A.17-26.
Payan, P., N. Mayer-Gostan, and P. Pang. 19bl. bite 01 calcium uptake
in ionic treshwater trout gill. J. Exp. Zool. 21b:345-347.
Perry, S.*1., M.S. Haswell, D.J. Randall, and A.P- Farrell. 19bl. Bran-
chial ionic uptake and acid-base regulation in the'rainbow trout
Salmo Gairdneri. J. Exp. Biol. 92:2b9-3U3.
Potts, W.T.W. and W. Fleming. (No year given;. The effect ot prolac-
tion and divalent ions on the permeability to water ot the Fundulus
Kansae, J. Exp. Biol. 53:317-327.
Randall, D.J. and J.W. Cameron. 1973. Respiratory control of arterial
pH as temperature changes in rainbow trout Saimo Gairdneri. Arner.
J. Physiol. 225(5;.997-1UU2.
Renzis, G. and J. Maetz. 1973. Studies on the mechanisms ol chloride
absorption by the goldtish gill, relation with acid—base regulation.
J. Exp. Biol. b9(,2; .339-3^6.
Shaw, J. 19i>9 . The absorption o± sodium ions in the craylish Astacus
Pallipes Lereboullet. 1: The erlect o± external and internal soaium
concentrations. J. Exp. Biol. 3b(. i ) : l2b-144 .
16
-------
Sokolov, V.A. and G.A. Vinograaov. i977. investigation of fish adapta-
tion to ditierent environmentai pti. In. biologiya vnutrennikh vod.
Inform, byui. (Biology of Iniand Waters. Information Bulletin). No.
33. Leningrad, USSR. pp. f>t>-b9.
Spry, D., C. Wood, C. and P- Hodson. iybl. The effects of environmentai
acid on freshwater fish with particular reference to the soft water
lakes in Ontario and the modifying effects of heavy metals. A liter-
ature review. In; Canadian Technical Report of fisheries and Aquatic
Sciences. Toronto, Canada. 999 pages.
Sutclitte, D.W. 19bb. Sodium regulation and adaptation to fresh water in
Gammarid crustaceans. J. Exp. Biol. 4b:33i>-3bU.
Sutclitte, D.W. 1971. Soaium intiux and loss in freshwater and brackish-
water population of the amphipod Gammarus Duebeni Littjebirg. J.
Exp. Jiioi. 54^1,1 .2bi>-2bb .
Thurston, K.V., G.A. Vinogradov, V.T. Komov, ana V.\u. Matey. 19/9.
Effect of low pH, ammonia salts and desalination on enzyme activity,
sodium metabolism in gills ana ultrastructure of chloride ceils in
freshwater fish. Report 1. in: Bioiogiya vnutrennikh vod. Iniorm.
byul. pp. 7b-bl.
Thurston, K.V., G.A. Vinogradov, V.T. Komov, and V.Ye. Matey. 197yb.
Effect of low pti, ammonia salts and desalination on enzyme activity,
sodium metabolism in gills and ultrastructure of chloride cells in
freshwater fish. Report 2. In: Bioiogiya vnutrennikh vod. Iniorm.
byul. No. 44. pp. 75-79.
Ultsch, G. and G. Gros (No year). Mucus as a diffusion barrier to oxygen:
Possible role in 0^ uptake at pH in carp (Cyprinus Caprio) gills.
Comp. Biochem. Physiol. b2A;bbb-bb9.
Vaala, S. and R. Mitchell. 197U. Blood oxygen tension changes in acid
exposed brook trout. Proceed. Pennsylv. Acad. Sci. 44:41-44.
Vinogradov, G.A. 197b. Osmotic reguiation of some relict glacier crus-
taceans as related to distinctions of their ecology and origin. In
Solenostnyye adaptatsu vodnykh organizmov (Salt Adaptation of Aquatic
Organisms). Leningrad, USSR. pp. I7b~209.
Vinograaov, G.A. J.979. Adaptation of aquatic animals with different types
of osmotic regulation to decline of environmentai pH. In. Fiziologiya
i parazitoiogiya presnovodnykh zhivotnykh (.Physiology and Parasitoiogy
of Freshwater Animals). Leningrad, USSR. pp. i7-2t>.
Vinogradov, G.A. 19bi . Processes of ionic regulation in freshwater animals
in waters with anthropogenic pollution. In Bioiogiya vnutrennikh voa
(Biology of Inland Waters). No. M. Leningrad, USSR. pp. i>3-5b.
17
-------
Vinogractov, G.A., P.A. Gdovskiy, and V.Ye Matey. 1979. Oxidation ot water
reservoirs and its effects on metabolism of freshwater animals. In:
Fiziologiya i parazitologiya presnovodnykh zhivotnykh. Leningrad,
USSR. pp. 9b-ii6.
Vinogradov, G.A., Ye.S. Dai', and V.T. Komov. 1983. Investigation ot
basic gill functions of crayfish exposed to ammonia salts and acidu-
lation of environment. In. Reaktsiya gidrobiontov na zagryazneniye
(Hydrobiont Reaction to Pollution). Moscow, USSR. pp. 168-i75.
Vinogradov, G.A., V.T. Komov, and V.Ye. Matey, 1964. Functional bases of
effect of low pH on fish and crustaceans. In: Fiziologicheskiye i
biokhimicheskiye aspekty toksikologii presnovoanykh zhivotnykh" (Physi-
ological and Biochemical Aspects ot Toxicology of Freshwater Animals)
Manuscript filed with V1NIT1 (Ail-Union Institute of Scientific and
Technical Information) , IhVV (.Institute of biology of Inland Waters) ,
USSR Acaaemy of Sciences, on ^5 March 1984, No. 163/64. pp. 147-190.
Vinogradov, G.A., V.Ye. Matey, V.Ye. and Ye.S. Dal1. 19&3. Effect ot
calcium on mineral metabolism and gill ultrastructure in freshwater
fish. In; Reaktsiya gidrobiontov na zagryazneniye. Moscow, USSR.
pp. 199-^07.
Vinogradov, G.A., V.A. Sokolov. and G.I. Fierova. 1978. Investigation ol
mechanism of action of low pH on freshwater fish. In; Teoreticheskiye
aspekty rybokhozyaystvennykh issledovaniy vodokhranilishch (Theoretical
Aspects of Fishery Studies of Reservoirs). Leningrad, USSR. pp. 168-
173.
Vinogradov, G.A., V.A. Sokolov, and G.I. Fierova. 1976. Loss of sodium
ion ana changes in blood pH ot fish exposed to acid environment.
In; Tret'ya Vsesoyuznaya konterentsiya po ekologicheskoy tiziologii
ryb. Tezisy dokl (Summaries of Papers Delivered at 3d All-Union Con-
ference on Ecological Physiology of Fish) Part 1. Kiev, USSR. pp.
76-77.
foiebe, A., A. McGavock, A. Fuller, and h. Markus. 1934. The ability or
freshwater fish to extract oxygen at different hydrogen ion concen-
trations. Physiol. Zool. 7:435-448.
18
-------
AMMONIA PRODUCTION AND EXCRETION BY FISH
by
D.J. Randall1
P.A. Wright
ABSTRACT
Ammonia is continually produced and excreted by fish. High environmental
ammonia levels, however, will reduce excretion resulting in body
accumulations. In this paper, published research on the production,
utilization, distribution, and excretion of ammonia by fish is reviewed.
INTRODUCTION
Ammonia is an end product of protein metabolism and, if allowed to
accumulate in the body, has a toxic action. Ammonia must, therefore, either
be excreted or be converted to less toxic compounds such as urea or glutamine.
Ammonia is a substrate as well as a product of protein metabolism and in some
tissues it may be utilized rather than produced. In general, however, there
is a continual production and excretion of ammonia, or the less toxic
substance urea, by the whole animal. Elevated environmental ammonia levels
will reduce excretion and result in ammonia accumulation in the body of fish.
Although ammonia is almost completely ionized at body pH values, the
gill is much more permeable to unionized ammonia and it appears that passive
diffusion of NH-j is the major form of ammonia excretion across the gills of
freshwater fish. Ammonium ions may also be excreted in exchange for Na~*"
ions. The pH gradients across the gills will influence ammonia excretion,
which will vary according to blood and environmental water pH changes.
AMMONIA PRODUCTION AND UTILIZATION
Amino acids in excess of those required for protein synthesis are
converted to ammonia in the liver. Figure 1 illustrates the sources and fate
of ammonia in fish. Transaminases in the liver convert amino acids into
glutamate for subsequent conversion into ammonia (Forster and Goldstein, 1969;
Watts and Watts, 1974). Ammonia is also produced by the deamination of
adenylates in fish muscle (Driedzic and Hochachka, 1976). Enzymes involved in
ammonia production have been located in the kidney, gill, and muscle, as well
as in the liver; however, the major site of ammonia production is probably the
liver.
Ammonia toxicity can be ameliorated by the formation of less toxic
compounds, namely glutamine and urea. Levi et al. (1974) recorded high
^Dept. of Zoology, University of British Columbia, Vancouver, B.C. Canada
19
-------
levels of glutamine in the brain of goldfish and found that brain
levels increased with ambient ammonia concentrations. Webb and Brown _
found high glutamine synthetase activity in the teleost and elasmobranch brain
and this may be important in protecting the brain from sudden surges in
ammonia concentration. Walton and Cowey (1977) were able to detect
glutaminase activity in the gills of trout, but were unable to measure any in
vivo utilization of glutamine by the gills.
Ammonia can be converted, through carbamyl phosphate, to urea either via
purines (uricolysis) or via the ornithine cycle. The enzymes required for
uricolysis have been found in most fishes (Forster and Goldstein, 1969; Watts
and Watts, 1974) but Florkin and Duchateau (1943) were unable to detect any
activity of uricolytic enzymes in the cyclostome, Lampetra. The ratio of urea
production via the ornithine cycle and uricolysis is about 100 to 1 in
elasmobranchs and dipnoi, whereas in teleosts, most of the urea is formed via
uricolysis (Gregory, 1977).
EXCRETION
INTO
ENVIRONMENT-<-
INGESTED
PROTEIN
WATER
UREA
uricolysis
PURINES
CREATINE
PORPHYRINES
PYRIMIDINES
AMINES
•NHj
NH
FISH
ornithine cycle
AMINO ACID
POOL
• PROTEIN
CARBON SKELETON
Figure 1. Production and excretion of nitrogenous compounds in fish.
AMMONIA DISTRIBUTTON
Ammonia exists in aqueous solution either as ammonia gas or as ammonium
ion. Trussell (1972) and Thurston et al. (1979) present tables of the percent
unionized ammonia in solutions of different pH and temperature; the percent
unionized ammonia increases with increasing pH and temperature but decreases
with increases in ionic strength of the solution. The combined concentrations
of ammonia gas in solution (NH3) and ammonium ions (NH^"1") will be
referred to as total ammonia.
20
-------
Cameron and Heisler (1983) found that ammonia was slightly more soluble
in fish plasma than in water, and they also constructed a nomogram to describe
the effects of ionic strength and temperature on the pK of the ammonia/
ammonium reaction (see also Kormanik and Cameron, 1981; Boutilier et al.
1984). The pK is around 9.5, so at the pH of fish tissues, nearly all of the
ammonia will be as ammonium ion. The pH of the tissue will be an important
determinant of the total ammonia level. Ammonia gas diffuses at about the
same rate as CC>2 (Cameron and Heisler, 1983) so it will rapidly equilibrate
between different tissue compartments. Ammonium ion is not so permeable and
large differences in levels may occur between compartments; in mammals and
birds it has been shown that ammonium ion levels reflect the pH of the
compartment, tissues with a lower pH having higher concentrations. This
situation probably holds true for fish (Figure 2). Thus, the heart of a fish
may have a much higher total ammonia concentration because the pH of this
tissue is lower than that of the blood; this is due to elevated ammonium ion
concentrations in heart muscle, with the ammonia gas levels being in
equilibrium with the blood. In general very little is known about the
difference in total ammonia content in different tissues in fish.
10
1400 -i
1200 -
1000 -
800 -
600 -
400 -
200 -
[ZNH,]i 1 +
[Z NH3]2 1 + 1o'PKa
NH3 = 2 pM
(pKa pHl)
pH 7.0
WATER
7.8 6.9
BLOOD
7.3 6.7
HEART
Figure 2. pH dependence of NH3:NH4+ ratio in water, blood, and heart.
pH values taken from Heisler (1980) for normal and acidotic
conditions in fish.
21
-------
AMMONIA EXCRETION
The excretion of ammonia by fish is variable, depending on the state of
the animal, the environmental conditions, and the species. Ammonia excretion
tripled in sockeye salmon following daily feeding (Brett and Zala, 1975) but
remained low and unchanging during 22 days of starvation. In freshwater fish
ammonia excretion increases in response to exercise (Sukumaren and Kutty
1977- Holeton et al. 1983), long-term acid exposure (McDonald and Wood, 1981;
Ultsch et al., 1981), hypercapnia (Claiborne and Heisler, 1984) and NH4C1
infusion (Hillaby and Randall, 1979). In contrast, short-term exposure to
acid or alkaline water caused a decrease in ammonia excretion in trout (Wright
and Wood, 1984). It is not known if these changes in excretion reflect
changes in the rate of ammonia production or in the ammonia content of the
body. The ammonia content of fish is likely to be the equivalent of the
ammonia excreted in about 2 hours, most of the ammonia being in the tissues
with a lower pH, like muscle. Blood levels are around 0.200 to 0.300 mMol,
but muscle at a lower pH may have concentrations up to 1 mMol. Thus a kg fish
may contain about 0.500 to 0.700 mMol of ammonia and have an excretion rate of
about 0.300 mMol per hour. There is increased ammonia production in muscle
during exercise (Driedzic and Hochachka, 1976). Ammonia excretion by the
dogfish in seawater is unaffected by temperature change, exercise, hyperoxia,
hypercapnia, or the infusion of either HC1 or NaHC03 or anything that
induces acid-base stress (see Heisler, 1984, for review). This is surprising,
because many of these changes affect pH and therefore would be expected to
alter the ammonia content of body compartments and therefore ammonia
excretion.
There is an elevation in blood ammonia during starvation (Hillaby and
Randall, 1979; Morii, 1979), which is perhaps surprising because ammonia
excretion declines (Brett and Zala, 1975). Blood ammonia levels also rise
with increases in temperature (Fauconneau and Luquet, 1979) and with higher
ammonia concentrations in the water (Fromm and Gillette, 1968). Exposure of
fish either to air (Gordon, 1970) or to increased ammonia levels in water
(Fromm, 1970; Guerin-Ancey, 1976), raises blood ammonia levels and reduces
ammonia excretion and this is associated with a rise in urea production in
many, but not all fish. Unlike the above studies, Buckley et al. (1979) found
no change in blood total ammonia when coho salmon were exposed to elevated
ammonia levels in the environment. They did observe a significant rise in
plasma sodium, however, indicating some coupling between sodium uptake and
ammonia excretion (see below).
AMMONIA EXCRETION ACROSS GILLS
Most of the ammonia produced by the fish is excreted across the gills.
Oxygen, carbon dioxide, ions and water also are transferred across the gills.
The movement of ammonia gas is largely independent of the transfer of other
molecules, and is a function of the unionized ammonia gradient. Ammonia entry
into the fish depends on this gradient (Wuhrmann et al., 1947; Wuhrmanp and
Woker, 1948; Fromm and Gillette, 1968). The rate of loss also depends on the
NH^ gradient, in most instances (Hillaby and Randall, 1979; Kormanik and
Cameron 1981; Cameron and Heisler, 1983). The excretion of ammonium ion is
strongly coupled to the movement of other ions. Membranes, including the
gills, are not very permeable to cations like ammonium ion. Ammonium ion
22
-------
displace potassium in many membrane processes, for example in squid giant axon
(Binstock and Lecar, 1969), and this is the probable reason that elevated
ammonia causes convulsions in so many vertebrates. In fish gills it is
possible that NH^"1" can substitute for potassium in oubain-sensitive
sodium/potassium exchange and also substitute for protons in amiloride
sensitive Na /H exchange, the former moving ammonium from blood into the
gill epithelium, the latter exchanging ammonium for sodium on the outer
surface of the gill epithelium (Maetz and Garcia-Romeu, 1964; Evans, 1977;
Girard and Payan, 1980; Wright and Wood, 1984). Either acid conditions or
amiloride in the water inhibit Na+ influx across the gills and both these
conditions result in a reduction of ammonia excretion (Wright and Wood, 1984)
and ammonia infusion will stimulate sodium influx even in seawater fish
(Evans, 1977). It is interesting to note that ammonia excretion was
maintained, even in the face of a calculated reversed ammonia gas gradient,
presumably by sodium/ammonium exchange. Cameron and Heisler (1983) could
account for ammonia excretion in trout, under most conditions, by the
diffusion of unionized ammonia, but in the presence of high external ammonia,
sodium/ammonium exchange may counter balance the diffusive uptake of ammonia
gas. Indeed, this would explain the unchanging blood ammonia but increased
sodium levels in coho salmon exposed to elevated water ammonia (Buckley et al.
1979). Ammonium ion efflux and sodium influx must be quantified to determine
the exact relationship between sodium and ammonium movements. It is
relatively easy to determine sodium influx using isotopes but it is difficult
to determine NH^ efflux. Methods exist for measuring total ammonia
excretion from measurements of total ammonia in blood or water. If the pH is
known in both blood and water, then the ammonia and ammonium ion
concentrations on both sides of the gills can be calculated and the ammonia gas
and ammonium ion differences across the gills estimated. Wright and Wood
(1984) did this for rainbow trout exposed to a variety of acid and alkaline
conditions or following inhibition of sodium uptake with amiloride. They
observed a positive correlation between total ammonia excretion and the
ammonia gas gradient. If the ammonia gas excretion is subtracted from total
ammonia excretion, then there was an approximate relationship between sodium
influx and ammonium ion efflux.
The difficulty with this approach is that the exact pH of water and blood
on either side of the gill epithelium is not known. The pH of afferent and
efferent blood can be measured and the mean of these two may be an
approximation of the pH of blood in the gills. Water pH is equally difficult
to determine. Firstly there are undoubtedly boundary layers next to the gill
surface and this will contain concentration gradients; secondly there will be
longitudinal gradients as material is added to the gill water.
Substances added to the water that will affect pH are C02, ammonia and
ammonium ion, and protons. There is no evidence that C02 hydration occurs
at a catalysed rate in gill water; in as much as the water is only in contact
with the gill for about 100 to 400 msec. (Randall, 1982), and at low pH the
uncatalysed C02 reaction takes several seconds, any effect of C02 on water
pH will occur after the water has passed over the gills. Any excretion of
protons will have an immediate effect lowering pH, whereas excretion of
ammonia gas will raise pH. Conversely, excretion of ammonium ion will lower
pH and, because ammonia/ammonium reactions are rapid, they will influence
water pH at the gills and therefore ammonia excretion. One might predict
that, under most conditions, ammonia excretion will exceed both ammonium ion
23
-------
and proton excretion, thus the effect will be to raise the pH of the water
in contact with the gills. Although the amount of C02 excreted will far
exceed that of ammonia, there will not be an appreciable reduction in
water pH until the water has left the gills. It is clear a more detailed
understanding of ammonia excretion requires a more detailed analysis of pH
gradients across the gills.
Wright and Wood (1984) measured ammonia excretion in trout exposed
to a variety of water pH conditions, and they found that sodium influx was
zero at pH 4.1. If it is assumed that under these conditions ammonium/
sodium exchange is also zero, then all ammonia excreted must have been
due to unionized ammonia diffusion. The unionized ammonia gradient was
calculated from measurements of pH and total ammonia content in water and
blood. The calculated permeation coefficient for ammonia diffusion across
the trout gill was 65% of that calculated by Cameron and Heisler (1983).
The unionized ammonia excretion was then calculated for trout under other
conditions using this permeation coefficient and the estimated NH3 gradi-
ents. The ammonium ion excretion was then determined by subtracting the
NH3 excretion from the measured total ammonia excretion.
A plot of ammonium ion excretion against sodium influx for trout
under a variety ot conditions is shown in Figure 3a. Considering only
control and acid exposed fish there is a close 1 to 1 correlation between
sodium influx and ammonium excretion (Figure 3b), indicating a tight
coupling of sodium and ammonium exchange. Under alkaline conditions,
however, the coupling was not obvious although clearly sodium influx was
reduced. If it is assumed that under all conditions there is a 1 to 1
relationship between sodium and ammonium ion flux (Figure 3c) then under
alkaline conditions there must have been an underestimate of the unionized
ammonia excretion (Figure 3a).
The estimation of unionized ammonia excretion depended on the
assumption that pH in the bulk medium was the same as that at the surface
of the gills. If, under alkaline conditions, there is still a tight
coupling of ammonium ion and sodium exchange—that is, ammonium efflux
equals sodium influx (see Figure 3c), we can estimate unionized ammonia
excretion by subtracting ammonium ion excretion from total ammonia excre-
tion. Table 1 lists the calculated NH3 and NH4+ movements across the
gills, note that in some cases there is an excretion of ammonium ion but
an uptake of ammonia by the fish. We have determined the effect of
ammonia transfer on water pH at the gills from these fluxes, the buffer-
ing capacity of the water, and the equilibrium constants for the NH3/NH4+
reaction (Thurston et al., 1979).
The effects of ammonia movements on water pH (Table 1) are small,
especially at water pH between 6.6 and 8.1. At water pH of 8.7 and 9.5,
however, ammonia transfer has some effect on the pH of water as it passes
over the gills. The NH^ gradient, and therefore NH3 excretion in the
study of Wright and Wood (1984), was based on the pH of the bulk phase
water rather than the pH of water at the gill surface. The fact that the
pH of water at the gill surface was affected by ammonia transfer at these
high water pH levels, may have led to an error in NH3 excretion and
24
-------
-BOO-i
Tj -600-
1
-«00_
i!
-200-
,pH=8.7
• pH=9.5
I Control
pH=8.1
«pH=6.6
Amilorlde
pH=8.1
pH=«.1
0 •
-800-1 B
4-200
+100
1—
+600
+800
Ll -600-
u>
jf
-«00-
-200—
, pH 6.6
pH 4.1
4-200
+400
+600
Figure 3. Relationship of Na+ influx to NH4+ efflux, and Na+ influx
to pH. Each dot represents mean of at least 8 fish.
25
-------
TABLE 1. ANALYSIS OF AMMONIA EXCRETION IN TERMS OF Na+/HN4+ EXCHANGE AND
WATER pH NEXT TO THE GILL.
Water
Acid
pH 4.1
1
1 Tout
M J
amm NH~
280.99 280.99
2
out
NH4+
0
APH3
.06
Moderate
acid 403.09 118.20 284.89 .02
pH 6.6
Control
pH b.l 393.58 -179.73 573.31 -.03
Amiloride
pH 8.1 232.00 193.04 38.96 +.03
Moderate
alkaline
pH 8.7 261.35 -231.00 492.35 -.10
Severe
alkaline
pH 9.5 67.19 -155.22 222.7b -.10
1. measured in umol.kg"1.h"1
2. measured in uequiv.kg~l.h~l
3. ApH = pH calculated -pH measured
therefore an overestimate of ammonium ion excretion. As a result, Wright
and Wood (1984) did not observe a one to one relationship between ammonium
ion and sodium flux at these elevated water pH levels (Figure 3a) .
Carbon dioxide in the water affects ammonia toxicity; if C0£ levels
are raised, total ammonia toxicity is decreased (Alabaster and Herbert,
1954). C02 causes a fall in pH and decreases the proportion of unionized
ammonia in solution. The unionized form has a greater toxic effect be-
cause ammonia must enter the fish to exert its toxic action and lipid mem-
branes are much more permeable to unionized ammonia (Wuhrmann et al.,
1947- Wuhrmann and Woker, 1948; Thurston et al . , 1981). Thus the reduction
in unionized ammonia associated with the fall in water pH caused by the
rise in C0£ decreases total ammonia toxicity. Lloyd and Herbert (I960)
found, however, that although total ammonia toxicity was reduced at high
C02 levels, the inverse was true when considering unionized ammonia alone.
More unionized ammonia is required in low C02~high pH water to exert the
26
-------
same toxic effect as seen in fish in high C02~low pH water. The explanation
presented by Lloyd and Herbert (1960) for the decreased toxicity of union-
ized ammonia in low CC>2 water was that C02 excretion across the gills would
reduce pH and therefore the concentration of unionized ammonia in water
flowing over the gills. This is consistent with our conclusions of the
effects of C02 at high pH, but not at water pH levels of below 8.10.
Another possible explanation is that the blood pH of the fish also varied
inversely with the C02 content of the water such that the total ammonia
content of the blood decreases with the C02 in water pH in Lloyd and
Herbert's experiments, and both these factors are known to reduce blood
pH in fish (Janssen and Randall, 1975; Randall et al., 1976; Heisler,
1980). The tish in Lloyd and Herbert's experiments were exposed to "water
of different pH and C02 levels for 18 hours and the blood pH of these fish
was probably inversely related to C02 levels in the water at the time of
ammonia exposure. Blood pH will be an important determinant of blood total
ammonia levels and this in turn is an important factor in its toxic action
(.Hillaby and Randall, 1979). Blood pH is probably decreased with increas-
ing CC>2 levels in the water, and this causes an increase in the blood total
ammonia levels for a given unionized ammonia concentration. This could
account for the differences in unionized ammonia toxicity observed. What
is required is accurate measurement of pH in both blood and water and
therefore NH3 gradients across the gill epithelium.
REFERENCES
Alabaster, J.S., and D.W. M. Herbert. 1954. Influence of carbon dioxide on
the toxicity of ammonia. Nature 174:404.
Binstock, L., and H. Lecar. 1969. Ammonium ion currents in the squid giant
axon. J. Gen. Physiol. 53. p. 342-361.
Boutilier, R.G., T.A. Heming, and G.K. Iwama. 1984. Appendix: Physiochemical
parameters for use in Fish Respiratory Physiology- Appendix: Fish
Physiology, Vol XA ed. W.S. Hoar and D.J. Randall. Academic Press Inc.
Brett, J.R., and C.A. Zala. 1975. Daily pattern of nitrogen excretion and
oxygen consumption of sockeye salmon (Oncorhynchus nerka) under controlled
conditions. J. Fish. Res. Biol. Can. 32:2479-2486.
Buckley, J.A., C.M. Whitmore., and B.D. Liming. 1979. Effects of prolonged
exposure to ammonia on the blood and liver glycogen of coho salmon
(Oncorhynchus kisutch). Comp. Biochem. Physiol. 63C:297-303.
Cameron, J.N., and N. Heisler. 1983. Studies of ammonia in the rainbow
trout: physio-chemical parameters, acid-base behavior, and respiratory
clearance. J. Exp. Biol. 105:107-125.
Claiborne, J.B., and N. Heisler. 1984. Acid-base regulation in the carp
(Cyprinus carpio) during and after exposure to environmental hypercapnia.
J. Exp. Biol. 108:25-43.
Dr'iedzic, W.R., and P.W. Hochachka. 1976. Control of energy metabolism In
fish white muscle. Am. J. Physiol. 230:579-582.
27
-------
Evans, D.H. 1977. Further evidence for Na/NH4 exchange in marine teleost
fish. J. exp. Biol. 70:213-220.
Fauconneau, B., and P. Luquet. 1979. Influence d'une elevation de
temperature sur 1'evolution de 1'aminoacidemie et de 1'ammoniemie apres le
repas chez la truite arc en ciel (Salmo gairdneri R.). Ann. Biol. anim.
Bioch. Biophys. 19(4A):1063-1079 .
Florkin, M., and G. Duchateau. 1943. Les formes du systeme enzymatique de
1'uricalyse et 1'evolution du catabolisme purique chez les animaux. Arch.
Int. Physiol. 53:267-307.
Forster, R.P., and L. Goldstein. 1969. Formation of excretory products. In:
Fish Physiology. Eds. W.S. Hoar and D.J. Randall, Academic Press, NY.
Vol I., pp. 313-350.
Fromm, P.O., and J.R. Gillette. 1968. Effect of ambient ammonia on blood
ammonia and nitrogen excretion of rainbow trout (Salmo gairdneri). Comp.
Biochem. Physiol. 26:887-896.
Fromm, P.O. 1970. Section III. Effect of ammonia on trout and goldfish. In:
Toxic action of water soluble pollutants on freshwater fish. EPA Water
Pollution Control Research Series. 18050 DST 12/70 pp. 9-22.
Girard, J.P., and P. Payan. 1980. Ion exchanges through respiratory and
chloride cells in freshwater and seawater adapted teleosteans. Am. J.
Physiol. 238: (Regulatory Integrative Comp. Physiol. 7) R260-R268.
Gordon, M.S. 1970. Patterns of nitrogen excretion in amphibious fishes.
Urea Kidney, Proc. Int. Colloq. 1968:238-242.
Gregory, R.B. 1977. Synthesis and total excretion of waste nitrogen by fish
of the Periophthalmus (mudskipper) and Scartelass families. Comp.
Biochem. Physiol. 57A:33-36.
Guerin-Ancey, 0. 1976. Etude experimentale de I1excretion azotee du bar
(Dicentrarchus labrax) en cours de croissance. II. Effets du jeune sur
1'excretion d'ammonia det d'uree. (Experimental study of the nitrogen
excretion of bass (Dicentrarchus labrax) during growth. II. Effects of
starvation on the excretion of ammonia and urea). Aquaculture
9(2):187-194.
Heisler, N. 1980. Regulation of the acid-base status in fishes. In:
Environmental physiology of fishes. Ed. M.A. Ali, Plenum Publishing Co.,
NY.
Heisler, N. 1984. Acid-base regulation in fishes. In: Fish Physiology Vol.
XA, ed. W.S. Hoar and D.J. Randall. Academic Press Inc., NY.
Hillaby, B.A., and D.J. Randall. 1979. Acute ammonia toxicity and ammonia
excretion in rainbow trout (Salmo gairdneri). J. Fish. Res. Biol. Canada
36:(6)621-629.
28
-------
Holeton, G.F., P. Neumann, and N. Heisler. 1983. Branchial ion exchange and
acid-base regulation after strenuous exercise in rainbow trout. Resp.
Physiol. 51,303-318.
Janssen, R.G., and D.J. Randall. 1975. The effects of changes in pH and
PCQO in blood and water on breathing in rainbow trout, Salmo
gairdneri. Resp. Physiol. 25:235-245.
Kormanik, G.A., and J.N. Cameron. 1981. Ammonia excretion in the FW catfish:
the role of diffusion. Amer. Soc. Zool. 21(4):1042.
Levi, G., G. Morisi, A. Coletti, and R. Catanzaro. 1974. Free amino acids in
fish brain: normal levels and changes upon exposure to high ammonia
concentrations in vivo, and upon incubation of brain slices. Comp.
Biochem. PhysioTT 49A:623-636.
Lloyd, R. , and D.W.M. Herbert. 1960. The influence of carbon dioxide on the
toxicity of un-ionized ammonia to rainbow trout (SaImo gairdneri
Richardson). Ann. appl. Biol. 48:399-404.
Maetz J., and F. Garcia-Romeu. 1964. The mechanism of sodium and chloride
uptake by the gills of a freshwater fish, Carassius auratus II Evidence
for NH4+/Na+ and HC03~/C1~ exchanges. J. Gen. Physiol.
47:1209-1227.
McDonald, D.G., and C.M. Wood. 1981. Branchial and renal acid and ion fluxes
in the rainbow trout at low environmental pH. J. Exp. Biol. 93:101-118.
Morii, H. 1979. Changes with time of ammonia and unea concentrations in the
blood and tissue of mudskipper fish, Periophthalmus cantonensis and
Boleophthalmus pectinirostris, kept in water and on land. Comp., Biochem.
Physiol. 64A:235-243.
Randall, D.J. 1982. The control of respiration and circulation in fish
during exercise and hypoxia. J. Exp. Biol. 100:275-288.
Randall, D.J., N. Heisler and F. Drees. 1976. Ventilatory response to
hypercapnia in the larger spotted dogfish Scyliorhinus stellaris. Am. J.
Physiol. 230:590-594.
Sukumaran, N., and M.N. Kutty. 1977. Oxygen consumption and ammonia
excretion in the catfish Mystus armatus, with special reference to
swimming' speed and ambient oxygen. Proc. Indian. Acad. Sci. 86B:195-206.
Thurston, R.V., R.C. Russo, and K. Emerson. 1979. Aqueous ammonia
equilibrium-tabulation of percent un-ionized ammonia. U.S. Environmental
Protection Agency, Duluth, Minnesota. EPA-600/3-79-091.
Thurston, R.V., R.C. Russo, and G.A. Vinogradov. 1981. Ammonia toxicity to
fishes: effect of pH on the toxicity of the un-ionized ammonia species.
Environ. Sci. Technol. 15(7): 837-840.
29
-------
Trussell, R.P- 1972. The percent unionized ammonia in aqueous ammonia
solutions at diffeent pH levels and temperatures. J. Fish. Res. Board
Can. 29(10):1505-1510.
Ultsch, G.R., B.C. Jackson, and R. Moalli. 1981. Metabolic oxygen conformity
among lower vertebrates: the toadfish revisited. J. Comp. Physiol.
142:439-443.
Walton, M.J., and C.B. Cowey. 1977. Aspects of ammoniogenesis in rainbow
trout, Salmo gairdneri. Comp. Biochem. Physiol. 576:143-149.
Watts, R.L., and D.C. Watts. 1974. Nitrogen metabolism in fishes. In:
Chemical zoology. Eds. M. Florkin and B.T. Scheer, Vol. VIII, Academic
Press, NY. pp. 369-446.
Webb, J.T., and G.W. Brown, Jr. 1976. Some properties and occurrence of
glutamine synthetase in fish. Comp. Biochem. Physiol. 548:171-175.
Wright, P.A., and C.H. Wood. 1984. An analysis of branchial ammonia
excretion in the freshwater rainbow trout: effects of environmental pH
change and sodium uptake blockade. J. Exp. Biol. (in press).
Wuhrmann, K., and Woker, H. 1948. Contributions to the toxicology of fishes.
II Experimental investigations on ammonia - and hydrocyanic acid
poisoning. Translation. Schweiz. Z. Hydrol. 11:210-244.
Wuhrmann, K., F. Zehender, and H. Woker. 1947. Biological significance for
fisheries of ammonium-ion and ammonia content of flowing bodies of water.
Translation of: "Uber die fischereibiologische Bedeutung des Ammonium-und
Ammoniakgehaltes fliessender Gewasser". Vierteljahrsschrift der Naturf.
Gesellschaft in Zurich 92:198-204.
30
-------
THE USE OF FISH BEHAVIOR IN COMPARING TOXIC EFFECTS OF THREE CHEMICALS
by
M.G. Henry1
ABSTRACT
The behavioral effects of three chemicals are compared using a method
developed to monitor changes in social groups of bluegill, Lepomis macrochirus.
Ten behaviors were observed once daily for 96-hr before and after treatment.
Methyl parathion induced hyperactivity followed by changes in frequency of
comfort movements. Copper sulfate disrupted respiration, but the most strik-
ing impact was an increase in aggression caused by chlordane. The distribu-
tion of toxicant-related responses was related to social rank in each of the
treated hierarchies. The dominant and subordinate fish were most affected.
The best indicator of toxicant effect across rank was respiratory in nature,
because coughs increased in all fish exposed to each of the three compounds.
This assay was sensitive enough to detect, behavioral alterations at concentra-
tions well below lethal levels. Behavioral bioassays, if selected to relate
to important life history characteristics of the species of interest, could
provide useful ecological links for interpreting laboratory effects and veri-
fying them in the field.
INTRODUCTION
Assessing the toxicity associated with environmental contaminants is
chiefly a biological problem because their impact on aquatic organisms is at
the heart of the hazard evaluation process. Despite good intentions and
sound hypotheses, however, the reality persists that our ability to relate
laboratory effects to field effects is very limited. Classical toxicity tests
examining mortality, changes in growth, and effects on reproduction have used
single species and single compounds under sterile test conditions. The pur-
pose they serve is that these tests produce toxicity values that are compar-
able in nature due to the use of standardized test methodology. Despite this
-"-Columbia National Fisheries Research Laboratory, Rt. 1, Columbia, MO 65201
Present address: Great Lakes Fishery Laboratory, 1451 Green Road, Ann
Arbor, MI 48105
31
-------
advantage, the issue of ecological significance remains. Ecological aspects
of an organism's life history, such as its behavior, need to be incorporated
into the development and standardization of testing protocols.
Behavior is considered the organismal level manifestation of the physi-
ologically and environmentally influenced state of the animal. Behavior
mediates survival and reproduction by influencing predator-prey interactions,
feeding efficiency, swimming performance, courtship, nest digging, etc. If
key behaviors are altered by a contaminant, then ultimately survival and re-
production could decline. Measurement of behavioral parameters can assist us
in interpreting toxicant effects within an ecological framework. In addition,
data from behavioral bioassays may aid us in designing more focused field
studies so that verification of laboratory effects is possible.
Keeping in mind the issues of ecological significance, cost effective-
ness, standardized methodology, and applicability over a range of compound
classes, we developed this test. Bluegill (Lepomis macrochirus) were used
because they are ubiquitous throughout North America, have been used in
classical toxicity tests, and are behaviorally diverse. The objectives of
this research were to compare the behavioral effects of three chemically
distinct compounds and evaluate the overall utility of this test method.
MATERIALS AND METHODS
Ten different behaviors were monitored in each established population
consisting of five adult bluegill. Each behavior related either to body
maintenance or social interaction. Daily observations (0.5 hr/tank) were
made directly and the frequencies of each behavior were recorded by hand.
Individual fish were recognized on the basis of s-ize, color and natural
markings. No tags, brands or fin clips were utilized in the event that they
would interfere with normal behavior patterns.
Fish were randomly assigned to four 295L aquaria, equipped with plants
(Elodea) and gravel to simulate some of the characteristics of a natural
habitat. Tanks were monitored for 96-hr before and after addition of the
toxicant. Each flow-through system utilized a modified Mount and Brungs (1967)
proportional diluter to deliver a different toxicant concentration to each
test tank. The lowest concentration selected for each toxicant examined was
based on a level (if available in the literature) that approximated the low
end of the Maximum Allowable Toxicant Concentration (MATC). The middle con-
centration approximated the upper end of the MATC and the high concentration
approached the reported Lethal Concentration (LC50). If a MATC was not avail-
able, a geometric progression from the LC50 was used instead. A well water
control was incorporated to monitor changes through time. The purpose of
selecting levels based on MATCs and LCSOs was to facilitate comparison of the
sensitivity of this new behavioral technique to endpoints obtained through
more classically accepted protocols.
Three separate experiments were conducted, each including two complete
replications. The three experiments were based on the behavioral evaluation
of three chemically distinct compounds: methyl parathion - an organophosphate
insecticide, copper sulfate - a heavy metal, and chlordane - an organochlorine
32
-------
insecticide. Water concentrations were determined using gas-liquid chroma-
tography or atomic absorption spectrophotometry.
Because the number of chemicals in the United States requiring screen-
ing is so vast, standardization of new methods is crucial so that industry,
universities and government agencies can utilize them. In over 40 replica-
tions of this behavioral bioassay, methods standardization was possible so
that the whole procedure could be completed in 17 to 21 days. This time
period is considerably less than that required in partial-chronic tests,
thereby increasing cost efficiency without sacrificing sensitivity. A more
detailed methods description and results/discussion can be found iri papers
by Henry and Atchison (1979a, 1979b, 1984, in review).
For comparative purposes, four behaviors affected by the middle concen-
tration of each chemical have been selected for discussion. Two behaviors
relate to respiratory processes and two relate to social interaction, spe-
cifically aggressive behaviors used in hierarchy establishment and maintenance.
The two respiratory behaviors that were examined were coughs and yawns.
Coughs (Henry and Atchison 1979a, 1979b, 1984) were identified as the rapid,
repeated opening and closing of the mouth and opercular coverings, accompa-
nied by partial extension of the paired fins. A yawn was conversely recog-
nized as a singular event typified by maximal opening of the mouth and oper-
cular covering accompanied by hyperextension of all fins, both paired and
medial.
The two socially induced behaviors we will discuss are nips (bites) and
nudges (contact between fish, the aggressor touching the recipient with a
closed mouth). Both of these behaviors are aggressive in nature.
Data were transformed (4x+1) and analyzed using analysis of .variance
and least significant difference determinations (Snedecor and Cochran 1967).
RESULTS AND DISCUSSION
The postures and forms of behaviors monitored did not change in the
presence of the toxicants used; however, frequency- and hierarchy-related
distribution throughout the test population was altered.
In general, hyperactivity was noticeable in all fish exposed to methyl
parathion. Two comfort movements (s-jerks and fin flicks) were most altered
in frequency, increasing .significantly (P=0.001) over control levels once
methyl parathion was introduced (Henry and Atchison 1984). Coughs also in-
creased significantly (P=0.01) in the presence of the toxicant (Table 1).
Copper sulfate did not induce hyperactivity but dramatically influenced res-
piratory disruptions, increasing the frequency of coughs (P=0.01) and yawns
(P=0.01) well beyond levels observed during the pre-exposure period. Chlor-
dane produced large elevations (P=0.01) in aggressive behaviors. The more
intense manifestations, nips, became so frequent that the most subordinate
fish was killed by the dominant in the 0.044 mg/L treatment.
33
-------
TABLE 1 MEAN FREQUENCIES OF FOUR BEHAVIORS EXAMINED BEFORE AND AFTER
TABLE i. M^AW y ADDITION OF THREE CONTAMINANTS
Behavior _^_______
Toxicant T**&"^^^^
(middle concentration)
Tol rng/Lf ^ 19-0/29.5 7.C0;8.5 39.COA25 8.0A.5
Copper sulfate 9.0/72.0 6.0/31.0 31.0/71.5 4.0/9.0
(0.057 mg/L)
chlordane 13.5/48.0 7.0/41.0 31.0/78.5 6.5/1.0
(0.002 mg/L) ^
aC:T represents mean control values compared to treatment values.
Differential response associated with social rank occurred at every
concentration for each toxicant examined. The dominant and the most subordi-
nate fish were usually affected to the greatest extent. Because fish of
those two ranks are initially under the greatest amount of social stress
(Sparks et al. 1972, Noakes and Leatherland 1977), the combination of social
stress and toxicant stress produced a more extreme response than was seen in
fish of intermediate rank. The dominant and subordinate fish of each experi-
ment are compared in Table 2 before and after toxicant addition. Subordinate
fish suffered more respiratory disruptions than dominants; however, aggres-
sive behaviors were significantly altered in top ranking individuals. Fish
were affected differently, therefore, depending on hierarchy position but all
responded to the presence of the toxicants. Coughs, although most altered in
subordinates, universally increased in fish of every rank exposed to each of
the three contaminants. If one behavior had to be selected for cursory ex-
amination across ranks, cough frequency would be the best overall indication
of effect.
We know that the toxic.mode of action for methyl parathion is inhibi-
tion of cholinesterase at synaptic and neuromuscular junctions. Severe inhi-
bition of this enzyme often results in induction of hyperactivity and muscular
spasms (Brown 1978). As was previously mentioned, hyperactivity was observed
in fish exposed to methyl parathion. Copper, on the other hand, has been
associated with causing changes in gill structure (Anderson and Spear 1980).
Increases in coughs and yawns relate to this suspected underlying change in
physiology and morphology. Chlordane also disrupts respiration. However, it
seems to do so by creating anaerobic conditions in gill tissue via inhibition
of acetylcholinesterase (Verma et al. 1981). Thus it is both a neurotoxin
and respiratory disrupter. Coughs and yawns increased in fish exposed to
chlordane but aggression did as well.
34
-------
These underlying physiological mechanisms associated with the induction
of toxic responses can be observed at an organismal level in the altered fre-
quency of behaviors described above. Thus, this technique appears to be not
only sensitive but applicable across different classes of chemicals. It re-
quires minimal equipment, can be conducted in 17 to 21 days, and is ecologi-
cally related to maintenance of social groups of bluegill.
This test is, however, only appropriate for examining species that form
hierarchies. It is not appropriate for solitary predators, schooling fishes,
etc. Consequently, use and standardization of other behavioral tests such as
predator-prey, swimming performance or avoidance tests could be utilized
depending on the life history characteristics of the species of interest.
For example, swimming performance should be evaluated when anadromous fish
are exposed to a toxicant. Sufficient swimming stamina allows them to reach
nursery spawning areas and if altered could reduce reproduction.
Behavioral bioassays, used in conjunction with other approaches such as
acute, chronic, and partial chronic tests, residue dynamic studies, and phy-
siological or biochemical assays will help us more realistically evaluate
chemicals and set ecologically meaningful water quality standards.
REFERENCES
Anderson, P. D., and P. A. Spear. 1980. Copper pharmaco-kinetics in fish
gills-II. Body size relationships for accumulation and tolerance.
Water Research 14(8):1107-1111.
TABLE 2. MEAN FREQUENCIES OF FOUR BEHAVIORS PERFORMED BY MOST DOMINANT AND
SUBORDINATE RANKED FISH EXPOSED TO THREE TOXICANTS
Behavior
Toxicant
Methyl parathion
(0.03 mg/L)
Control
Treated
Cough
D:S
0.5/8.0
4.5/16.0
Yawn
D:S
0.0/5.0
1.0/3.5
Nip
D:S
14.5/0.0
4.5/0.0
Nudge
D:S
4.0/0.0
0.5/0.0
Copper sulfate
(0.057 mg/L)
Control 3.5/3.0 0.0/2.5 16.5/0.5 0.5/1.0
Treated 14.5/30.5 7.0/12.0 52.5/1.5 1.0/0.0
Chlordane
(0.002 mg/L)
Control 2.0/4.5 1.0/2.0 19.0/0.0 0.25/0.5
Treated 14.0/22.0 10.5/19.5 62.5/0.0 0.25/0.25
35
-------
Brown, A. W. A. 1978. Ecology of pesticides. John Wiley and Sons, New York,
New York.
Henry, M. G., and G. J. Atchison. 1979a. Behavioral changes in bluegill
(Lepomis macrochirus) as indicators of sublethal effects of metals.
Environmental Biology of Fishes 4:37-42.
Henry, M. G., and G. J. Atchison. 1979b. Influence of social rank on the
behavior of bluegill Lepomis macrochirus Rafinesque, exposed to sub-
lethal concentrations of cadmium and zinc. Journal of Fish Biology
15:309-315.
Henry, M. G., and G. J. Atchison. 1984. Behavioral effects of methyl para-
thion on social groups of bluegill (Lepomis macrochirus). Environmental
Toxicology and Chemistry 3:399-408.
Henry, M. G. , and G. J. Atchison. In review. The effects of copper on the
behavior of bluegill, Lepomis macrochirus. Transactions of the American
Fisheries Society.
Noakes, D. L. G., and J. F. Leatherland. 1977. Social dominance and inter-
renal cell activity in rainbow trout, Salmo gairdneri. Environmental
Biology of Fishes 2:131-136.
Mount, D. I., and W. A. Brungs. 1967. A simplified dosing apparatus for
fish toxicology studies. Water Research 1:21—29.
Snedecor, G. W., and W. G. Cochran. 1967. Statistical methods. Iowa State
University Press, Ames, Iowa.
Sparks, R. E., W. T. Waller, and J. Cairns, Jr. 1972. Effects of shelters
on the resistance of dominant and submissive bluegills (Lepomis macro-
chirus) to a lethal concentration of zinc. Journal of the Fisheries
Research Board of Canada 29:1356-1358.
Verma, S. R., I. P. Tonk, A. K. Gupta, and R. C. Dalela. 1981. In-vivo
enzymatic alterations in certain tissues of Saccobranchus fossilis
following exposure to four toxic substances. Environmental Pollution
Series A: Ecological Biology 26(2):121-128.
36
-------
RESISTANCE OF AQUATIC ANIMALS TO ORGANOPHOSPHORUS
PESTICIDES AND ITS MECHANISMS
by
V.I. Kozlovskaya, G.M. Chuyko, L.N. Lapkina
and V.A. Nepomnyashchikh.1
ABSTRACT
Aquatic animals possess varying resistance to organophosphorus pesti-
cides. Fish are more resistant than invertebrates. The toxicity of organ-
ophosphorus pesticides is caused by their ability to inhibit cholinester-
ase, primarily acetylchoiinesterase of the nervous system. This leads to
a disruption ot the nervous system. The molecular forms ot acetylchoiines-
terase from various animal species differ in terms of sensitivity to
organophosphorus compounds. A dependence between the sensitivity of an
enzyme to organophosphorus pesticides and the toxicity of preparations
for an organism is apparent between invertebrates and tish, among species
close in taxonomic position as well as in individuals in the same species
for various toxic substances. The resistance of aquatic animals, like
ground insects and mammals, depends upon the sensitivity of their choiin-
esterase. But this is not the only mechanism of resistance. In the
resistance of animals to organophosphorus pesticides, a definite role is
also played by the processes of metabolism and permeability.
INTRODUCTION AND DISCUSSION
Water pollution by pesticides, including organophosphorus compounds,
has an adverse eftect on hydrobionts. To effectively monitor these agents
in water and.forecast their eftect on fauna, it is necessary to determine
the resistance ot different aquatic animals to pesticides and investigate
its mechanisms.
Aquatic animals differ in resistance to this group of compounds. Thus,
among the 12 species ot leeches referable to 2 orders and 4 tamilies, there
are species with high (Protoclepsis tessulata, Hemiclepsis marginata) and
•'•Institute of Biology of Inland Waters, USSR Academy of Sciences, Borok, USSR
37
-------
low (.Caspiobdella fadejewi, Herpobdella nigricollis) resistance (Lopkina
and Flerov 1979). Among crustaceans and mollusks, the water flea (Daphnia
pulex), hog slater (Asellus aquaticus) and large snail (Limnaea stagnalis)
are the least resistant. The trumpet snail (Planorbis corneus) dies at
high concentrations, the same as fish including crucian carp (Carassius
carassius), roach (Rutilus rutilus) and zope (Abramic ballerus). The carp
ICyprinuT carpio) is the most resistant species. Chlorofos toxicity is
100,000 times higher tor the water tlea than the carp. There are also fish
species with low resistance. They are representatives of the Salmonidae (£.
irrideus) and Percidae (P. tluviatilis) families (Table 1 and 2). Our data
on toxicity of organophosphorus pesticides for aquatic animals are essenti-
ally in agreement with the data in surveys IHogan and Knowles 19bb, Chuyko
et al. 19b3j prepared on the basis of results of American studies.
Organophosphorus compounds have a dissimilar effect on different
stages of the life cycle of aquatic animals. In leeches, newly hatched
young specimens that have not begun independent feeding are the least
resistant. (4b-h LC5 are 0.003 and 0.03 mg/& for Caspiobdella fadejevi
and Piscicola geometra, respectively. Leech cocoons are the most resis-
tant to toxicants. Chlorofos in concentrations of 0.5 and 1 mg/& , with
4b-h exposure, which has a devastating effect on adult specimens, is
virtually harmless for cocoons. Resistance of eggs in cocoons is at a
maximum at the early stages of embryonic development and it diminishes by
the end of development (Table 3) (Lapkina 1983).
For fish, organophosphorus pesticides are the most hazardous during
the embryonic (from the time of fertilization to cleavage and at organo-
genesis stage), larval, and young fry periods of development. In concen-
trations of 0.01-0.001 mg/Jl, metaphos, phosalone, methylnitrophos, ioso-
phos, and sayfos impair embryogenesis and cause hatching of malformed
larvae. At the prelarval stage, exposure to a weaker toxic agent does
not reveal visible disturbances; however, by the larval stage there are
pathological deviations of development. One of the distinctions of effect
of organophosphorus compounds on embryogenesis is slower embryonic devel-
opment, which is associated with loss of a large number of embryos.
TABLE 1. TOXICITY OF ORGANOPHOSPHORUS PESTICIDES FOR AQUATIC INVERTEBRATES,
4b-h EXPOSURE AND TEMPERATURE OF lb-21°C, mg/£
Stage, Maximum tolerated
size, concentration
Pesticides and Organisms mm (MTC) LCso
Chlorofos
Leeches
Protoclepsis tessulata 15 5 105 300
Hemiclepsis marginata 15-22 20 100 300
38
-------
TABLE i. (Continued)
Pesticides and Organisms
Glossiphonia complanata
Helobdelia stagnalis
Caspiobdella fadejevi
Piscicola geometra
Hiruao medicinalis
haemopis sanguisuga
Herpobdella octoculata
H. testacea
H. nigricollis
H. (Dina) lineata
Crustaceans
Strep tocephalus
torvicornis
Daphnia pulex
Asellus aquaticus
Mollusks
Limnaea stagnalis
Planorbis corneus
Carbotos
Leeches
Hirudo medicinalis
Herpobdella octoculata
H. nigricollis
Stage,
size,
mm
15-25
5-10
15-25
20-35
eO
70-90
30-45
30-45
25-35
20-40
Adult
Adult
Adult
Adult
Adult
00-80
30-45
20-30
Maximum tolerated
concentration
(MTC)
15
3
0.003
0.01
0.05
2.5
0.5
0.1
0.03
1
0.00007
0.00515
1
fa. 5
4.5
1
80
50
0.07
0.8
0.3
10
1.5
O.fa
0.2
3.5
0.04
0.00028
0.4
0.5
50
11
b.5
4.7
300
250
0.5
1.5
0.6
25
1.5
1
0.8
8
0.00141
51.5
250
14
8
7
39
-------
TABLE
(Continued)
Pesticides and Organisms
Stage, Maximum tolerated
size, concentration
mm (.MTC)
LC
51L
LC"in
Crustaceans
Daphnia pulex Adult
Asellus aquaticus Adult
Moilusks
Limnaea stagnalis Adult
Planorbis corneus Adult
Rogor
Leeches
Hlrudo medicinalis 60-bO
tierpobdella octoculata 30-45
H. nigricollis 20-30
O.OUU1
0.011
11.2
72.9
2
5
3
0.0013 0.0177
0.27 b
36.7
14b.O
2b
183.2
359.7
40
100
100
Roe had the lowest resistance from the time or fertilization to cleav-
age and at the organogenesis stage. Larvae on exogenous nutrition and fry
at early stages of development also have low resistance (.Guseva 1980).
Among larvae and fry, the younger specimens perish taster (.Table 4). For
try, 9b-h LC^u of most agents constitutes 0.1-iO mg/Jl (.Johnson and Finley
1964, Post and Schroeder 1971, Grischenko et al. 1975, Prokopenko et al.
1975).
Current conceptions ol the mechanisms of animal resistance to organo-
phosphorus compounds are based on results obtained primarily for mammals
and terrestrial anthropods (O'Brian 1971, Rozengart and Shestobitov 1978).
It was established that resistance involved many factors, among which we
can single out three basic mechanisms: difference in rate of penetration
ot toxicant into the body, metabolism of compounds in the body, and sensi-
tivity of cholinesterases, which are the "targeted" enzymes. The last is
the most important, because affinity ot the enzyme tor a toxicant deter-
mines the toxicity of organophosphorus pesticides.
In aquatic animals, as in mammals, choiinesterases are represented by
two types: acetyicholinesterase (ACE) and cholinesterase (CE). The enzyme
40
-------
TABLE 2. TOXICITY OF ORGANOPHOSPHORUS PESTICIDES FOR FISH WITH 4b-h
EXPOSURE AT TEMPERATURE OF lb-21°C,
Fish species
Stage, age, and size,
mm
MTC
LC
Chlorotos
Cyprinus carpio Current Year Brood (CY)
2-year old
Carassius carasslus CY
2-year old
90
340
100
1157
500
150
Abramis brama
A. ballerus
Rutllus rutilus
Perca tluviatilis
Salmo irrideus
Lebistes reticulatus
DlJVP [dichlorvosj
Cyprinus carpio
Perca tluviatilis
Carbotos
Cyprinus carpio
Perca fluviatilis
200 — — 200
200 — — 70
140 — 30 60
120 0.25 O.fa2 1.95
CY — 1
Adult — 14
CY i2.1 21.9 44.1
CY 0.37 0.59 1
CY 30 50 100
CY — 0.034*
*According to Prokopenko et al. 1975
41
-------
TABLE 3. SURVIVAL OF FISH LEECH COCOONS IN CHLOROFOS SOLUTIONS, 48-h
EXPOSURE
Chlorof os
concentration
mg/
Caspiobdeila
20
20
20
10
10
10
2
2
2
Control
Cocoon
age,
days
f adejevi
(period o± embryonic
1-4
10-11
lb-20
1-4
9-11
17-20
1-4
9-12
lb-20
—
Number
of
cocoons
development ,
33
40
27
44
50
30
84
50
31
175
Hatched
quantity
20 days)
3
0
0
44
18
2
84
36
15
175
fry
%
y
0
0
100
36
6.7
100
70
48.5
100
Pisciocola geometra
50
50
50
5
5
5
Control
(.period of embryonic
1-2
7-8
10-12
3-4
7-8
11-12
—
development ,
24
25
41
44
48
35
65
14 days)
10
0
0
44
28
13
65
41.5
0
0
100
58.5
37
100
42
-------
TABLE 4. TOXICITY OF ORGANOPHOSPtiORUS COMPOUNDS FOR FISH AT DIFFERENT STAGES OF LIFE CYCLE, 96-H.
Fish species
Salmo clarki
Salvelinus fontinalis
S. namaycush
Acipenser stellatus
Cyprinus carpio
^
-------
ol the lish brain is typical ACE. There is ACE in biood serum ol lish, the
properties ol which resemble the enzyme ol the brain. At the same time,
there is a blooa serum enzyme relerable to the CE type in some representa-
tives ol the lamily ol Cyprinidae Uope, roach, bream) L1UJ. In the tested
aquatic invertebrates (with the exception ol the iresh-water oligochaetous
worm, Tubilex tubifex), there is one enzyme ol the ACE type represented in
dilterent molecular lorms. Homogenates ol the oligochaetous worm have high
BuCE and PrCE activity, which warrants the beiiel that this species has
two enzymes (ACE ana CE,), or one enzyme that occupies an intermediate
position, between ACE and CE, with regard to its properties figures 1
and 2).
Data on inhibition ol fish brain ACE by chlorofos, DDVP—metabolite
oi chlorolos and carbofos—are listed in Table 5. All oi the tested spe-
cies showed similar extent ol inhibition ol this enzyme by chlorolos. The
values ol pl^y constituted 3.6 lor the enzyme of roach (Rutilus rutilus)
and bream lAbramis brama) , and about 4.0 for the enzyme of carp (Cyprinus
carpio) and perch (Perca lluviatilis). The values of p^Q and K^ are
higher with carbolos and DDVP than with chlorolos, which is indicative of
the more marked inhibitory eliect of these toxicants. We also failed to
observe species-specific dillerences in enzyme sensitivity to DDVP and
carbolos.
TABLE 5. VALUES OF p!5u AND K^ FOR FISH BRAIN ACETYLCHOL1NESTERASE
DURING INTERACTION WITH CHLOROFOS, DDVP, AND CARBOFOS IN VITRO
Chlorolos
Fish species pl
-------
There also is a correlation between enzyme sensitivity ana product
toxicity in species that are close in their systematic place. Thus, in two
species or gastropod moliusks, the large snail ILimnaea stagnalisj and the
trumpet snail (.Planorbis corneus), which differ by 1UU times in resistance
to chlorotos, an enzyme of the ACE type is present in nerve ganglia, which
ditfers in relative eiectrophoretic mobility (REM) and sensitivity to the
toxicant. In the less resistant species, the large snail, this enzyme is
more inhibited by chlorofos (.Figure 3).
In the species of the family Cyprinidae that we studied, brain and
blood serum ACE had the same sensitivity to chlorotos. At the same1time,
there was a correlation between fish resistance to this toxic agent and
blood serum CE sensitivity. For example, blood serum contains toxicant-
sensitive CE in the fish species that is less resistant to chlorofos (zope,
4tt-h LC^QQ = 7U mg/ i) , and the more resistant species (.carp, 4b-h L
5UU mg/^> does not have this enzyme (.Figure 4).
W
'in
~o
-o
I*
CD
"5
"w
JD
3
CO
a>
o
OC
A. Cyprinus carpio
.(brain)
1400
1200
1000
800
600
. 400
200
0
D. Not identified
336r
280
224
168
I 12
56
0,
ATCBr
B. Perca f luviatilis
(brain)
350
300
250
200
150
100
50
0
E. Not identified
168
140
I 12
84
56
28
0
ATCBr
5 4
C.Tubifex tubifex
(whole)
798
684
570
456
342
j
228
I 14
0
ATCBr
PrTCBr
BuTCBr
F Asellus aquaticus
(head]
ATCBr/
456
380
304
228
152
76
0
BuTCBr
0—0—O-rO-Oi
54321
Negative Logarithm of Molar Concentration of Substrate
Figure 1. Substrate specificity of aquatic animal cholinesterases at 3U°,
pH 7.5.
45
-------
REM 0.07-acetyicholinesterase
REM 0.26-0.29-cholinesterase
0.07
0.26
0.29
UJ
on
sss
+ 1 234
l-Cyprinus carpio 2-Abramis ballerus
3-Rutilus rutilus 4-Perco fluviatilis
Figure 2. hlectrophoregram ot fish blood serum cholinesterases,
005
0.195
UJ
ir
&2600
\
E 2200
=L
.« 1800
1000
k_
2 600
3
o 200
^ 0
ct:
A. Limnea staanalis
B. Planorbis corneus
4321
II I
I-Cholinesterase Electrophoregram
II-Substrate Specificity of Cholinesterases
Figure 3. Acetylcholinesterase of mollusk ganglia.
46
-------
TABLE 6. INHIBITION OF ACETYLCHOL1NESTERASE OF SOME SPECIES OF AQUATIC
INVERTEBRATES AND FISH IN VITRO BY CHLOROFOS, CARBOFOS AND DDVP,
I50 (INHIBITOR CONCENTRATION ELICITING 50% DEPRESSION OF ENZYME
ACTIVITY) IN 30 MIN.
Animal species
Daphnia magna
Tubitex tubifex
Limnaea stagnalis
Chironomus plumosus
Planorbis corneus
Asellus aquaticus
Cyclops sp.
Parca fluviatilis
Abramis brama
Cyprinus carpio
Rutilus rutilus
Carassius carassius
Source ot Chlorofos, DDVP Carbotos
enzyme M M M
Whole body 7 .5 -10~b
Whole body y-10~7 — 5-iU~b
Ganglia 5-10~~t> 1.1 °10~7 7*1U~^
Whole body 10"^ 5.1-lO~b 5'10~5
Ganglia 5- 10"-5 — S'lO"-3
Head b'10~^ — —
Whole body 3°10~4 — ti°10~4
Brain 10~4 4.1-iO~b 2.y-10~3
Brain 5-10"4
jirain 10~4 2.5-10~b 10~5
Brain 5-10~i*
Brain — — l.btt-lU"-*
A correlation between enzyme sensitivity to organophosphorus pesti-
cides and systemic resistance is demonstrable in the same species with
regard to ditterent toxicants. Thus, T^. tubitex and L_. stagnalis are less
resistant to chlorolos than carootos, and the enzyme ot these species is
also more sensitive to chlorofos (.Table 7).
Regardless ot degree ot ACE sensitivity to the toxic agents in vitro,
a decline in enzyme activity occurs in the case ot both acute and sublethai
intoxication, and it precedes appearance ot external signs of poisoning.
Thus, in carp exposed to acutely lethal concentrations ot carbotos (.A^-h
LC^y = 50 mg/£, in a state ol heightened excitability ; hydrolyzing capacity
ol brain ACE decreased by 54.7/i, with 73. 5£ loss ot equilibrium retlex.
Thirty percent ot the tish died within 24 h, and the rest were on their
47
-------
side at the bottom ot the aquarium.
low—7.6% (Table «).
In them, enzyme activity was very
With sublethal concentrations, upon appearance or the tirst signs ol
poisoning there were also reliable ditterences in ACE activity between the
experiment ana the control. When loss ot the equilibrium ret lex was ob-
served, enzyme activity constituted only 15.5%. Thereatter, the condition
ot the tish improved. After 2 days there was recovery ot the equilibrium
retlex and ACE activity also rose somewhat, but on the whole, enzyme level
was very low alter both 2 and 5 days—19.5 and 22.2%, respectively. Enzyme
activity was not tully restored at a later time (1U-2U days), although out-
wardly the condition ot the tish did not ditfer from the control (Table 9)
(Kozlovsk.aya 1983).
In gastropod mollusks CL. stagnalis and P^. corneus) , the tirst symptom
of chiorofos intoxication was weight gain due to excessive accummulation
ot fluid in the body. A comparison ot changes of weight and ACE activity
ot nerve ganglia revealed that the decline in enzyme activity preceded the
increase in mass (.Table 1U) (Kozlovskaya et al. 19b2).
Hog slater (A. aquaticus) specimens placed in glass containers without
rough surfaces are capable ot aggregating (forming collections). In the
presence oi chlorolos poisoning, depending on its severity, the animals
separate (crawl arouna the dish) due to their excited state. Reliable
inhibition ot ACE (by 45-50%) occurs arter 4 h, with U.01 mg/Ji chlorotos
concentration, which is distinct disaggregation.
A. Abramis ballerus
B. Cyprinus carpio
Acetylcholinesterase, REM 0.07 QCholinesterase, REM 0.26
0.07
0.26
UJ
o:
1
_L
-
W\
\\
100
Qt~\
80
^^
+-
1 60
o
o) 40
E
>»
c 20
UJ
0
-
"
•
.
V
^
V
V-
x;
|
^
'////////////////a"..
^
^
V;
^
V-
1
^
|
\
I
1
'&2W//////////A ,
SS\
_
-
-
.
^
|
I
10-6 10-5 IQ-4 iQ-3
Chiorofos Concentration, mol
Figure 4. Fish blood serum cholinesterases and their
chlorotos.
sensitivity to
48
-------
TABLE 7. TOX1CITY OF CHLOROFOS AND CARBOFOS FOR TUBIFEX TUBIFEX AND LIMNAEA
STAGNALIS, AND CHOLINESTERASE SENSITIVITY TO THESE AGENTS
48-h LCsd, mg/£
Animal species
T.
L.
tubif ex
stagnalis
chlorof os
0.029
0.5
carbof os
11.75
36.7
Enzyme inhibition,
1^0 in 30 min, M
chlorofos carbofos
9-10~7 5-10~6
All of the foregoing warrants the conclusion that the resistance ot
aquatic animals to organophosphorus pesticides depends on sensitivity ot
their cholinesterases. However, one cannot fail to consider other factors
as well. Resistance could be aue to the rate ot penetration of the toxic
agent into the body and to detoxification processes, as indicated by the
experiments dealing with toxicity oi chlorotos and DDVP tor iish as re-
lated to ditterent routes ot administration.
The carp and perch ditter signincantiy in resistance to chlorotos and
DDVP (, Table ll). Investigation ot sensitivity ot brain ACE in tne carp ana
perch to chlorotos revealed that pl^(j values are similar. 4.0 lor the carp
ana 4.2 tor the perch, brain and blood serum enzymes ot these two species
also tail to diller in sensitivity to DDVP- The K^^ values found tor them
are ot the same order, and are in the range ot 5.t>-10-1< to 9. 2-10^ m"1
TABLE b. CARP BRAIN ACE ACTIVITY IN THE PRESENCE OF ACUTE CARBOFOS
POISONING, 4b-H LCu,0 = 50 mg/£
Dynamics ot manifestation of
poisoning symptoms
35 min — heightened excitability
1.5 h — loss ot equilibrium reflex
24 h — loss of sensitivity
24 h — deaths
48 h — loss of sensitivity
Number
of fish
10
10
10
10
10
Acetylcholinesterase activity
jjmol ATC/(g-h)
168.9+39.5*
98.4+26.3*
29.2+9.9*
45.3+11.0*
24.3+4.07*
% of control
45.3
2b.5
7.8
12.2
6.5
*Ditterence between control and experiment is reliable with p 0.001,
49
-------
TABLE 9. CARP BRAIN ACETYLCHOLINESTERASE ACTIVITY WITH SUBLETHAL CONCEN-
TRATION OF CARBOFOS, 30 mg/£
Dynamics o± manelestation
Number Acetylcholinesterase activity
of fish ymol ATC/Cg'h) | % of control
UJ
1
i
2
h — heightened excitability
day — loss ot equilibrium reflex
days — restored equilibrium re-
10
1U
10
206.5+54.01*
57.8+15.02*
72.7+17.39*
55.4
15.5
19.5
flex, but body is arched when
fish swims
5 days—normal tish behavior, but 10
integument is dark
10 days—condition does not differ 10
from control fish
20 days—condition does not differ 10
from control fish
02.6+22.9*
151.5+40.5*
210.0+69.96*
40.6
56.4
*Difference between control and experiment is reliable with p 0.001
With the intraperitoneai route of intoxication, the carp also is more
resistant to DDVP than the perch. In this case, however, it is only 10
times more resis'tant than the perch (.Table 12) .
Intraperitoneai injection ot this toxic agent precludes the route ot
its usual penetration into fish. For this reason, the higher value lor
the ratio of LC^u or MVT tor carp to LC^u tor perch with the usual mode
ot intoxication, as compared to the LD^Q ratio with intraperitoneai injec-
tion is indicative of taster penetration ot the toxic agent into the perch.
This warrants the conclusion that the rate ot DDVP penetration determines
the ditterences in tish resistance to it. Water ana substances it contains
penetrates into all fresh-water tish mainly through the gills. The skin is
virtually uninvolved in this process (.Prosser 1977, Schmidt-Nielson 19b2j.
The difference noted between DDVP resistance ot carp and perch when
given by intraperitoneai injection can apparently be attributed to the
difference in rate of its detoxification in fish, which occurs enzymatic-
aliy, mainly in the liver. Fish liver homogenates actively break down
organophosphorus compounds, including DDVP (.Hogan and Knowies l96b). It
was shown that this process is taster in carp than in other tish species
(.Fujii and Asaka 1902).
50
-------
TABLE 10. CHANGE IN WEIGHT AND CHOL1NESTERASE ACTIVITY OF GANGLIA IN
LIMNAEA STAGNALIS WITH CHLOROFOS INTOXICATION
Time ot
experiment , h
0
3
b
24
4b**
^jj***
Number
of animals
15
15
15
15
15
12
Average
weight , %
100
100
10b.4
I2t>.3*
160.5*
97.1
Chonnesterase
activity, %
100
62.3*
45.7*
26. 9*
15.2*
y.i
*Uitlerence between control and experiment is reliable with p 0.01.
**Live moliusks
***Deaa moliusks
The inrormation about toxicity ot organophosphorus pesticides tor
aquatic animals makes it possible to assess, to some extent, their adverse
effect on aquatic fauna. Thus, judging by the results ot investigating
acute toxicity, we should expect that the presence ot pesticides in reser-
voirs would have a devastating etlect on leeches ot the herpobdeilidae and
Ichthyobdellidae families and, among those in the Hirudinidae tamily, it
would attect Hirudo medicinalis, which is useful in medicine, but would
not have a deleterious ettect on the predatory leech Haemopis sanguisuga.
Among fish, representatives of the Salmonidae and Percidae families would
be the most susceptible to organophosphorus pesticides. Fish would be
less resistant to toxic agents at early stages of ontogenesis. Ot all the
animals studied, crustaceans would be the most vulnerable.
Cladocera and, in particular, Daphnia pulex, which have low resistance
to organophosphorus compounds, are the most suitable as test objects to reg-
ulate levels of water pollution by organophosphorus compounds and assess the
toxicity of sewage trom industrial enterprises. One should bear in mind,
however, that plankton crustaceans have low resistance to a number of pollu-
tants, tor this reason, in order to .demonstrate expressly organophosphorus
compounds in water it is necessary to have adequate methods ot identifying
them. The enzymatic method could be used with success for this purpose, it
is based on measurement of cholinesterase activity in animals exposed to a
toxic agent or with direct introduction ot the enzyme into a toxic medium.
Commercial preparations and homogenates of aquatic animal organs and tis-
sues, which contain CE that is more sensitive to organophosphorus compounds
than ACE, can serve as sources ot enzyme lor such purposes.
51
-------
Because production ana use ot organophosphorus pesticides will remain
high in the tuture, along with regulations for their use, it is important to
search tor highly selective agents, with consideration ot their potential
toxicity tor aquatic animals. It is possible to solve the problem 01 syn-
thesizing agents with selective action only it the mechanisms ot resistance
to organophosphorus compounds ol animals in difterent systematic groups are
known.
In aquatic animals, resistance to organophosphorus compounds is
largely determined by the sensitivity of their target enzymes, AcE and CE.
In aquatic invertebrates that have low resistance to organophosphorus com-
pounds, ACE is more sensitive to toxic agents than in tish. Resistance
of gastropod moliusks also depends on the senstnvity of ACE of their
ganglia. Resistance of fish in the Cyprinidae family is unrelated to
sensitivity ot ACE of the brain, but is related to sensitivity ot blood
serum CE. As shown by the results ot the experiments with the carp and
perch, there may also be mechanisms other than sensitivity of target en-
zymes', such as rate of penetration ot the toxic agent into the body and
processes ot its detoxification, that determine the resistance ot tish
to organophosphorus compounds.
11. TOXICll"* OF CHLOROFOS AND DDVP FOR PERCH CPERCA FLUV1AT1L1S) AND
CARP (.CY.PR1NUS CARP10) WHEN FISH ARE PLACED IN TOXIC ENVIRONMENT
Fish Number Length ot Ratio ot carp
species ot tish tish, mm M+m 48-h LC^p, mg/& to perch L(
Chlorolos
Carp 28 b2+2 340.0 b48
(2by.8r428.4)*
Perch 48 b4+l U.b2
(O.bb-rO.70)
DDVP
Carp 3b b7+l 21.y 37
(20.2-T23.8;
Perch 42 b3+i O.by
(,0.b4-r0.b4)
*Coniidence intervals ot
52
-------
TABLE 12. TOXICITY OF DDVP FOR PERCH (PERCA FLUVIAT1LIS) AND CARP
(CYPRINUS CARPIO) WHEN GIVEN BY INTRAPERlTONEAL INJECTION
iish Number Length of Ratio o± carp
species ot fish fish, mm M+m LD^^pg/g to perch
Carp 20 109+2 292.0
(254.0T33b.O)* 9.b
)
Perch 42 109+2 30.4
(23.0-40.1)
*Lonfidence interval or
BIBLIOGRAPHY
Chuyko, G.M., V.I. Kozlovskaya, and V.M. Stepanova. 1983. Blood serum
carbonate esterases in the zope (Abramis ballerus), roach (Rutiius
rutilus) , bream (Abramis brama) and perch (Perca f luviatilis). Man-
uscript filea with the All-Union Institute of Scientific and Techni-
cal Information, IBVV [Institute of Biology of Inland Waters], USSR
Academy of Sciences, 22 February 1983, File No. bi93-83. 20 p.
Fujii, Y. and S. Asaka. 1982. Metabolism ot diazinon and diazoxon in
fish liver preparations. Bull. Environ. Contam. Toxicol. 29(4).
455-460.
Grishchenko, L.I., A.P- Verkhovskiy, and G.A. Trondina. 1975. Toxicity
of benzophosphate (Phosalone) t-or fish and detection ot poisoning.
In; Byulieten1 Vsesoyuznogo instituta ekspenmental'noy veternarii
(Bulletin ot the All-Union Institute ot Experimental Veterinary Sci-
ence). Moscow, USSR. pp. 58-bl.
Guseva, S.S. 1980. Effect of organophosphorus compounds on embryonic
and early postembryonic development of the carp. Abstract of candi-
datorial dissertation—biological sciences. Moscow, USSR. 22 pages.
Hogan, J.W. .and C.O. Knowles. 19b8. Degradation of organophosphates by
fish liver phosphatases. J. Fish Res. Board Canada. 25(8):1571-1579.
Johnson, W.W. and M.T. Finley. 19b4. Handbook of acute toxicity ot chemi-
cals to tish and aquatic invertebrates No. 137. Washington, USA. 9tt
P-
Koziovskaya, V.I., T.V. Volkova, andV.T. Komov. 1982. Choiinesterase
ot neural ganglia ana water metabolism in Limnaea stagnalis in the
presence ot chlorofos poisoning. In. Biologiya vnutrennikh vod. in-
to rm. byul. (Biology of Inland Waters. Information Bulletin) No. 55.
Leningrad, USSR. pp. 49-51.
53
-------
Kozlovskaya, V.I., V.M. Stepanova, ana G.M. Chuyko. iyb3. Reversibility
of carbotos poisoning ot carp. In; Reaktsii gidrobiontov na zagry-
azneniye (Hydrobiont Reactions to Poiiution). Leningrad, USSR. pp.
191-198.
Lapkina, L.N. 19b3. Leeches in the Rybinskiy Reservoir and their resis-
tance to toxic agents. Abstract of candidatoriai dissertation—bio-
logical sciences. Moscow, USSR. 23 p.
Lapkina, L.N. and B.A. Fierov. 1979. Investigation of acute poisoning
by certain toxic agents of leeches. In: iiziologiya i parazitolo-
giya presnovodnykh zhivotnykh (Physiology and Parasitology of Fresh-
Water Animals). Leningrad, USSR. pp. 50-59.
O'Brian, R. 1971. Toxic phosphates. Moscow, USSR. 631 p.
Perevoznikov, M.S. 1979. Ichthyocidai properties of carbofos. GosMIORKh
No. 14b. Leningrad, USSR. pp. 42-52.
Post, G. and T.R. Schroeder. 1971. The toxicity of lour insecticides to
tour salmonid species. Bull. Environ. Contain. Toxicology. 6(2):
144-145.
Prokopenko, V.A., N.P. Sokol'skaya, S.S. Nikulina, and N.R. Kosinova. 1975.
Comparative ichthyotoxicological characteristics of methylnitro-
phos and methathion. In. Problemy vodnoy toksikologii (.Problems of
Marine Toxicology) Vol. 1. Petrozavodsk, USSR. pp. 11U-112.
Prosser, L. 1977. Fluid metabolism: Osmotic balance and hormonal regu-
lation. In: Sravnitel'naya riziologiya zhivotnykh (Comparative
Animal Physiology) Vol. 1. Moscow, USSR. pp. 21-11b.
Rozengart, V.I. and O.Ye. Shestobitov. 1978. izbiratel'naya tofcsicnnost'
foslororganicheskikh insektoakaritsidov (Selective Toxicity of Organo-
phosphorus Insecticides-Acaricides). Leningrad, USSR. 173 p.
Schmidt-Nielsen, K. 19»2. Animal physiology. Moscow, USSR. 800 p.
54
-------
ACUTE TOXICITY OF 1ROM CYANIDES AND IhlOCYANAIE TO TROUT
by
R.V. Thurston
T.A. Heming2
ABSTRACT
Toxic effects to rainbow trout and brook trout of exposure to terro-
cyanide, ferricyanide, and thiocyanate were examined, under controlled light
conditions in a series of 96-hour laboratory tests. Solutions of both iron
cyanides were more acutely toxic to rainbow trout when tested under light
than when tested in total darkness. Thiocyanate was acutely toxic at con-
centrations lower than suggested in a previous study, but its effects
appeared to be relatively unpredictable.
INTRODUCTION
The toxic etfects of cyanides arise primarily trom the formation of
CN~ complexes with the metal ions present in many proteins and the subse-
quent inhibition of protein function (for review see Vennesland et al.
19bl). In vertebrates, cytochrome c is the protein most sensitive to cya-
nide and, as a consequence, the principal effect ot cyanide is to inhibit
oxidative phosphorylation at the level ot the mitochondria. The major
pathway for detoxification of cyanide in most animals is via conversion of
cyanide to thiocyanate (SCN~), a reaction that is catalyzed by the enzyme
rhodanase. The resultant thiocyanate is excreted via the urinary tract.
Acute and chronic toxic effects of cyanides, and the effect levels,
have been extensively" studied in most groups of animals including fishes
(for review see Doudoroff 197b, Towill et al. 1978). Similar information
regarding iron cyanides, thiocyanate, and cyanate is limited, however,
especially for fishes and other lower vertebrates. This lack of data is
^-Fisheries Bioassay Laboratory, Montana State University, Bozeman, MT 59717
2Aquatic Toxicology Section, Alberta Environmental Centre, Vegreville,
Alberta, Canada TOB 4LO
55
-------
especially apparent with respect to the characteristics and kinetics of up-
take and excretion, tissue distribution, and metabolism.
Ferrocyanide LFe(CN)64~J and ferricyanide [FeCCN)^3"] are byproducts
of the cyanidation process employed in the extraction of gold and silver
±rom ore. Both complexes are present frequently in ore mine effluents.
Environmental concerns about these iron cyanide complexes generally center
around their photodegradation to free cyanide. Light of wavelength less
than 420 nm is active in the decomposition of ferrocyanide, and of less
than 480 nm in the decomposition of ferricyanide (Broderius and Smith,
1980). Light source and intensity, therefore, are critical in any toxicity
testing with these two anions. Free cyanide concentration, as well as
total cyanide concentration, is important in interpreting test results.
Although quality and intensity of light appear to be the most important
factors in the decomposition process, temperature, pH, and dissolved oxy-
gen also play significant roles.
Thiocyanate enters the aquatic biosphere directly as SCNT and indi-
rectly as a product of the metabolism of cyanide. The fact that endogenous
conversion ot cyanide to thiocyanate is the principal pathway tor cyanide
detoxification in many animals does not inter that thiocyanate is not
toxic. Thiocyanate inhibits transport of halides in the thyroid (Wolff,
1964), stomach (Davenport 1940, Davies 1951), cornea (Zadunaisky et al.
1971), and gills (Epstein et al. 1973, 1975) and also inhibits various
enzymes such as carbonic anhydrase (Davenport 1940) , succinic dehydro-
genase (Solomon et al. 1973), and moieties of ATPase (Katz and Epstein
1971, Solomon et al. 1973).
Thiocyanate has been used therapeutically in the treatment of hyper-
tension in humans, occasionally with unpredictable and tatal consequences
(Garvin 1939. Goodman and Gilman 1970). Toxic effects to humans are the
result of an action of SCN~ on the central nervous system and include
irritability, nervousness, hallucinations, psychosis, mania, delirium,
and convulsions (Barnett et al. 1951).
Evidence of the conversion of inorganic and organic thiocyanates to
cyanide has been found in mammals (Boxer and Rickards 1952a, 1952b, Gold-
stein and Reiders 1951, Pines and Crymble 1952). It has been proposed
that these conversions involve a red cell enzyme, thiocyanate oxidase
(Goldstein and Reiders 1953), and the enzyme glutathione-S-transferase
(Habig et al. 1975).
Information on the toxicity of thiocyanate to aquatic life forms is
scant and contains conflicting results (for reviews see Doudoroff 197b,
Huiatt et al. 1983, Towill et al. 1978). Doudoroft (1976), for example,
noted ettect levels under various conditions to occur at concentrations
ranging trom 29 to 5000 mg/liter SCN~; he concluded that "thiocyanate is
somewhat toxic." APHA et al. (1980) have dismissed the toxicity ot thio-
cyanate as being relatively unimportant unless subsequent chlorination,
which would produce cyanogen chloride, is anticipated. The U.S. Environ-
mental Protection Agency (U.S. EPA 1980) does not list a water quality
criterion for thiocyanate. Further examination ot the aquatic toxicity
56
-------
of Chiocyanate is timely because of the recent development of the INCO
SC>2/air process for cyanide removal from industrial effluents. This pro-
cess selectively oxidizes cyanide and metal cyanide complexes, but has
little effect on thiocyanate (Devuyst et al. 1982, Huiatt et al. 1983).
In this presentation, we will discuss the results of recent unpub-
lished research at Fisheries Bioassay Laboratory (FBL) on the acute tox-
icity to fishes of both iron cyanide complexes and thiocyanate. The toxi-
cants we tested were reagent grade potassium ferrocyanide, potassium ferro-
cyanate, and potassium thiocyanate. The test fish were hatchery-reared
rainbow trout (Salmo gairdneri) and brook trout (Salvelinus f ontinalis).
Chemical analyses were conducted according to methods prescribed by APHA
et al. (1980), ASTM (.1981). and U.S. EPA (1974). Fish were tested using
diluters either of the design o± Benoit (1982) or Mount and Brungs (1967).
Each diluter delivered water to one control tank and five test tanks;
each of the test tanks was maintained at a different concentration of toxi-
cant. Except as noted, ten fish were tested in each tank. Median lethal
concentration (LC50) values were determined according to the Trimmed
SpearmanKarber method (Hamilton et al. 1977). More complete details of
the testing methods have been, or will be, reported elsewhere (Meyn et al.
1984, Kerning et al. Submitted).
FERROCYAN1DE AND FERR1CYANIDE TOXICITY TESTS
TEST PROCEDURES
A series of 96-hour, flow-through toxicity tests was conducted on
rainbow trout to determine the toxicity of ferrocyanide and ferricyanide
under two light regimes: total darkness and a cycle of 18 hours light/
6 hours darkness. A minimum of two tests were run with each chemical under
each light regime. Four additional tests, two each on ferrocyanide~and
ferricyanide, were conducted at an elevated temperature under a cyclic
regime of 18 hours light/6 hours darkness. The tanks were glass aquaria
with glass covers, having a water volume of approximately 14 liters. The
diluters had a flow rate of 30 ml/minute and the turnover time was approx-
imately 8 hours. Tests were started with the test concentrations already
established in the tanks. Fish for a given test were from a single pool
and were distributed at random among the control and test tanks. The
tests were conducted in a room equipped with wide-spectrum bulbs (Spectra-
lite, Long-Lite Lighting Products) that were reported to produce a spec-
tral energy distribution similar to daylight. For tests conducted in
total darkness, a red bulb was used for illumination during the collection
of water samples and during mortality checks.
In designing the experiment, it was assumed that there might be
marked differences in photodecomposition of the toxicants between a test
regime of total darkness and one that, provided light during a simulated
daytime period. It was further assumed that, at relatively low light
intensities (as compared with the intensity of bright sunlight, modest
differences in intensity between tanks would not significantly affect test
results. As a consequence, no special attempt was made to ensure that
57
-------
light intensities at the water surface of the different tanks in each test
were identical. Two sets of test tanks were used; the range of light
intensity at the tank water surface for one of these was 230 to 400 lux
(mean 330 lux) and for the other was 550 to 780 lux (mean 680 lux). It
subsequently developed that these light intensity differences most probably
played a tar greater role in determining the test results than originally
anticipated; these differences may have been sufficient to mask differ-
ences in results attributable to temperature within the range tested (9 to
16°C).
Water samples for cyanide analysis were collected from the effluent
of each tank three times during each test. Sodium hydroxide was added to
each sample to raise the pH to 12; samples were then refrigerated and
shipped to Homestake Mining Company at Lead, South Dakota, for analysis.
Total cyanide (total CN~) was determined using a modified acid reflux/
distillation method (ASTM 1981), and "weak acid dissociable cyanides"
were determined by "Method C" (ASTM 1982). Under our test conditions
using FBL water, which contained only insignificant quantities of metals,
"weak acid dissociable cyanides" was essentially tree cyanide. Results
tor all tests were reported in terms of total CN~; results for two tests
in the dark and all tests in the light also were reported in terms of
Method-C CN~. All other water measurements were made at FBL, and included
dissolved oxygen, temperature, pH, alkalinity, and hardness (Meyn et al.
1984).
It should be noted that solutions of ferricyanide ion are dis-
tinctly yellow. The intensity of color increases with increasing concen-
trations and was almost orange at the highest concentrations tested. By
contrast, ferrocyanide ion imparts only a pale yellow color to its solu-
tions, even at very high concentrations. The presence of iron inter-
fered with alkalinity and hardness determinations in all but the control
tanks. Light intensity measurements were made with a Gossen Panlux
Electronic FootLambert meter, recorded as foot-candles and converted to
lux.
The LC50 values derived in terms of either total CN~ or Method-C CN~
should be considered valid only under the conditions of testing, i.e.,
total darkness or 18-hour day, and at the specified temperature and con-
centration of dissolved oxygen. The LC50 values expressed in terms of
Method-C CN~ are further restricted by being computed from the means of
concentrations that increased during some of the tests.
TEST RESULTS
Comparison of the results of the ferrocyanide tests conducted in
total darkness (Tests 1074, 1077, 1078) and those under an intermittent
light regime of either 330 lux (Tests 1088, 1090) or 680 lux (Tests
1080, 1083) shows the fish were more sensitive to the test water in those
tests conducted under light (Table 1). Comparison of results of the
tests conducted under light further shows that the fish were more sensi-
tive to the test water at 680 lux than at 330 lux. A variable introduced
58
-------
TABLE 1. FERROCYAN1DE TOXICITY TESTS - SUMMARY OF TEST CONDITIONS AND 96-HOUR LC50 VALUES OBTAINED
(Ten fish per tank)
Ln
ID
Test
no.
1U74
1077
1078
1088
1090
1080
1083
Light regime,
hours of
light/dark
0/24
0/24
0/24
18/6
18/6
18/6
18/6
Mean
light
intensity,
lux
0
0
0
330
330
680
680
Mean
pH
(range)
-
7.95
(7.89-8.01)
7.77
(7.66-7.94)
7.83
(7.75-7.95)
7.33
(7.30-7.34)
7.55
(7.50-7.57)
Mean
temperature ,
°C
(range)
9.4
(9.4-9.5)
-
9.4
(9.4-9.5)
15.7
(15.4-16.1)
15.2
(15.2-15.4)
9.6
(9.6-9.7)
9.5
(9.5-9.6)
Mean
D.O.,
nig/liter
(range)
8.44
(7.03-8.85)
-
8.83
(8.76-8.97)
8.22
(8.01-8.49)
7.85
(7.64-8.11)
7.93
(7.77-8.17)
8.37
(8.16-8.58)
Method-C CN~
96-h LC50,
mg/liter
(95% C.I.)
-
1.10
(0.99-1.23)
0.79
(0.63-1.0)
0.71
(0.51-0.98)
0.15
(0.13-0.17)
0.17
( - )*
Total CN~
96-h LC50,
mg/liter
(95% C.I.)
752
(474-1190)
> 867
939
(828-1064)
220
(184-261)
232
(178-302)
33.0
(28.7-37.9
37.4
( - )*
*Confidence interval not calculable by method used.
-------
in this second comparison, however, is that the 330-lux tests were con-
ducted at a higher temperature (15 to 16°C) than those at 680 lux (9 to
10°C).
Comparison of the results of the ferricyanide tests conducted in
total darkness (.Tests 1075, 1076, 1079) with those conducted under an
intermittent light regime of 680 lux (Tests 1085, 1087, 1089) shows
markedly greater toxicity under light (Table 2). Of the three tests
conducted at 680 lux, one was conducted at 10°C (Test 1085) and the
other two at 16°C (Tests 1087, 1089). Results show a slight increase
in toxicity at the higher temperature, but the overlap in LC50 confi-
dence intervals for these tests suggests that the finding is not con-
clusive.
To see whether there are additional conclusions that might be
drawn from these data, we combined the results ot all tests and made the
assumption that differences in toxicity, if anyt resulting from tempera-
ture differences were over-ridden by other factors. We looked at all
the data for each test, and from each we selected the tank with the high-
est Method-C CN~ concentration at which no more than 10% mortality
occurred (Table 3). For two of the ferricyanide tests, more than 10%
of the fish died at even the lowest concentration tested. These were
Test 1087 with 40% mortality in Tank 2, and Test 1089 with 30% mortality
in Tank 2. To include these tests in our data pool it was necessary to
use these tanks with > 10% mortality, presumably <^ 10% mortality would
have occurred at a lower Method-C CN~ concentration. We then plotted
the measured light intensity for each of these test tanks versus its
measured Method-C CN~ concentration, using the highest single measurement
of Method-C CN~ in each tank as opposed to the mean of three measurements.
The rationale for selecting the highest measured Method-C CN~ value was
that the toxic action of cyanide is rather rapid, and the fish that
died were probably most affected by that highest concentration, rather
than by a 96-hour average.
The results of this plot (Figure 1) are revealing. The estimated
curves for both sets ot tests (ferrocyanide and ferricyanide) are simi-
lar, and one may conjecture that there may not be a significant differ-
ence between them. The implications are that the toxicity of Method-C CN~
is light dependent, i.e., there is a correlation between the toxicity of
Method-C CN~ and light intensity.
The behaviour and distress symptoms of fish in the ferrocyanide and
ferricyanide acute toxicity tests were similar to, but less dramatic than,
those that we also observed during acute toxicity tests on thiocyanate.
These symptoms included flaring of the gills, gaping of the mouth, erratic
and sudden movement, and immediate stiffness ot the body when the tish
died. The symptoms were particularly noticeable at the termination of a
test when all remaining fish were anesthetized with tricainemethanesulfonate
prior to weighing and measuring. The degree to which these symptoms were
evident was directly related to toxicant concentration; at the lowest
concentrations tested, and in the control tanks, fish did not evidence
these symptoms.
60
-------
TABLE 2. FERR1CYANIDE TOXICITY TESTS - SUMMARY OF TEST CONDITIONS AND 96-HOUR LC50 VALUES OBTAINED
(Ten fish per tank;
en
Light regime,
Test hours of
no. light/dark
1075 0/24
1076 0/24
1079 0/24
L081 18/6
1082 18/6
1084 18/6
1085 18/6
1087 18/6
1089 18/6
Mean
light Mean
intensity, pH
lux (range)
0 7.87
(7.46-8.23)
0 7.96
(7.75-8.17)
0
330 7.33
(7.30-7.34)
330 7.56
(7.53-7.58)
330
680 7.78
(7.76-7.80)
680 7.96
(7.89-8.06)
680 7.89
(7.79-8.10)
Mean
temperature ,
°C
(range)
9.3
(9.1-9.4)
9.3
(9.2-9.4)
-
9.6
(9.5-9.7)
9.6
(9.5-9.7)
-
10.3
(10.1-10.7)
16.0
(15.8-16.1)
16.0
(15.5-16.4)
Mean
D.O. ,
mg/liter
(range)
8.18
(7.03-8.85)
8.88
(8.70-9.01)
-
7.91
(7.61-8.11)
8.28
(8.09-8.43)
-
8.26
(8.13-8.34)
8.69
(8.58-8.83)
8.16
(8.05-8.37)
Method-C CN~ Total CN~
96-h LC50, 96-h LC50 ,
mg/liter mg/liter
(95% C.I.) (95% C.I.)
1210
(1060-1380)
869
(680-1110)
> 1.14 > 877
0.24 69.6
(0.23-0.26) (54.7-88.5)
> 0.50 > 541
> 0.61 > 731
0.42 44.2
(0.36-0.50) (38.2-51.0)
0.40 38.8
(0.36-0.43) (25.0-60.2)
0.31 10.8
(0.30-0.32) (8.96-13.1)
-------
TABLE 3. LIGHT INTENSITY IN TANKS WITH HIGHEST METHOD-C CM" CONCENTRATION
AT WHICH NO GREATER THAN 10% MORTALITY OCCURRED
" ~ Tank with <^ 10% mortality
Chemical and Highest measuredLight intensity,
test no. Method-C CN~, mg/liter lux
Ferrocyanide
1074 1.15 0
1078 1-13 0
1080 0.13 780
1083 0.14 590
1088 0.35 250
1090 0.49 310
Ferricyanide
1079 1.30 0
1081 0.22 310
1082 0.64 400
1084 0.58 390
1085 0.46 780
1087 0.47* 780
1089 0.33T 780
*Test tank reported is that with lowest mortality (40%). [See text]
TTest tank reported is that with lowest mortality (.30%). (.See text]
THIOCYANATE TOXICITY TESTS
TEST PROCEDURES
Acute (96-hour) toxicity tests were conducted on both brook trout
and rainbow trout under continuous-flow conditions as described above,
except that these tests were started at zero toxicant concentration, and
full test toxicant concentrations were effectively reached within 18 hours
Tests were conducted under two conditions: an ambient photoperiod with
fish visually exposed to normal laboratory traffic and activities and con-
trolled photoperiods with fish isolated from these activities by curtains.
Controlled photoperiods were 24-hours white light (223 lux), 12-hours
light/12-hours dark, and 24-hours dim red light (0.2 lux). Effects of
stress on thiocyanate toxicity were assessed in tests in which fish were
forced to swim vigorously for 30 seconds, by inserting a hand-held dip net
into the tank water and moving it around taking care not to touch the fish.
62
-------
—O FERROCYANIDE
—,Q__ FERRICYANIDE
0.0 O.I 0.2 0.3 0.4 0.5 0.6 0.7 O.8 0.9 1.0 I.I i.2 1.3
METHOD-C CN" (mg/liter)
Figure 1. Test tank concentrations with £ 10% mortality vs. light intensity.
(Ferricyanide data points with arrows indicate > 10% mortality.)
TEST RESULTS
In conducting our initial acute toxicity tests on brook trout, we
observed that, although there was an overall dose response correlation,
some fish survived the 96-hour test period at thiocyanate concentrations
an order of magnitude greater than those that caused a high percentage
of mortality to others (Table 4). We also observed that the death of fish
frequently would be sudden, i.e. fish that evidenced little or no sign of
distress at the time of routine scheduled mortality observations were
under distress or dead shortly thereafter. We further observed that fish
with relatively placid behaviour in a given tank markedly increased their
swimming activity when we were collecting water samples, or when we were
removing the fish from the tank at the end of the test; some of the seem-
ingly healthy fish died in the dip net during removal from the tank.
These mortalities in the tank or upon removal were characterized by sudden
rapid movement, onset of tonic convulsions, loss of equilibrium and
buoyancy, gasping, flaring of the operculae, darkening of the skin epi-
thelia, and finally cessation of ventilation and extreme rigor. We
termed this occurrence "sudden death syndrome" (SDS).
63
-------
TABLE 4. PERCENT MORTALITY OF BROOK TROUT* IN EACH TEST TANK DURING
96-HOUR ACUTE TOXICITY TESTS ON THIOCYANATEt (Tests 651,
652, 653, 655)
Test no. -
Tank no.
651-6
651-5
651-4
651-3
652-6
651-2
652-5
652-4
653-6
652-3
653-5
652-2
653-4
653-3
653-2
655-6
655-5
655-4
655-3
655-2
All tests-1
Exposure
concentration,
mg/ liter SCN~
517
375
298
232
198
163
148
115
96.4
90.8
72.1
65.4
56.1
44.3
31.6
31.3
24.1
18.5
14.8
10.5
0.0
No. o±
fish
10
10
10
10
10
10
10
10
15
10
15
9
15
15
15
10
10
10
10
10
45
Percent
mortality
at 96 h
90
90
80
80
90
80
70
90
73
90
47
56
73
53
47
50
70
30
20
0
0
*Mean fish size (.and range): 6.7(5.56-7.94) g, 8.2(6.1-10.3) cm.
tOther measured water variables, mean values and ranges for all tanks:
temperature 15.1(12.9-17.3)°C; dissolved oxygen 7.70(6.38-8.31) mg/liter;
PH 7.88(7.83-7.90); alkalinity 176(167-181) mg/liter as CaCO-i: hardness
203(195-208) mg/liter as CaC03.
To test the hypothesis that SDS was related to the fish being dis-
tressed by activity near the test area, two 96-hour acute toxicity tests,
one each on brook trout (Test 658) and rainbow trout (Test 723), were con-
ducted during which activity near the test area was reduced to a necessary
minimum. At the close of the standard 96-hour test periods, the fish were
stressed with a dip net, and mortality counts taken 15 minutes later (Test
658) and 75 minutes later (.Test 723) showed a dramatic increase over those
taken at 96 hours (Table 5). Two additional 96-hour tests on brook trout
(Tests 657 and 659) were conducted during which the fish (10 fish per
tank) were stressed with a dip net at 24, 48, 72, and 95 hours. Fish
64
-------
size and measured water chemistry variables for these tests were similar
to those reported tor Test 658. The 96-hour LC50 for Test 657 was 13.8
mg/liter SCN~ and that for Test 659 was 13.3 mg/liter SCN~.
To determine whether the toxicity of thiocyanate might differ with
different photoperiods, duplicate 96-hour tests were conducted on rainbow
trout under each of three different light regimes: continuous exposure
under red light, a daily regime of 12 hours white light and 12 hours in the
dark, and continuous exposure to white light. The fish were least sensitive
to thiocyanate toxicity under red light, more so under the alternate light/
dark regime, and most sensitive under continuous exposure to white light
(Table 6). At the end of the 96-hour test periods, the fish tested under
continuous red light were briefly exposed to white light, and the light
was briefly turned off on the fish tested continuously under white light;
mortality counts were taken again 45 minutes later tor each of these tests.
Although not as dramatic as in the tests with trout subjected to dip net
stress at the close of the 96-hour test period (Table 5), there was an
TABLE 5. PERCENT MORTALITY BEFORE AND AFTER STRESS OF FISH EXPOSED
TO THIOCYANATE* (Ten fish/tank)
Tank
no.
Brook troutt (Test 658)
Exposure
concentration
mg/liter SCN~
Percent
mortality
96h
96.25h
Rainbow trout** (Test 723)
Exposure
concentration
mg/liter SCN~
Percent
mortality
96h
97.25h
1
2
3
4
5
6
0.00
5.52
8.00
9.99
13.0
16.7
0
0
10
10
0
10
0
10
70
70
80
100
0.02
7.70
9.90
11.8
15.3
20.8
0
0
0
30
0
50
0
100
100
80
100
90
*0ther measured water chemistry variables, mean values, and ranges for all
tanks: Test 658 - temperature 16.3 (15.9-16.7)°C, dissolved oxygen
7.54(7.20-7.93) mg/liter; pH 7.86(7.83-7.95), alkalinity 164 mg/liter as
CaC03 (all measurements), hardness 194(194-195) mg/liter as CaC03; Test
723 - temperature 11.8(11.3-12.5)°C, dissolved oxygen 9.35(9.20-9.78) mg/
liter, pH 7.95(7.92-8.00), alkalinity 175(172-178) mg/liter as CaC03,
hardness 202(198-206) mg/liter as CaC03.
TMean fish size (and range) 8.9 g, 9.1(7.8-10.7) cm; 96-h LC50 > 16.7 mg/
liter SCN~, 96.25-h LC50 7.79 mg/iiter SCN~ (95% C.I. 6.62-9.16).
**Mean fish size (and range) 0.72(0.32-1.42) g, 4.16(3.15-5.25) cm; 96-h
LC-50 20.8 mg/liter SCN~ (95% C.I. not calculable by method used),
97.25-h LC50 < 7.70 mg/liter SCN~.
65
-------
TABLE 6. ACUTE TOX1CITY OF THIOCYANATE. TO RAINBOW TROUT* DURING TESTSt
WITH DIFFERENT LIGHT REGIMES (Ten fish/tank)
LC50 mg/liter SCN~ (95% C.I.)
Test
no.
721
725
727
728
720
724
Light regime
Continuous red light
Continuous red light
12h white light/12h dark
12h white light/ 12h dark
Continuous white light
Continuous white light
96-h
159 (119-213)
64.6 (13.2-316)
83.5 **
55.7 (39.0-79.6)
48.3 (34.6-67.5)
44.9 (30.0-67.2)
9b.75-h
76.3 (48.7-120)
64.6 (13.2-316)
3b.7 (25.7-58.3)
41.4 (27.4-62.5)
*Mean fish size (and range) 0.83(0.28-1.79) g, 4.42(3.10-5.55) cm.
tOther measured water variables, mean values, and ranges for all tanks:
temperature 13.0(12.2-13.5)°C, dissolved oxygen 8.96(8.66-9.30) mg/liter,
pH. 7.93 (7.77-8.87), alkalinity 174(168-206) mg/liter as CaC03, hardness
202(200-206) mg/liter as CaC03.
**Confidence intervals not calculable by method used.
increase in mortality in three out of four of these tests following the
sudden change in light regime.
The persistence of SDS was tested by stressing (dip net method) sep-
arate groups of juvenile rainbow trout (10 fish per group) at 0, 4, 12,
24, and 48 hours after exposure to thiocyanate for 96 hours under flow-
through conditions, a single control group was stressed at each of these
time periods. After the 96-hour exposure period, thiocyanate addition to
the incoming water to each test tank was discontinued, concentrations in
the test tanks at 4 hours were approximately 4 mg/liter SCN~, and < 1 mg/
liter SCN~ by 12 hours. The group stressed at 4 hours suffered 60% mor-
tality, and none of the fish died after stress at 48 hours (Table 7).
DISCUSSION
Solutions of both ferrocyanide and ferricyanide proved to be more
acutely toxic to rainbow trout when tested under light than when tested in
total darkness. Further, there was a trend toward greater toxicity of
either of these solutions at greater light intensity. Both of these poten-
tial toxicants are known to decompose under light to produce free cyanide.
Although it was not possible for us to measure free cyanide, the concentra-
66
-------
tion of Method-C CN~ was assumed to approximate the concentration of free
cyanide.
If free cyanide alone were the toxic agent, and it Method-C CN~
concentrations approximate free cyanide concentrations, then calculation
of LC50 values in terms of Method-C CN~ should have been comparable for
all tests regardless of light regime. This was not the case; toxicity
of the solutions tested in terms of Method-C CN~, whether considering
the LC50 values for either or both sets of tests, or considering the
individual tanks with _< 10% mortality for both sets of tests combined,
revealed a relationship between available light and mortality of fish.
We conclude that Method-C CN~ is not a true (approximate) measure of
free-cyanide under the conditions of our tests and/or that free cyanide was
not the sole toxic agent in the tests conducted. Additional research will
be necessary to quantify better the toxicity of solutions of iron cyanide
complexes, and the conditions of their photodegradation. In conducting
this additional research, greater care should be exercised to standardize
light conditions both among tanks for any given test, and between tests
that are to be compared for other variables such as temperature. Further,
the procedures for analysis of different forms of cyanide in the test
waters should be reviewed to ensure that samples collected have not appre-
ciably altered by the time of analysis.
TABLE 7. PERSISTENCE OF SUDDEN DEATH SYNDROME IN JUVENILE RAINBOW TROUT*
MAINTAINED IN FRESH WATER FOR DIFFERENT TIME PERIODS AFTER 96-
HOUR EXPOSURE TO SCN~ (Test 726)
Group
no.
1
2
3
4
5
6
Exposure
concentration,
mg/liter
SCN-
0.0
6.7
7.3
6.y
b.o
7.7
-Mortalities
during
96-h
exposure
0
2
0
1
1
1
Hours in
replacement
water after
SCN~ exposure
and before
30s stressT
0
0
4
12
24
4b
Additional
mortalities
within
45-m
after
stress
0
b
6
2
3
0
Total
mortalities
during
entire
test
period
0
b
6
3
4
1
*Ten fish per group, mean size and range 1.03(0.37-2.4b) g, 4.57(3.25-6.00)
cm.
TNo fish died during these periods.
57
-------
The results of our tests on thiocyanate demonstrate it to be acutely
toxic to trout at concentrations lower than those suggested by previous
studies (see Doudoroff 1976). The acute toxicity of thiocyanate to trout,
however, would appear to be relatively unpredictable. Exposed fish, which
seemed healthy, would suddenly and often tor no apparent reason begin to
convulse and quickly expire.
The signs of SDS in trout (convulsions, loss of equilibrium and
buoyancy, extreme rigor, among others) are virtually identical to the
characteristics of acute thiocyanate toxicity in mammals, including humans
(convulsions, vertigo, coma, hyper-reactivity, extensor rigidity) (Garvin
1939, Smith 1973). In mammals, these effects are probably due to the SCN~
ion itself, rather than to its metabolic products, HCN, OCN , and NH-j
(Smith 1973); administration of sodium thiosulphate, an antidote for cya-
nide poisoning, has no effect on thiocyanate toxicity in mammals. Thiocya-
nate toxicity in mammals would appear to be influenced by neural activity,
inasmuch as sympathomimetic stimulants (e.g. amphetamines) increase the
toxicity of SCN~, whereas barbiturate anaesthetics (e.g. phenobarbital)
decrease its toxicity (Smith 1973).
SDS in fish may represent a direct effect of the thiocyanate anion
on neuromuscular functioning; SCN~ has been shown to enhance the con-
tractility and electrical conduction of frog muscle fibers (Lubin 1957),
and to increase the excitability and electrical transmission of spinal
neural pathways and motoneurons (Goto and Esplin 1961). Regardless of the
exact mechanism involved in SDS, normal rapid movements of wild fish
during feeding and predator avoidance probably would be sufficient to
trigger SDS as a consequence of thiocyanate exposure.
Standard acute toxicity tests (96-hour LC5U determinations) are of
limited value in predicting the toxicity of thiocyanate because of the
unpredictability of SDS. Subsequent to the studies reported here, we de-
termined that the correlation between plasma thiocyanate concentrations and
SDS provides a more certain measure of thiocyanate toxicity (Heming et al.
Submitted). Those results demonstrated that fish readily accumulate SCN~
and 50% of the fish tested would be at risk of SDS when plasma SCN~ concen-
trations reached approximately 4 mmol/liter. The duration of thiocyanate
exposure necessary to attain that plasma concentration depends upon the
exposure concentration, fish size, water temperature, and ambient water
quality.
It is possible that, under conditions when uptake of SCN~ is extreme-
ly slow, excretion of SCN~ would be able to keep pace with SCN~ uptake and
plasma concentrations of thiocyanate would remain low. Under such circum-
stances, SDS would not be expected to occur. None of our experiments was
run at a sufficiently low concentration of thiocyanate to test this hypoth-
esis; the lowest concentration of toxicant we tested was 17 umol/liter
KSCN. At that level of exposure, 6.7 g rainbow trout accumulated thiocya-
nate at a rate of approximately 15 mmol/liter plasma/day/kg fish. Given
such a rate of SCN~ accumulation, one would expect that 50% of the exposed
population would be at risk of SDS within 1-2 months. During such long
term exposures, fish probably would encounter the chronic toxic effects of
68
-------
thiocyanate. In mammals, and presumably in fishes as well, SCN~ inhibits
incorporation of I~ by the thyroid gland, which in the long term results
in goiter (Wolff 1964).
ACKNOWLEDGMENTS
Elizabeth L. Meyn, Donald R. Skaar, and Richard K. Zajdel were
principally responsible tor performing the acute toxicity tests, and Ms.
Meyn and Terry L. Mudder (Homestake Mining Company) for the water cjiem-
istry analyses. This research was funded by the U.S. Environmental Pro-
tection Agency, Environmental Research Laboratory, Duluth, Minnesota,
Research Grant CRbU7240, and by the Homestake Mining Company, Lead, South
Dakota.
REFERENCES
APHA (American Public Health Association), American Water Works Associa-
tion, and Water Pollution Control Federation. 1980. Standard
methods for the examination of water and wastewater. 15th ed.
American Public Health Association, Washington, D.C.
ASTM (American Society for Testing and Materials). 1981. Proposed re-
vision of standard methods of test for cyanide in water, Standard
D-2036-81. American Society for Testing and Materials, Philadelphia,
Pennsylvania.
ASTM (American Society for Testing and Materials). 1982. Designation
D2036. Annual book of standards. American Society for Testing and
Materials. Philadelphia, Pennsylvania.
Barnett, H.J.M., M.V. Jackson, and W.B. Spaulding. 1951. Thiocyanate
psychosis. J. Am. Med. Assoc. 147: 1554-1558.
Benoit, D.A., V.R. Mattson, and D.L. Olson. 1982. A continuous-flow mini-
diluter system for toxicity testing. Water Res. 16: 457-464.
Boxer, G.E., and J.C. Rickards. 1952a. Determination of traces of hydro-
gen cyanide in respiratory air. Archs. Biochem. Biophys. 39: 287-
291.
Boxer, G.E., and J.C. Rickards. 1952b. Determination of thiocyanate in
body fluids. Archs. Biochem. Biophys. 39: 292-300.
Broderius, S.J., and L.L. Smith. 1980. Final Report. Grant No. R805291.
U.S. Environmental Protection Agency, Duluth, Minnesota.
Davenport, H.W. 1940. The inhibition of carbonic anhydrase and of gastric
secretion by thiocyanate. Am. J. Physiol. 129: 505-514.
Davies, R.E. 1951. The mechanism of hydrochloric acid production by the
stomach. Biol. Rev. 26: 87-120.
69
-------
Devuyst, E.A., B.R. Conard, and V.A. Ettel. 1982. Pilot plant operation
of the INCO SC>2/air cyanide removal process. Can. Mining J. August
1982: 27-30.
Douaoroff, P. 1976. Toxicity to fish of cyanides and related compounds.
A review. U.S. Environmental Protection Agency, Duluth, Minnesota.
EPA-600/3-76-038.
Epstein, F.H., J. Maetz, and G. de Renzis. 1973. Active transport of
chloride by the teleost gill: inhibition by thiocyanate. Am. J.
Physiol. 224: 1295-1299.
Epstein, F.H., P- Silva, J.M. Forrest, and R.J. Solomon. 1975. Chloride
transport and its inhibition by thiocyanate in gills of seawater
teleosts. Comp. Biochem. Physiol. 52A: 681-683.
Garvin, C.F. 1939. The fatal toxic manifestations of the thiocyanates.
J. Am. Med. Assoc. 112; 1125-1127.
Goldstein, F., and F. Reiders. 1951. Formation of cyanide in dog and man
following administration of thiocyanate. Am. J. Physiol. 167: 47-51.
Goldstein, F., and F. Reiders. 1953. Conversion of thiocyanate to cyanide
by an erythrocytic enzyme. Am. J. Physiol. 173: 287-290.
Goodman, L., and A. Gilman. 1970. The pharmacological basis of thera-
peutics. 4th ed. The Macmillan Co., London, England.
Goto, K., and D.W. Esplin. 1961. Excitatory effects of thiocyanate on
spinal cord. J. Pharmacol. Exp. Ther. 133: 129-136.
Habig, W.H., J.H. Keen, and W.B. Jakoby. 1975. Glutathion S-transferase
in the formation of cyanide from organic thiocyanates and as an
organic nitrate reductase. Biochem. Biophys. Res. Commun. 64: 501-
506.
Hamilton, M.A., R.C. Russo, and R.V. Thurston. 1977. Trimmed Spearman-
Karber method for estimating median lethal concentrations in
toxicity bioassays. Environ. Sci. Technol. 11:714-719.
Correction — Environ. Sci. Technol. (1978) 12: 417.
Heming, T.A., R.V. Thurston, E.L. Meyn, and R.K. Zajdel. (Submitted).
Acute toxicity of thiocyanate to fishes: some physiological impli-
cations. Trans. Am. Fish. Soc.
Huiatt, J.L., J.E. Kerrigan, F.A. Olson, and G.L. Potter. 1983. Work-
shop — Cyanide from Mineral Processing. Utah Mining and Mineral
Resources Research Institute, Salt Lake City, Utah.
Katz, A.I., and F.H. Epstein. 1971. Effect of anions on adenosine
triphosphatase of kidney tissue. Enzymes 12: 499-507.
70
-------
Lubin, M. 1957- The effect of iodide and thiocyanate ions on the mechani-
cal and electrical properties of frog muscle. J. Cell Comp. Physiol.
49: 335-349.
Meyn, E.L., R.K. Zajdel, and R.V. Thurston. 1984. Acute toxicity of
ferrocyanide and ferricyanide to rainbow trout (Salmo gairdneri).
Technical Report No. 84-1, Fisheries Bioassay Laboratory, Montana
State University, Bozeman, Montana.
Mount, D.I., and W.A. Brungs. 1967. A simplified closing apparatus tor
fish toxicology studies. Water Res. 1: 21-29.
Pines, K.L., and M.M. Crymble. 1952. In vitro conversion of thiocyanate
to cyanide in the presence of erythrocytes. Proc. Soc. Exp. Biol.
Med. 81: 160-163.
Smith, R.P- 1973. Cyanate and thiocyanate: acute toxicity. Proc. Soc.
Exp. Biol. Med. 142: 1041-1044.
Solomon, R.J., P- Silva, and F.H. Epstein. 1973. Thiocyanate inhabitable
ATPase in the gill of Anguilla rostrata. Bull. Mt. Desert Island
Biol. Lab. 13: 115-117.
Towill, L.E., J.S. Drury, B.L. Whittield, E.B. Lewis, E.L. Galyan, and A.S.
Hammons. 1978. Reviews of the environmental effects of pollutants:
V. Cyanide. U.S. Environmental Protection Agency, Cincinnati, Ohio.
EPA-600/1-78-027.
U.S. EPA (U.S. Environmental Protection Agency). 1974. Methods for chemi-
cal analysis of water and wastes. U.S. Environmental Protection
Agency, Washington, D.C. EPA-625/t>-74-003.
U.S. EPA (U.S. Environmental Protection Agency). 1980. Ambient water qual-
ity criteria tor cyanides. U.S. Environmental Protection Agency,
Washington, D.C. EPA-440/5-80-037.
Vennesland, B., E.E. Conn, C.J. Knowles, J. Westley, and F. Wissing. 1981.
Cyanides in biology. Academic Press, London, England.
Wolff, J. 1964. Transport of iodide and other anions in the thyroid gland.
Physiol. Rev. 44: 45-90.
Zadunaisky, J.A., M.A. Lande, and J. Hafner. 1971. Further studies on
chloride transport in the frog cornea. Am. J. Physiol. 221: 1832-
1836.
71
-------
PROCESSES OF FORMATION Ot NATURAL WATER QUALITY
by
M.I. Kuz'menko and A.I. Merezhko1
ABSTRACT
This paper develops the notion of a unity in the processes that deter-
mine water quality in bodies of water and water courses. Relationships are
examined among the productivity of a body of water, the processes of self-
purification in it, and the resultant water quality. Using individual hy-
drobionts, in particular the higher aquatic plants and certain invertebrates,
as examples, an analyis is made of their vital activities in relation to
the processes that determine natural water quality. Significant attention
is given to a description of gas conditions in the body of water in relation
to photosynthesis in the higher aquatic plants. The role of individual
hydrobionts in the absorption ana accumulation of biogenic elements and
toxic compounds is shown. Particular attention was directed to the role of
the root systems of plants and to the filtration capacity of dreissensia
(.zebra mussel) in these processes. The conclusion is drawn that water
quality is a product of the functioning of the ecosystem as a whole.
INTRODUCTION AND DISCUSSION
A water reservoir is a complicated ecological system with more or less
marked and, not infrequently, quite discrete links between its elements.
The linkages are determined by the nature of hydroiogical, physicochemical
and biological processes that provide tor the function of the ecosystem as
an integral whole.
The productivity of the body of water and the quality of water in it
are integral indicators of function of aquatic ecosystems. While reservoir
productivity is closely related to its trophism, water quality is related
to its self-puritication processes. The latter takes place as a result of
biotic circulation of matter (Vinberg 1976) , which includes formation,
Institute of Hydrobiology, Ukrainian Academy of Sciences,
72
-------
conversion, and destruction of organic compounds. The quantitative ex-
pression of these processes and the relationship between them determines
the direction of self-purification.
Formation of an organic substance in water (.photosynthesis) is re-
lated to carbon dioxide uptake, use of solar energy, and absorption ot
biogenic elements. The relationship between these processes indicates
that the more intensive plant photosynthesis is, the more active is the
absorption ot biogenic elements.
An excess or shortage of these elements leads to disturbances of in-
trareservoir processes related to self-purification and tormation ot the
quality of naturally occurring water.
As a result of photosynthesis by aquatic plants alone, more than 400
billion tons of oxygen is discharged per year, and it would be difficult
to exaggerate its role in formation of water quality. According to our
data, perfoliate pondweed gives oft about 50 mg oxygen/h/100 g wet weight
during photosynthesis. The discharged oxygen is instrumental in reaeration
of water (.Vinberg iy7b) and provides rheophil conditions for hydrobionts.
Not infrequently, one observes oversaturation of water with oxygen
in macrophyte thickets, particularly around noon. Such conditions provide
tor normal vital function of hydrobionts and supply oxygen tor processes
ot oxidation of organic substances, especially in the summer. With ele-
vation ot temperature in thickets ot macrophytes, there is activation or
microbiological processes. There is intensification of vital functions of
periphyton, which also improves oxygen conditions. Even if there are in-
stances ol oxygen deficiency in macrophyte thickets, they are local and
temporary. According to our data, the oxygen shortage disappears with the
first rays ot the sun, and by noon to 1 pm one observes oversaturation ot
water with oxygen in thickets of macrophytes. The prevalence ot processes
ot oxygen uptake by higher aquatic plants over its production is observed
tor only 5-fa h (.Figure 1).
Analysis of the curves of output and uptake of oxygen by higher aqua-
tic plants warrants the conclusion that the amount of oxygen discharged
by submerged macrophyte species is more than Ib times greater than the
amount taken up. This can be expressed as 02/1^2 = 3.1 for perfoliate
pondweed and 02/^2 = 16 for hornwort. Hence, it can be concluded that
oxygen given ott during photosynthesis in higher aquatic plants is used
to a significant extent, not only tor plant respiration, but for formation
of oxygen conditions in the water.
One also observes a correlation between photosynthesis of higher aqua-
tic plants and formation ot chemical composition ol water. This link is
manifested by a change in carbonate equilibrium and assimilation of differ-
ent torms ot carbon dioxide by assimilating plants. Bicarbonates are almost
always present in natural waters. When there is intensive uptake by plants,
rree carbon dioxide disappears very rapiaiy from water, and the plants
change to assimilation of bicarbonate carbon dioxide, which again leads to
tormation of carbonates. This is associated with change in either direction
73
-------
CM
o
OT
"OT
o>
£.
200 P
100 -
CM
O
O
o>
E
c
o
C O
OT "Q.
O OT
"rt V
J tr
Q.
-100 -
4 8 12 16 20
Time, hour of day
24
FIGURE 1. Daily exchange ot gases in the hornwort.
in concentration of hydrogen ions (Merezhko 1967, Shiuyan and Merezhko
iy/2), which in turn, atlects the direction of processes of formation ot
water quality. But the biological essence ot photosynthesis in processes
ot formation of quality of natural water is not limited to this.
One of the distinctions ot all photosynthetic organisms is their
capacity to absorb and accumulate in tissues and organs a significant quan-
tity of various biogenic elements. The role of this process in the cycle
ot biogenic elements and in forming water quality is unquestionable. The
figure characterizing accumulation of biogens by aquatic plants fluctuates
over a rather wide range, and it depends on a number ot biotic and abiotic
factors. For this reason, the participation of higher aquatic plants in
processes ot formation ot water quality is permanent.
The direction ot intrawater processes determining iormation ot water
quality depends largely on the period ot turnover of biogenic elements.
The latter, in turn, is determined by the physiological activity ot biogens,
74
-------
intensity of their absorption, and duration ol the period of accumulation
in plants. Nitrogen, phosphorus, potassium, iron, chlorine, and manganese
are utilized the most actively by plants. The same elements are deciding
factors in processes ot formation ot water quality (Merezhko 1973).
Nitrogen and phosphorus accumulate in virtually all plants in the same
amounts (Table 1). All other biogens are absorbed and accumulated by dif-
ferent species ot higher aquatic plants in difterent amounts. There is
particularly distinct demonstration ot ditterences in uptake and accumula-
tion ot potassium, magnesium, calcium, manganese, and iron. They are taken
up the most intensively by the lesser reedmace, bulrush, flowering rush,
common water plantain and perfoliate pondweed, and to a lesser extent by
the common reed.
We should dwell in particular on chlorine uptake by plants. Virtu-
ally all ot the tested plant species absorb and accumulate it in a rather
large quantity. The amount ot chlorine in plants ranges from 1.U to 1.8%.
It is absorbed the most by the common water plantain.
Absorbed biogens are unevenly distributed in the plant (Table 2).
Thus, in August, maximum levels of nitrogen, iron, and manganese were dem-
onstrated in leaves and reproductive organs of the reed. There are lower
levels of these biogens in its stems and rhizomes. The opposite is ob-
served at the end of the vegetation period. At this time, there is efflux
of biogens from leaves and stems into the root system. The latter is ex-
tremely important to processes ot formation of water quality, since a
surplus ot biogenous elements in a reservoir is bound, to some extent,
with plant root systems and is thus excluded from circulation for a rela-
tively long time.
Other autotrophic organisms, including algae, play an analogous role
in processes ot circulation ot biogens in aquatic ecosystems. The gener-
ation period is very short, however, in algae and other autotrophic organ-
isms.
Alter the plants die otf, absorbed minerals and organic matter return
into water and become actively involved in the cycle ol matter. The role
ot this group of autotrophs in formation ot water quality, like that ol
higher aquatic plants, is also manifested by their participation in pro-
cesses ot tormation of biologically valuable water. Probably, algae play
the tirst and foremost role in this process. Their role in discharging
biologically active compounds, including enzymes, into the environment and
in the tormation ol water quality is well-known and proven (Telitchenko
1972, Telitchenko and Telitchenko 1972). Biologically valuable water can
be formed only under the influence of hydrobionts.
Concurrently with uptake ot biogenous elements, autotrophs take up,
accumulate, and transform toxic compounds. The most importance in this
process is given to higher aquatic plants (Merezhko et al. 1975). There
is both passive and active absorption ot these compounds by aquatic macro-
phytes. Passive absorption occurs during transpiration, when one cannot
demonstrate translormation ot these compounds in plants. Consequently,
75
-------
TABLE 1. ACCUMULATION OF SOME EXOGENOUS ELEMENTS IN HIGHER AQUATIC PLANTS, % DRY MATTER
Object of
study
Common reed
Lesser reedmace
-J Bulrush
<7*
Flowering rush
Common water
plantain
Perloliate
pondweed
2.17
+0.04
2.52
+0.06
2.34
+0.00
2.bfa
+0.07
2.oy
+o.oa
2.02
+0.02
p
0.35
+0.02
0.41
+0.01
o.3y
+0.01
0.40
+0.01
0.55
+0.02
0.53
+0.03
K
1.70
+0.03
i.iy
+0.01
2.35
+0.07
4.3b
+o.oy
2. by
+0.0b
2.01
+0.05
Ca
0.38
+0.002
1.07
+0.01
o.t>y
+0.02
1.3(3
+0.0t>
1.20
+0.4
o.y5
+0.02
Mg
0.10
+0.01
0.15
+0.01
0.12
+0.01
0.21
+0.02
O.lb
+0.01
0.33
-rO.02
Na
0.14
+0.02
0.51
+0.02
0.40
+0.03
0.43
+0.01
0.3b
+0.02
0.33
+0.01
Cl
1.36
+0.01
1.20
+0.02
1.5b
+0.04
1.17
+0.02
1.B7
+0.03
1.55
+0.02
Si
1.13
+0.02
0.12
+0.01
0.31
+0.01
0.34
+0.03
0.73
+0.04
0.42
+0.03
Fe,
mg%
0.005
+o.oooy
0.01
+0.001
O.OOb
+0.0007
0.03
+0.002
0.01
+0.0001
O.Ol
+0.001
Mn,
mg/o
0.02
+0.001
O.Ob
+0.001
0.03
+0.0001
o.oy
+0.002
O.Ob
+0.0001
0.03
+0.0001
-------
TABLE 2. CHEMICAL COMPOSITION OF COMMON REED ORGANS (.AUGUST;
Composition
Ash
Compound ,
% dry matter
N
P
K
Ca
Mg
Na
Ci
bi
Compound ,
mg% dry
matter
Fe
Mn
Leaves
6.91
2.48
0.26
U.7U
0.96
0.11
O.Ob
0.57
0.24
14.04
4b.b4
Plant
Stems
3.69
0.65
0.13
0.4b
0.2b
0.05
0.05
0.74
0.17
11.56
ll. 6b
Organs
Panicles
7.96
1.45
0.39
0.99
0.35
0.09
0.03
0.73
2.32
21.75
25. 5b
Rhizomes
b.17
1.46
0.19
2.2b
0.15
0.03
0.15
1.17
6.17
6.07
b.20
passive absorption ol toxic compounds by macrophytes provides, to some
extent, only for migration o± toxic agents trom one environment to another
—in this case, from water to air. however, active absorption ot toxic
compounds by macrophytes leads to their complete or partial detoxification.
This already has a direct bearing on processes of self-purification and
formation ot water quality.
Carbon-labeled DDT and sevin are demonstrable in plants as early as
1 day atter their introduction into the environment. The highest concen-
trations of these agents have been found in the root system of plants and
somewhat lower ones in leaves and stems (Table 3).
The data on presence ot the radioactive tracer in reed organs warrant
the assumption that DDT and sevin probably enter the plant through its
root system. The arrangement of the reed's root system is somewhat ditter-
77
-------
TABLE 3. UPTAKE AND ACCUMULATION OF 14C-LABELED DDT AND SEV1N BY COMMON
REED (Activity of specimens in thousands of pulses/min-g)
Pesticide
DDT
Sevin
Plan
Leaves
2.4 + U.05
2.1 -1- 0.9
:s with watei
Stems
5.b + 0.9
2.8 + 0.7
1
r roots (
Water roots J_
!
540.2 +11.0 |
1
29.7 + 2.5 I
— 1
Plants \
water
Leaves
7.3 + 0.5
1.7 + 0.3
without
roots
Stems
8.4 + 0.3
0.9 + 0.2
ent from that of other plants. Part of its roots lie along the bottom of
the body of water (water roots). These roots are markedly ramitied, and
they absorb the most actively both biogenous elements and toxic compounds
(Table 4, Figure 2). Toxic compounds are accumulated just as actively by
reed rhizomes, and this is important to processes of formation of water
quality. It can be assumed that, because of this, higher aquatic plants
can serve under certain conditions for detoxification of some deleterious
substances in the water environment (Merezhko 1973).
CVJ
E
•
c
200 r
to
9>
jo
•3
Q.
o ,00
T3
C
O
to
C
Q)
I, Stem-1 ike water roots
2. Rhizomai water roots
3. Rhizomai soil roots
Time, min
60
FIGURE 2. Intensity of uptake of 14C of alanine-I—L4C by adventitious
roots of common reed.
78
-------
TABLE 4. DISTRIBUTION OF 14C IN COMMON REED ROOT SYSTEM, THOUS. PULSKS/M1N*G DRY WEIGHT
Pesticide
DDT
HCCH*
Control
Pesticide
Concentration
tag/ £
0.50
1.00
2.00
0.50
0.25
1.00
—
Radioactivity
Rhizomes
0.83 + 0.03
O.bl + 0.02
1.01 -t- 0.04
1.20 + 0.03
1.11 + 0.06
1.53 + 0.07
0.98 + 0.01
of leaves,
%
6.7
6.8
10.6
16.9
15.0
24.3
5.9
Water roots
0.45 + 0.00
0.53 + 0.0
0.93 + 0.01
1.02 + 0.01
0.86 + 0.0
1.45 + 0.05
0.37 + 0.0
Radioactivity
ot leaves,
%
3.6
4.4
9.8
14.3
11.6
23.0
2.2
Soil roots
O.Ofa + 0.01
0.12 + 0.0
0.18 + 0.0
0.09 + 0.0
0.06 + 0.0
0.12 + 0.0
0.70 + 0.0
Radioactivity
of leaves,
%
0.69
'i.O
1.9
1.2
0.8
1.9
0'.4
*H.exachiorocyciohexane
-------
While we wish to stress the roie ol higher aquatic plants, as well as
other autotrophic organisms, in processes of formation of surface water
quality, it must be borne in mind that this roie can be positive only if
plants are promptly harvested from reservoirs (Frantsev 19t>9, 1972). Thus,
according to some simple estimates, more than 200 kg nitrogen, 60 kg phos-
phorus, over 130 kg chlorine, 60 kg calcium, etc., are removed per hectare
when harvesting reed vegetation from water.
In describing processes of formation of quality of natural waters as
related to vital functions of hydrobionts, special attention must be given
to microorganisms and invertebrates with respect to uptake and accumulation
of biogenous elements and organic compounds (Kryuchkova 1972, Vinberg 1972).
Mollusks play a special role in these processes as filters and sedimenta-
tors of suspended substances.
The results ot many years of investigation of Dreissena polymorpha
(Shevtsova and Kharchenko 1981) indicate that this mollusk is a powerful
filter tor suspended particles. On this basis, some authors believe that
the Dreissena plays an important part in biological self-purification
(Stanezykowska 1975, Kondrat'yev 1977, Shevtsova and Kharchenko 1981).
We cannot help but agree with this, but one should not forget that, in
the absence of a consumer of Ureissena, its roie in both selt-puritication
processes and formation ot water quality could be virtually eliminated
and, in some cases (hyperproduction) it could be negative. Nevertheless,
the significance ot this moilusk in processes of formation of water qual-
ity in regions where it develops on a mass scale is quite obvious.
According to the data ot Shevtsova and Kharchenko (.1981), 1 m-^ of Dreis-
sena on the bottom ot a canal filters 405.9 nP of water per year. This
is associated with up to 0.747 kg mineralization and precipitation ot 7.7
kg organic substances in the form of agglutinates. The role of inverte-
brates in processes ot formation of water quality, like that of plants,
is determined by the direction and intensity ot their metabolic processes.
The role of higher aquatic plants has also been proven in processes ot
precipitation of suspended matter (Figure 3) (Merezhko 1978).
Equal importance to formation of water quality is attributed to pro-
cesses of destruction and mineralization of organic matter. It should be
borne in mind, however, that not all organic substances ot autochthonous
and allochthonous origin undergo mineralization. A significant part of
them accumulates in the form ot inoxidizabie tractions, which otten have
a direct bearing on processes of formation of water quality (Telitchenko
1972).
Microtlora and chemical oxidation piay the leading role in destruc-
tion and mineralization ot organic matter. The relative involvement ot
ditterent groups ot microorganisms in destruction and mineralization of
organic matter is not the same, and it is determined by the type 01 reser-
voir and amount ot substance being mineralized (Vinberg 1972).
Thus, the structural and tunctional organization ot circulation ot
matter in aquatic ecosystems is attected by hydrobionts on ditterent tro-
phic levels by the principle ol waste-tree technology.
80
-------
o>
0)
E
TJ
0)
"5 2
I I
I. Common reed
2. Perfoliote pondweed
3.Lesser reedmace
VI VII
Time, month
VIII
FIGURL, ji. Precipitation o± suspended matter on higher aquatic piants
during vegetation period.
In the case ot intensive operation of water resources for recovery of
the basic product of an aquatic ecosystem—biologically valuable water,
it is necessary to control quantitative and qualitative development ot
biohydrocenoses. Consideration of specifics ot function and control of
development ot dominant representatives of producers ana consumers as a
function of time and space provides a possibility for selecting optimum
methods ot formation of water quality and improving the efficiency ot oper-
ating marine ecosystems.
BIBLIOGRAPHY
Frantsev, A.V. 196^. One must help nature! Khimiya I Zhizn'. No. 4.
Frantsev, A.V. lb»72. Some problems of controlling water quality. In;
Teoriya'i praktika biologicheskogo samoochishcheniya zagryaznennykh
vod. Moscow, USSR.
Frobisher, M. l^bb. Osnovy mikrobiologii (.Bases of Microbiology), Moscow,
USSR.
Kondrat'yev, G.P. iy?7. Biofiitration. In. Volgogradskoye vodokhran-
ilishche: naseleniye, bioiogicheskoye prognozirovaniye i samoochish-
cheniye (.The Volgograd Reservoir. Population, Biological Forecasting
and Seit-Puritication) . Saratov.. USSR.
81
-------
Kryuchkova, N.M. 1972. Zooplankton as an agent in self-purification of
water reservoirs. In: eoriya ± praktika biologicheskogo samoochish-
cheniya prirodnyKh vod (Theory and Practice of Biological Self-Purifi-
cation of Naturally Occurring Waters). Moscow, USSR.
Merezhko, A.I. 1967. Ecological and physiological distinctions of blue-
green algae. Abstract of candidatoriai dissertation—biological
sciences. Kiev, USSR.
Merezhko, A.I. 1973. Role of higher aquatic plants in formation of water
quality. Gidrobiol. Zhurn. 9(.4).
Merezhko, A.I. 197b. Ecological ana physiological distinctions of higher
aquatic plants and their role in formation ol water quality. Abstract
o± doctoral dissertation—biological sciences. Kiev, USSR.
Shevtsova, L.V. and T.A. Kharchenko, T.A. 19ttl. Role of Dreissena in
processing suspended organic matter in the North Crimea Canal. Gid-
robiol. Zhurn., 17(5).
Shiyan, P.N. and A.I. Merezhko. 1972. Effect of concentration of hydro-
gen ions on photosynthesis and metabolism ol radioactive carbon in
aquatic plants. Gidrobiol. Zhurn. 8(2). Ibid, Vol. tt, No. 2, 1972.
Stanezykowska, A. 1975. Ecosystem of the Mikolajekie Lake. Regularities
of the Dreissena Polymorpha Poll. (Bivalvia) Occurrence and its
function in the lake. Pol. Arch. Hydrobiol. No. 22.
Telitchenito, M.M. 1972. Possibility of controlling self-purification pro-
cesses using biological methods. In: Teoriya i praktika biologich-
eskogo samoochishcheniya zagryaznennykh vod. Moscow, USSR.
Telitchenko, M.M. and L.A. Telitchenko. 1972. The problems of water
quality and current methods of solving them. In; Teoriya i praktika
biologicheskogo samoochischen iya zagryaznennykh vod. Moscow, USSR.
Vinberg, G.G. 1972. Significance of hydrobiology to solution of water
management problems. In: Teoriya i praktika bioiogicheskogo samoo-
chishcheniya zagryaznennykh vod (Theory and Practice of Biological
Self-Purification of Polluted Water). Moscow, USSR.
Vinberg, G.G. I97b. Biologicheskiye protsessy samoochishcheniya na
zagryaznennom uchastke reki (Biological Self-Purification Processes
in Polluted River Section). Minsk, USSR.
82
-------
C02 EXCRETION AND AMMONIA TOXICITY IN FISHES;
IS THERE A RELATIONSHIP?
by
T.A. Heming1
ABSTRACT
The aquatic toxicity of ammonia is pH-dependent and, therefore, may be
modulated by chemical-biological reactions, such as hydration of excreted
C02> which alter the pH of water adjacent to the gill surface. Effects
of CC>2 excretion on the pH of interlamellar water of rainbow trout (Salmo
gairdneri) were investigated by following changes in the pH of mixed expired
water (pHex) using a stopped-flow apparatus. The pHex immediately postgill
was approximately the same as inspired water pH, but pHex decreased as a
nonlinear function of time downstream of the gill. Half-time ot the
acidification reaction was 240 sec, in excellent agreement with the half-time
of uncatalyzed COyiHCOo" equilibrium. Acetazolamide (0.44 mM) in the bathing
medium had no effect on slow downstream changes in pHex, indicating that
branchial carbonic anhydrase was not available to catalyse hydration of
excreted CC>2' Downstream changes in pHex were abolished by external carbonic
anhydrase (28,080 Wilbur-Anderson units/L). These results suggest that C02
excretion has little, if any, acidifying effect on water adjacent to the gill
surface. Measurements of bulk water pH are probably the best possible
indicators of boundary layer pH. Attempts to explain the effect of pH on
the LC50 concentration for ammonia in terms of a significant difference
between the pH of bulk water and gill water are over-simplified to the point
of being erroneous.
INTRODUCTION
The relative toxicity of ammonia solutions (NH^ + H20 = NH^ + H^O ,
HoO+ = H+) to fishes increases when environmental pH increases (Lloyd and
^Department of Zoology, University of British Columbia, Vancouver, B.C.,
Canada VbT 2Ay. Present address: Aquatic Toxicology Section, Alberta
Environmental Centre, Vegreville, AB, Canada TOB 4LO.
83
-------
Herbert I960, U.S. EPA 1977). This pH-de pendency is related to an increase
in the ratio of non-ionized to ionized ammonia (NH^NH^ ) at higher pH values
(Emerson et al. 1975). Biological membranes, such as the fish gill, are
generally more permeable to non-ionized compounds than to ionized compounds
(Jacobs 1940). If one assumes, on the basis of this difference in gill
permeability, that NH3 is the sole toxic species in ammonia solutions then it
follows that the LC50 concentration for such solutions in terms of the
concentration of NH3 should be independent of pH. Various studies (Lloyd and
Herbert 1960, Tabata 1962, Tomasso et al. 1980, Thurston et al. 1981), how-
ever, have shown that the LC50 concentration for NH3 is positively correlated
with environmental pH, which suggests that NH3 is more toxic to fish at lower
pH values. There are several possible explanations for this trend. Firstly,
the assumption that NH3 is the sole toxic species in ammonia solutions may be
incorrect. NH,+ may also exert a toxic effect that would become more appar-
ent at lower pH values as the relative proportion of NH^ in solution in-
creases. Secondly, past calculations and measurements of the concentration
of NH3 in bulk water may be inappropriate estimates of the concentration of
at the gill surface.
The ratio of NHo:NH/ in solution is dependent upon the solution pH,
temperature and ionic strength (Emerson et al. 1975). Because of heat and
mass transfer across the gill, the chemical characteristics of boundary layer
water at the gill surface must differ from that of bulk water for thermody-
namic reasons (Buysman and Koide 1971). Such differences in water tempera-
ture and ionic strength probably are extremely minute and have negligible
effects on NH-, :NH^ equilibrium. More importantly, several authors (Lloyd
and Herbert 1960, Szumski et al. 1982) have suggested that water adjacent to
the gill surface is significantly more acidic than bulk water due to excre-
tion of C02 and a resultant CC>2 gradient normal to the gill surface. The
proposed differences are as large as 0.31 pH units (Lloyd and Herbert
1960).
For hydration of metabolic C02 (C02 + H20 = H+ + HC03~) to have such an
appreciable effect on boundary layer pH, the hydration reaction must occur
rapidly enough to alter the proton concentration within the interlamellar
transit time of gill water (0.10 - 10.15 sec, Randall et al. 1982). This
is much faster than the uncatalysed rate of C02 hydration (Kern 1960).
Szumski et al. (1982), however, have proposed that this C02 reaction is
catalysed by gill carbonic anhydrase. These authors have developed a model
to estimate the concentration of NH3 at the gill surface taking into account
biological C02 excretion and local water quality. Depending upon water
quality characteristics, the model of Szumski et al. (1982, see their Figure
6) predicts that the U.S. Environmental Protection Agency's criterion for
total ammonia (U.S. EPA 1977) for protection of warm water fishes is 6 to
39 times too restrictive. O^-induced acidification of water adjacent to
the gill, however, does not fully account for observed effects of pH on the
toxicity of NH3 (Roseboom 1983) or other acids/base (Broderius et al. 1977).
Moreover, there is nothing in the experimental literature that allows one
to quantify the effects of C02 excretion on the pH of interlamellar water.
The aim of the present study was to investigate the effects of C02
excretion on the pH of water adjacent to the gill of rainbow trout (Salmo
84
-------
gairdneri) by following changes in the pH of mixed expired water. If CC>2
hydration at the gill surface occurs rapidly enough to have an appreciable
effect on interlamellar water pH, the pH of mixed expired water (pHex) will
be lower than inspired water pH (pHin) and no downstream changes in pHex
will occur. On the other hand, if C02 hydration at the gill surface occurs
slowly, pHex immediately postgill will be approximately equal to pHin and
pHex will decrease slowly downstream of the gill due to continued C02 hydra-
tion. Downstream changes in pHex were followed using a stopped-flow appara-
tus. Effects of external carbonic anhydrase and acetazolamide, a specific
inhibitor of carbonic anhydrase, on downstream changes in pHex were also
examined.
METHODS
Rainbow trout (Salmo gairdneri) weighing between 200 and 400 g were
obtained from Sun Valley Hatchery (Mission, B.C.). The fish were housed
outdoors in tanks supplied with continuous flows of dechlorinated Vancouver
tapwater (.temperature 6 - 10°C) and were fed to satiation daily with a
commercial trout food. Fish were starved for at least 48 hours prior to
and during the experiments.
Fish were anaesthetized (50 - 67 mg/L tricaine methanesulfonate, pH
adjusted to 7.5 with MaHCOj, temperature 8°G) and fitted with catheters for
sampling of mixed expired water. Catheters were constructed from polyethyl-
ene tubes (inner diameter 0.86 mm, length 15 - 20 cm) that had been heat-
sealed at one end. Numerous small holes were made along a 3 to 4 cm length
of the tube near the sealed end. Catheters were sutured to the fish, one
catheter per fish, in such a way that these holes were located immediately
behind the left opercular cover and spanned the entire length of the opercu-
lar cleft, each hole facing anteriorly into the opercular cavity. Following
surgery, each fish was transferred to a 2-L tank supplied with a 1.5 to 2.0
L/min flow of test water that was recirculated from a thermostated C8 +
0.5°C) 120-L reservoir. Test water was renewed at a rate of approximately
25% every 2 days. The test water was a balanced salt solution consisting of
40 mM NaCl, 1.6 mM KC1, 0.47 mM CaCl2, 0.62 mM MgS04, 5.5 mM NaHC03 and
0.95 mM NaH2?04 in dechlorinated Vancouver tapwater. The buffering value
of this test water was about 0.508 mM H+/pH unit in the pH range 7.2 to 8.5.
Experiments were conducted in this test water, rather than in normal fresh-
water, because the presence of dissolved salts greatly improved the stabili-
ty, reproducibility and response time of water pH measurements. Fish were
allowed 7 to 10 days to acclimate to the test water before experiments were
begun. This period of time is more than adequate to allow rainbow trout to
re-establish'ionic, osmotic and respiratory steady-state conditions after
transfer from freshwater to balanced salt solutions (Perry and Heming
1981).
Water pH was measured using a polymer body, sealed reference combina-
tion pH microelectrode (Canlab) and a Radiometer PHM64 meter, and was simul-
taneously recorded using a chart recorder. The pH electrode was housed in
an acrylic plastic measurement chamber (measurement volume 0.1 mL) that was
situated in the test water upstream of the fish (Figure 1). The inlet port
of this chamber communicated directly with the test water when measurements
85
-------
Figure 1. Outline of experimental apparatus used to measure the pH of mixed
expired water of trout. A. pH meter; B. chart recorder; C. peri-
staltic withdrawal pump; D. reference combination pH electrode;
E, F, and G, reservoirs of test water with and without 28,080
Wilbur-Anderson units/L bovine carbonic anhydrase or 0.44 mM
acetazolamide.
of inspired water pH (pHin) were made and was connected to the opercular
catheter when measurements of mixed expired water pH (pHex) were made. In
both situations, the outlet port of the chamber was connected to a peristal-
tic pump that withdrew water past the electrode at a rate of 5 mL/min. At
that pumping rate, water samples required less than 1.8 sec to transit the
opercular catheter and about 1.2 sec to fill the measurement chamber. Re-
sponse time of the pH-measuring system was determined by rapidly switching
the inflow to the measurement chamber between 20 mM phosphate buffers with pH
values of 4.05 and 7.00; half-time of the resultant response was 12.5 sec.
This response time was 10 to 20 times faster than the expected half-time of
the uncatalysed C02:HC03 reaction and so was deemed adequate for my purposes.
A typical experiment consisted of the following measurements. In-
spired water pH was measured under continuous-flow conditions for 20 to 30
minutes. The withdrawal pump was then switched off, suddenly stopping water
flow past the electrode, and pHin was followed under stopped-flow conditions.
The electrode assembly was then connected to the opercular catheter and pHex
was measured under identical continuous-flow and stopped-flow conditions.
y«t!rn£i°V!lrOUSh the flSh tank WaS subsequently switched to a thermostatted
(.B + U.5 C) 4-L reservoir of test water that contained 28,080 Wilbur-Anderson
units/L; bovine carbonic anhydrase, and pHin and pHex were again measured un-
der continuous-flow and stopped-flow conditions. Water flow through the fish
tank was then returned to the control reservoir for a period of 1 to 2 hours.
Finally, water flow through the fish tank was switched to a thermostatted
86
-------
(b + U.5°C) 4-L reservoir of test water that contained 0.44 mM acetazolamide,
and pHin and pHex were again measured under continuous-flow and stopped-flow
conditions.
Observed changes in water pH were quantified by two methods. Firstly,
the value for dpH/dtime was calculated at the point when water flow past the
electrode was first stopped. The initial rate of HT1" generation was calcu-
lated as the product of that value and the buffering value of the test
water. Secondly, the quantity of H+ generated at 1-minute intervals after
electrode flow was stopped was calculated as vBCpHex11 - pHex°), where pHex°
was the pHex under continuous-flow conditions, pHexc was the pHex at known
time intervals after electrode flow was stopped, v was the measurement
volume, and B was the buffering value of the test water. These data were
fitted by the methods of least squares to a double-reciprocal plot, which
allowed calculation of the half-time of the H+ generating reaction.
The data are presented as arithmetic means + their standard errors
(SE). Means were compared statistically using paired Students t-tests. A
5% level of probability (P _< U.05) was adopted as the fiducial level of sig-
nificance .
RESULTS
The pH of inspired water was unaffected by water flow past the pH
electrode, pHin measurements made under continuous-flow and stopped-flow
conditions were stable with time and did not differ from one another. The
pH of mixed expired water, on the other hand, was affected by flow conditions
(continuous versus stopped) in control and acetazolamide studies (Figure 2).
In all control and acetazolamide studies, pHex decreased as a nonlinear
function of time once water flow past the electrode was stopped. Flow con-
ditions had no effect on pHex in carbonic anhydrase studies.
Kinetics of the changes in pHex under stopped-flow conditions are
summarized in Table 1. No statistical differences were present between the
pHex changes in control and acetazolamide studies. In these studies, pHex°
(mixed expired water pH under continuous-flow conditions) did not difter
significantly from pHin. The pHex became progressively more acidic than
pHin, however, once electrode flow was stopped in control and acetazolamide
studies. The initial rate of acidification in these studies averaged 0.943
nmol H^/min, the half-time of the acidification reaction averaged 240 sec,
and the final equilibrium pHex averaged 0.12 units lower than pHin. Similar
changes in pHex were not seen in carbonic anhydrase studies. In carbonic
anhydrase studies, pHex° was consistently lower than pHin by about 0.12
units, and pHex remained stable at that value under stopped-flow conditions.
DISCUSSION
Mixed expired water of rainbow trout was not in pH equilibrium when
it exited the opercular cavity. The pH of mixed expired water immediately
upon exiting the opercular cavity was approximately that of inspired water.
Mixed expired water became progressively more acidic than inspired water
downstream of the opercular cavity at a relatively slow rate. The half-time
87
-------
7.64 r
7.62
7.60
7.58
756
CONTROL
7.54
7.62
7.60
7.58
ACETAZOLAMIDE
7.56
7.54
7.59
CARBONIC
ANHYDRASE
7.57
0
4
TIME, min
8
Figure 2. Typical changes in the pH of mixed expired water ot a single
rainbow trout at 8°C, alter water flow past the pH electrode was
stopped at time 0. The trout was exposed sequentially to control
test water, test water containing 28 ,UbO Wilbur-Anderson units/L
bovine carbonic anhydrase, (90 minutes rinse in control test
water, and test water containing 0.44 mM acetazolamide.
88
-------
TABLE 1. KINETICS OF CHANGES IN THE pH OF MIXED EXPIRED WATER OF RAINBOW
TROUT AT 8°C, MEASURED UNDER STOPPED-FLOW CONDITIONS. VALUES
ARE MEANS + SE (N = 4). pHin, INSPIRED WATER pH; pHex°, EXPIRED
WATER pH UNDER CONTINUOUS-FLOW CONDITIONS; NS, NO SIGNIFICANT
TRENDS IN pH WERE OBSERVED.
pHin
pHex
Final Initial rate of Half-time of
equilibrium H~*~ generation H+ generation
pH (nmol/min) (sec)
Control
7.64 + 0.03
7.63 + 0.06
7.52 + 0.07
O.fa9 + 0.04
270 + 32
Acetazolamide
7.67 + 0.04
7.66 + 0.01
7.56 + 0.02
1.00 + 0.07
210 + 13
Carbonic anyhdrase
7.69 + 0.04
7.57 + 0.03
7.58 + 0.04
0.05 + 0.02
NS
of the acidification reaction was approximately 240 sec at 8°C and an
average pHin of 7.66. This value is in excellent agreement with the half-
time for the uncatalysed equilibrium of C02 and HC03~ (Table 2). Downstream
acidification of mixed expired water indicates that the C02:HC03 equilibrium
was dominated by hydration of excreted C02. This is consistent with the
model of Cameron and Polhemus (1974) for C02 excretion in fish, which pre-
dicts that molecular C02 accounts for 95% of total carbon dioxide excretion
while HCOo" efflux accounts for only the remaining 5%. Assuming the observed
acidification reaction obeyed first-order kinetics and given the one-for-one
stoichiometry between C02 and H4" in the C02 hydration reaction, one calculates
from the initial rate of H generation, the reaction half-time and the C02
solubility coefficient tor water (Boutilier et al. 1984) that the change in
Pco2 (partial pressure of C02) between inspired and mixed expired water of
89
-------
TABLE 2. HALF-TIME (SEC) FOR UNCATALYSED C02:HC03 EQUILIBRIUM IN AQUEOUS
SOLUTIONS AT INFINITE DILUTION, DETERMINED BY LINEAR INTERPOLATION
OF DATA OF KERN (J. Chem. Educ., 37 :14-23, 1960) AT 0 and 25°C.
Temperature (°C)
pH
4.00
5.00
6.00
7.00
7.64
8.00
9.00
10.00
11.00
0
1.3
10
74
240
—
300
200
47
5.4
8
0.94
7.2
53
172
198
212
139
32
3.7
25
0.18
1.3
9.1
26
—
25
8.7
1.1
0.12
rainbow trout at 8°C was about 0.7 torr*. This is a reasonable value consid-
ering the arterial-venous difference in blood PcO2 across the trout gill (0.5-
1.0 torr, Heming 1984), C0£ production by gill tissue and the diluting effect
*According to first-order kinetics, reaction velocity (R, mol/L/sec) is re-
lated to substrate concentration (c, mol/L) as
R = kc,
where k is the rate constant. The half-time (t, sec) of such a reaction is
calculated as
t = In 2 / k.
Given values for R and t, these equations can be rearranged to calculate the
initial substrate concentration as
c = Rt / In 2.
The concentration of C02 determined in this way was converted to its partial
pressure_by division with the appropriate solubility coefficient. Since (X>2
and HC03 in inspired water were in complete equilibrium and, therefore, did
not contribute to the observed changes in pHex, the above calculations do not
require that the Pco2 of inspired water be known. Rather, these calculations
yield the increase in Pco£ as water passes over the gills.
90
-------
of inspired water that bypasses the secondary gill lamellae. This suggests
that the majority of excreted CC>2 was hydrated at an uncatalysed rate down-
stream of the gill; one would have expected an unreasonably low estimate of
the difference in inspired:expired water Pco2 if excreted C02 was hydrated in
a partially catalysed equilibrium at the gill surtace. The finding that ex-
ternal acetazolamide had no effect on downstream changes in pHex indicates
that branchial carbonic anhydrase was not available to catalyse C02:HC03
interconversions within the boundary layer at the gill surface. This is
consistent with the results of histochemical surveys of carbonic anhydrase in
branchial and opercular epithelia of fish (Haswell et al. 1980, Dimberg et
al. 1981, Lacy 1983). These studies have shown that the enzyme is localized
in the cytoplasm and nuclei of epithelial cells; no evidence of carbonic
anhydrase on the apical surface of branchial or opercular epithelia has been
found. Results of the present carbonic anhydrase studies provide further
evidence that slow changes observed in pHex were due to continued hydration
of CC>2 downstream of the opercular cavity. Mixed expired water attained pH
equilibrium prior to exiting the opercular cavity only when external carbonic
anhydrase was available to catalyse CC>2 :HC03 interconversions within the
interlamellar and opercular spaces. Table 2 demonstrates that, in the ab-
sence of such an external catalyst, water temperature and/or pH would have
to be outside the lethal limits of many aquatic species for uncatalysed
CC^iHCOo reactions to occur rapidly enough to have an appreciable effect on
water pH within the interlamellar transit time for gill water (0.10 - 0.15
sec, Randall et al. 1982).
What is the pE of water adjacent to the gill surface? One would pre-
dict, given the slow rate of uncatalysed C02 reactions, that CX>2 excretion
would have little, if any, effect on boundary layer pH. Of course, C02 and
HC03 are not the only acia-base relevant molecules transfered across the
gills. Movements of NHo, NH^+, H+ and OH~, amongst others, across the
gills will also influence boundary layer pH. The net effect of all of these
transfer processes on boundary layer pH will depend upon the direction and
magnitude of the fluxes, the interlamellar transit time of gill water, the
uncatalysed rates of COoiHCO^" and NH^NH.^ interconversions, the buffering
capacity of the ambient water, and the volume of the boundary layer. As
such, boundary layer pH must be a function of ambient water quality (tempera-
ture, pH, ionic strength and composition), ventilation rate and volume, and
the mode of ventilation (buccal pumping versus ram ventilation). The mass
ratio for branchial excretion of C02:NH3:H+:HC03~ in rainbow trout at rest in
freshwater is about 19:3:3:1, assuming (a) a ratio for C02:HC03 excretion of
19:1 (Cameron and Polhemus 1974), (b) a ratio for total C02 production to
total ammonia production of 8:1 at a steady state respiratory quotient of 0^8
(Heisler 1984), (c) a'one-for-one branchial exchange of Na+/H+ and HC03~/C1~
with the rate of Na+ influx being about three times that of Cl~ (Eddy 1982),
and (d) that simple diffusion of NH3 accounts for ammonia excretion (Cameron
and Heisler 1983). Protonation of excreted NH3 (NH3 + H+ = NH^+) within the
boundary layer will ameliorate acidifying effects of H excretion and C02 hy-
dration; this protonation reaction occurs rapidly enough to be complete with-
in the interlamellar transit time of gill water. The precise effect of gill
fluxes on boundary layer pH will be extremely temporal, however, since the
magnitude and even the direction of these fluxes are sensitive to many biotic
and abiotic variables including exercise, respiratory acidosis, and water
91
-------
ammonia levels and salinity (Eddy 1982, Heisler 1984, Cameron and Heisler
1983). Moreover, enhanced interlamellar turbulence during periods of rapid
or vigorous buccal pumping may eradicate gill boundary layers and directly
expose the gill surface to bulk water. Considering the sensitivity limits of
research pH meters (+ U.001 units) and the accuracy of precision calibration
buffers (+ 0.005 units), measurement of bulk water pH is probably the best
possible Tndicator of the pH of water adjacent to the gill surface. Models
that attempt to explain the observed effects of pH on the aquatic toxicity of
compounds, such as ammonia (Lloyd and Herbert I960, Szumski et al. 1982), in
terms of a large C02~induced acidification of the gill boundary layer must be
regarded as being over-simplified to the point of being erroneous.
ACKNOWLEDGMENTS
Dr. D.J. Randall is thanked for his intellectual and material support
of this work. Financial support was provided by a University of British
Columbia Graduate Fellowship to the author. The work described in this paper
was not funded by the U.S. Environmental Protection Agency and therefore the
contents do not necessarily reflect the views of the Agency and no official
endorsement should be inferred.
REFERENCES
Broderius, S.J., L.L. Smith Jr., and D.T. Lind. 1977. Relative toxicity of
free cyanide and dissolved sulfide forms to the fathead minnow
(Pimephales promelas). J. Fish Res. Board Can. 34, 2323-2332.
Boutilier, R.G., T.A. Heming, and G.K. Iwama. 1984. Physiochemical param-
eters for use in fish respiratory physiology, pg. 403-430. In Fish
physiology, volume X, in W.S. Hoar and D.J. Randall, editors. Academic
Press, New York, New York.
Buysman, J.R., and F.T. Koide. 1971. Ion concentration profile normal to
cell membrane. J. Theor. Biol. 32: 1-23.
Cameron, J.N., and N. Heisler. 1983. Studies of ammonia in rainbow trout:
Physico-chemical parameters, acid-base behaviour and respiratory clear-
ance. J. Exp. Biol. 105, 107-125.
Cameron, J.N., and J.A. Polhemus. 1974. Theory of C02 exchange in trout
gills. J. Exp. Biol. 60, 183-194.
Dimberg, K., L.B. Hoglund, P.G. Knutsson, and Y. Ridderstrale. 1981. Histo-
chemical localization of carbonic anhydrase in gill lamellae from
young salmon (Salmo salar L.) adapted to fresh and saltwater. Acta.
Physiol. Scand. 112, 218-220.
Eddy, F.B. 1982. Osmotic and ionic regulation in captive fish with partic-
ular reference to salmonids. Comp. Biochem. Physiol. 73B, p. 125-
141.
92
-------
Emerson, K., R.C. Russo, R.G. Lund, and R.V- Thurston. 1975= Aqueous
ammonia equilibrium calculations: Effect of pH and temperature. J.
Fish. Res. Board Can. 32: 2379-2383.
Haswell, M.S., D.J. Randall, and S.F. Perry. 1980. Fish gill carbonic
anhydrase: Acid-base regulation or salt transport? Amer. J. Physiol.
238, R240-R245.
Heming, T.A. 1984. The role of fish erythrocytes in transport and excre-
tion of carbon dioxide. Ph.D. thesis, University of British Columbia,
Vancouver, B.C.
Heisler, N. 1982. Transepithelial ion transfer processes as mechanisms
for fish acid-base regulation in hypercapnia and lactacidosis. Can.
J. Zool. 60: 1108-1122.
Heisler, N. 1984. Acid base regulation in fishes, pg. 315-407. In Fish
physiology, volume X, W.S. Hoar and D.J. Randall, editors. Academic
Press, New York, New York.
Jacobs, M.H. 1940. Some aspects of cell permeability to weak electrolytes.
Cold Spring Harbour Sym. Quant. Biol. 8, 30-39.
Kern, D.M. i960. The hydration of carbon dioxide. J. Chem. Educ. 37,
14-23.
Lacy, E.R. 1983. Histochemical and biochemical studies of carbonic anhy-
drase in the opercular epithelium of the euryhaline teleost, Fundulus
heteroclitus. Amer. J. Anatomy 166, 19-39.
Lloyd, R., and D.W.M. Herbert. 1960. The influence of carbon dioxide on the
toxicity of un-ionized ammonia to rainbow trout (Salmo gairdneri
Richardson). Ann. Appl. Biol. 48, 399-404.
Perry, S.F., and T.A. Heming. 1981. Blood ionic and acid-base status in
rainbow trout (Salmo gairdneri) following rapid transfer from fresh-
water to seawater: Effect of pseudobranch denervation. Can. J. Zool.
59, 1126-1132.
Randall, D.J., S.F. Perry, and T.A. Heming. 1982. Gas transfer and acid/
base regulation in salmonids. Comp. Biochem. Physiol. 73B, 93-103.
Roseboom, D.P. 1983. Discussion of: Evaluation of EPA un-ionized ammonia
toxicity criteria. J. Water Poll. Control Fed. 55: 420-421.
Szumski, D.S., D.A. Barton, H.D. Putnam, and R.C. Polta. 1982. Evaluation
of EPA un-ionized ammonia toxicity criteria. J. Water Poll. Control
Fed. 54, 281-291.
Tabata, K. 1962. Toxicity of ammonia to aquatic animals with reference to
the effect of pH and carbon dioxide. Bull. Tokai Reg. Fish Res. Lab.
34: 67-74 (Transl. from Japanese).
93
-------
Thurston, R.V., R.C. Russo, and G.A. Vinogradov. 1981. Ammonia toxicity to
fishes. Effect on pH on the toxicity of the un-ionized ammonia
species. Environ. Sci. Techn. 15, 837-840.
Tomasso, J.R. , C.A. Goudie, B.A. Simco, and K.B. Davis. 1980. Effects of
environmental pH and calcium on ammonia toxicity in channel catfish.
Trans. Amer. Fish. Soc. 109: 229-234.
U.S. EPA (U.S. Environmental Protection Agency). 1977. Quality criteria for
water. Office of Water and Hazardous Materials, U.S. Environmental
Protection Agency, Washington, B.C.
94
-------
USE OF INDICATORS OF FUNCTIONAL STATE OF BIOTA
IN BIOLOGICAL MONITORING OF SURFACE WATERS
by
V.A, Bryzgalo, L.S. Fedorova, T.A. Khoruzhaya,
L.S. Kosmenko and L.P. Sokolova^
ABSTRACT
Discussed is the use of various functional characteristics of aquatic
ecosystems in the biomonitoring of surface waters. A study of the system
as a whole with an assessment of the interaction between the elements of
the system and between the system and the surrounding environment has pro-
vided an opportunity to assess the intensity and direction of the basic
processes involved in the transformation of matter and energy in the com-
munities. These processes include photosynthesis, fixation of C02, con-
sumption of organic matter, etc.
Particular attention was given to evaluating the information pro-
vided by the individual physiological and biochemical indicators lor the
state of the biota. Research was carried out on aquatic objects under
varying conditions ot thropogenic stress. In addition the main criteria
were established for assessing the trend and intensity of production-
destruction processes.
INTRODUCTION AND DISCUSSION
Biological monitoring of surface waters is done in the USSR by the
hydrobiologicai network, ot the National Service for Observation and
Monitoring of the Environment (OGSNK) on three levels; impact, regional,
and background. At the present time, methods used for this purpose are
based on traditional hydrobiologicai observations ot systematic identifica-
tion ot species, as well as saprobiological analysis (Anon. lS»bJJ)- Studies
are made ot total number of organisms, biomass, number of species, marker
significance, etc.
1Hydrochemical Institute, Rustov-on-Don, USSR
95
-------
Such an approach makes it possible to assess the entire ecosystem
ot a body ot water on the basis ot data concerning the condition of phy-
toplankton, zooplankton, bacteriopiankton, periphyton, zoobenthos, and
macrophytes. Saprobic indexes, Vudiviss's biotic index and several other
procedures, which permit comparing both different parts of reservoirs to
one another and different water systems, are used tor a formalized evalua-
tion of these hydrobiocenoses.
All these methods and approaches make it possible to describe the
structure ot an ecosystem, but lurnish virtually no information about its
function. Among the functional methods recommended tor use in the OGSNK
network, we should include those for determination ol primary production
and destruction of organic matter ot phytoplankton, bacterioplankton and
zooplankton, as well as pigments of phytoplankton, that are used by dif-
ferent network hydrobiological laboratories.
One of the reasons tor insufticient use of functional indicators in
the system ot biological labeling of surtace water is the highly dynamic
nature of aquatic communities, as well as the wide spatial and time heter-
ogeneity of their characteristics, and significant seasonal fluctuations
in development of the different elements ot an ecosystem. At the same
time, the need to use the functional approach tor biological monitoring
of surface waters is growing increasingly pressing and it is dictated by
a number of factors. The very definition ot the concept of "norm" as a
zone ot optimum function of a system requires use primarily ot functional
characteristics to assess the ecological state ot water systems. Function-
al parameters are also important from the standpoint ot the basic water
protection tasks, because they make it possible to turn to evaluation ot
the process ot tormation ot water quality, the capacity ot hydrobiocenoses
to maintain stability ot quality characteristics of water.
The speed of functional methods, as compared to structural ones, is
one ot the appreciable advantages ot the former particularly in effective
monitoring (.Israel1
Successful use ot functional parameters can be made only if distinct
correlations are established between changes in them and changes in the
corresponding parameters ot the environment caused by exogenous factors.
However, even with this formulation of the problem, a number ot difficul-
ties arise, lor example, it is quite obvious that direct measurement ot
results of the dose-ettect reaction in a water system, particularly in
studies ot plankton communities, is impossible at the present time. As
tor model experiments, it is hardly valid to extrapolate data to natural
water ecosystems due to the absence or more or less distinct ecological
criteria ot similarity. Finally, use ot the probabilistic statistical
approach requires comprehensive and detailed analysis ot aquatic ecosys-
tems, which is hardly expedient tor economic considerations.
One ot the routes ot solving this problem is to tind methods of
evaluating the results ot population (.or even community) activity, pro-
vided there is substantial reduction 01 spatial resolution ot information
This becomes feasible when studies are planned in such a way that the
96
-------
water system is considered, on the one hand, as a unique ecosystem and,
on the other hand, as separate ecosystems in the form ot contrasting water
masses characterized by specific physicochemicai conditions ana populations,
on the assumption that there is relatively little interaction between them.
It is possible to implement such an approach if we proceed from the
following. Different physicochemicai conditions determine formation ot
different ecosystems. This permits selection of representatives for a
given type of study ot a region, which reduces spatial resolution of the
study and increases time resolution. And, using structural and functional
parameters, one can gain an idea about the extent and direction ot ,main
processes of transformation of matter and energy in aquatic ecosystems,
and assess the informativeness of biochemical parameters in biomonitoring
of water quality.
Properly validated conclusions about the ecological state of water
systems are difficult to derive because ot the lack of unified and ade-
quate methodological approaches tor quantitative description of processes
that are of interest from the standpoint ot trophodynamxcs. To date,
quite a few tunctional characteristics ot state ot marine ecosystems have
been proposed and developed. A significant number of studies have dealt
with investigation ot different tunctionai parameters ot phytopiankton.
As tar back as the turn ot the century, hydrobioiogists began to view
phytopiankton as the main producer of organic matter—and the only one in
vast expanses of ocean—on the basis of which the entire diversity of
aquatic life is tormed. This was the reason for the heightened interest
in investigating not only the species composition of phytopiankton and
its quantitative distribution, but various parameters characterizing its
function. We are referring, first of all, to output', i.e., rate of produc-
tion ot organic matter by phytopiankton. To determine this, one generally
uses physiological methods based on comparing the results ot determining
intensity of photosynthesis (production) and respiration (.destruction) of
communities according to oxygen balance (Federov iy?9).
In spite of the fact that the oxygen method used in most cases has
some flaws, as well as of the dissatisfaction of many researchers with
obtained results, it became possible to formulate tasks such as determin-
ation of productivity ot autotrophic and heterotrophic organisms. It is
expressly in trying to assess their contribution to overall production of
an aquatic ecosystem that more sensitive radioactive carbon methods were
developed for measuring phytopiankton output. Functional methods, which
characterize the state and activity of phytopiankton and, in particular,
different parameters or intensity ot photosynthesis, underwent the great-
est development. We should mention here tiuorescence and luminescence
techniques, which are highly sensitive and permit use ot rather simple
equipment and instruments (Sirenko i9b3).
We should also include among the iunctional methods those used to
investigate processes of transformation and utilization ot organic mat-
ter by hydrobionts. The mechanisms ot these processes and conditions
that make them efficient are being studied intensively, although we are
97
-------
stiil tar from comprehending the key aspects of the problem (Pursons et
al. 1982). Of definite interest are studies of heterotrophic activity
o± aquatic cenoses both to assess their role in forming the qualitative
composition of natural waters and as an informative integral indicator
of the conditon and stablity of aquatic ecosystems.
Interest in studying bacteriopiankton as related to water systems
differing in trophism was prompted by the substantial contribution of this
community to processes of utilization of organic matter. Measurement of
uptake of glucose and comparing it to the total quantity of ATP-containing
bacteria, concentration of chlorophyll a, changes in pH, and levels of oxy-
gen, carbon dioxide, nitrogen and phosphorus made it possible to define the
trophic status of 21 lakes in a mountain province of New Zealand (.Spenser
197tt). Apparently, one can define the time and space parameters of distri-
bution ot pollutants according to rate of uptake of organic matter. Thus,
Pearl and Goldman (.1972) obtained encouraging results using the rate of
acetate uptake as an indicator of the condition of the lake's water masses.
In recent years, there has been increasing discussion ot the possi-
bility ot using kinetic parameters of uptake ot organic matter by natural
waters based on examination oi dynamics of assimilation of 1\) from water
samples with addition of labeled organic compound (Wright and Hobble 19bb).
By using structural and functional methods of studying aquatic com-
munities, one can gain an idea about the extent and direction of basic
processes in aquatic ecosystems. This is possible thanks to the fact that,
by studying the dynamics of functional parameters against the background
of change in structure of communities, we obtain a description of proces-
ses—in particular, of photosynthesis, fixation of CO^ , uptake of organic
matter, etc. It is expedient to use various coetficients—tor example,
the ratio of destruction to biomass (.D/B) and production to biomass (.P/B).
Such estimates bring us to a description of extent of thermodynamic
organization ot the ecosystem or hydrobiocenosis in question (Odum 1975).
Some results of using these coetficients tor biomonitoring were recently
summarized in an article by Vernichenko and Starko (.1902). It should be
stressed that a more or less distinct range of changes in these coeffi-
cients has not yet been defined due to insufficient development o± both
the methods of measuring primary production and destruction, and biomass.
In the OGSNK network, the P/D coefficient is used to determine the intens-
ity ot production-destruction processes. The range ot changes in this in-
dicator has been studied more, it is used, for example, by specialists at
the IGB ot the Ukrainian Academy ot Sciences in a plan for a system oi in-
tegrated evaluation ot quality of surface waters (.Zhukinskiy et al. 1976).
Primary production and destruction, as well as biomass levels, are
very dynamic and subject to substantial time and space changes, they de-
pend on seasonal fluctuations, hydrological conditions in the reservoir
and other factors. All this makes it considerably more complicated to
use them for biomonitoring. The validity ot this statement is well-illus-
trated in the studies we conducted on reservoirs differing in trophism
using the above-described approaches.
98
-------
The purpose ot the study included demonstration ot the main structural
and functional indicators ol the state ot the biota, which are related to
formation oi quality ot natural waters as related to ditterent anthropo-
genic loads, as well as determination ot the main criteria tor assessing
the direction and intensity ot production and destruction processes.
The studies were conducted in a "conditionally pure" eutrophic reser-
voir, with predominantly natural eutrophication, and an oligotrophic one
with an anthropogenic load, during a period ot natural alternation tor the
reservoir oi dominant groups ol algae, starting with diatoms and ending
with blue-green algae, covering the entire period ot development ot the
latter (from U to 9b£ ot total phytoplankton biomass).
In the period ot phytoplankton development we studied, there were
three peak, levels ot total biomass, one ot which was attributable to de-
velopment ot diatoms (7 July), and the two others to biomass ot blue-green
algae (25 July and 15-17 August). The natural replacement of dominants
(diatoms—blue-green algae) occurs in active competition between green and
blue-green algae, and it is notable for drastic fluctuations in biomass
level and photosynthetic activity per unit phytoplankton biomass (P/B coef-
ficient) (Table 1). Development ot blue-green algae is associated with
relatively uniform increase in total biomass and production of phytoplank-
ton. The P/B coefficient also rises, and during the period of drastic
dominance of blue-green algae, it changes by no more than 4 times. The
changes in primary production are characterized by two maximums, one ot
which corresponds to the period of diatom "blooming" (7 July), and the
second, which is more significant, corresponds to the period of marked
"blooming" of blue-green algae (15-19 August).
Biologically, production is the most important function in the struc-
ture of the community, when the structure is stable, the functional param-
eters, in particular P/B, remain constant. Any change in structure ot
the community leads to change in production rate. We found that the in-
crease in phytoplankton biomass is taster at the early stage ot "blooming"
than the increase in production. For this reason, there is decline ot
photosyntnetic activity per unit phytoplankton biomass (P/B coefficient).
While we can judge the status ol an ecosystem as a whole from the P/B
ratio, the ratio of phytoplankton production to its population size (P/S)
enables us to turn to a description of the condition ot cells ot individ-
ual specimens, which is determined primarily by presence of nutrients in
their environment.
Evidently, this ratio could indicate, indirectly, the purpose ot phy-
toplankton activity: intensive division and increase in number or accumu-
lation ot nutrients in the cell, i.e, increase of its biomass. Indeed,
during the period of diatom and green algae blooming, when activity is
directed toward retaining their populations, there is a maximal Pp/S ratio
(U.2-U.b) '1U~3. Under favorable ambient conditions, the dominant blue-
green algae expend a significant part ot the newly formed organic matter
on cell division and increase in population size. The Pp/S ratio is then
smaller by a factor of 10 (0.2-0.3)-11T3 -
99
-------
TABLE 1. CHARACTERISTICS OF PHYTOPLANKTON (p) ACTIVITY ACCORDING TO SOME PARAMETERS
o
o
Date of
taking
sample
7 Jul
9 Jul
11 Jul
14 Jul
25 Jul
30 Jul
10 Aug
12 Aug
15 Aug
iy Aug
22 Aug
Total number (S) , Biomass
thousands (Bp)
cells/m& mg C/£
2
1
2
Ib
14
50
b2
95
493
706
b
.0
.96
.00
.70
.55
.60
.60
.00
.00
.00
.50
0
0
.145
.047
0.045
0
1
0
0
1
3
5
0
.470
.100
.790
.760
.140
.600
.060
.160
Production
W p) >
mg C/(£ day)
1
0
0
0
0
1
2
2
29
13
1
.60
.30
.47
.49
.32
.25
.3b
.bb
.30
.10
.16
w
mg C/
(mg -day)
11
6
10
1
0
1
3
2
b
2
7
.0
.0
.0
.0
.3
.6
.0
.5
.0
.6
.0
Pp/S, Heterotrophic
yg C/ pyruvate uptake,
(cells -day) yg C/(&-day)
O.bO
0.15
0.24
0.03
0.02
0.02
0.03
0.03
0.06
0.02
0.13
10-3
10-3
10-3
10-3
10-3
10-3
10-3
10-3
10-3
10-3
10-3
—
0.16
0
—
0.12
4.74
0
5.3
14.5
0
0
The rate ol pyruvate uptake by blue-green algae reached 5-15 yg C/(£-day).
-------
Hydrochemicai studies revealed that, during the period of prevalence
ot blue-green algae, levels of mineral forms of nitrogen and phosphorus
were favorable for their development, and the C:N:P ratio was optimum.
However, the results of measuring heterotrophic activity ot algae (assay
of phytoplankton uptake of iZtC-pyruvate) revealed that, even without a
shortage ot mineral nutrients, blue-green algae at the stage of maximum
photosynthetic activity are capable ot utilizing readily available organic
substances.
As we know, intensity ot metabolism, i.e, ratio ot respiration to bio-
mass (R/B) determines the extent or thermodynamic organization (Wright
and Hobble 1966), which could characterize the reaction of the ecosystem
to an exogenous factor. Most often, one uses the P/B ratio, rather than
R/B, and the former expresses the increment ot production per unit initial
or mean value. P/B, the coetticient or its reciprocal B/P-coefticient of
turnover of biomass—are considered by many researchers to be the most
informative parameters characterizing the production capacities of a given
species or entire community.
During the period of intensive phytoplankton blooming, bacterioplank-
ton reacts to its environment by increasing its functional activity (Table
2). Bactenoplanfcton production reaches a maximum during the period of
maximum development of blue-green algae (15-19 August). However, it we
calculate averaged activity per cell (ratio of P to total population size) ,
the result will indicate that, in spite of the large number of cells,
bacterioplankton is in a relatively depressed state during the period ot
dominance of blue-green algae. Activity drops from 6-10~* yg C/day to
U.7b-lu~^ yg C/aay.
The ratio ot heterotrophic ^C-pyruvate heterotrophic uptake to bac-
terial biomass production, as determined from dark assimilation of CC>2, can
also serve as an indirect indicator of the state of bacterioplankton. The
higher the activity or bacterial cells, the closer the ratio is to 1.
These studies enabled us' to conclude that neither total population
size nor biomass, nor even such functional parameters as production of
phytoplankton and bacterioplankton could yield sufficient information about
the state of the natural population. It is only by turning to a descrip-
tion ot activity of the population unit (one cell) that we come close to
characterizing the actual response of organisms to a change in their en-
vironment, in both the case of its natural change at the early stage of the
eutrophication process and in the case of "severe" anthropogenic pollution.
The seeming "wellbeing" ol the ecosystem, which can be assessed by the
structural organization of the cenosis, conceals appreciable changes in
condition of organisms in different trophic chains and the population as a
whole (specific activity, rate of biomass turnover, doubling time, intens-
ity of respiration, coefficient of energy metabolism, etc.). These changes
in status of the ecosystem can be detected only if studies are made with
sutticient spatial and time resolution.
Approaches based on evaluation ot ditferent physiological and bio-
chemical parameters are among promising methods that have not yet been
101
-------
TABLE 2. CHARACTERISTICS OF BACTERIOPLANKTON ACTIVITY ACCORDING TO SOME FUNCTIONAL PARAMETERS
Date 01
Taking Sample
9 Jul
11 Jul
14 Jul
25 Jul
29 Jul
j
I 30 Jul
31 Jul
10 Aug
12 Aug
15 Aug
19 Aug
21 Aug
22 Aug
Total number (S) ,
millions
ceils/m&
0.54
0.79
2.01
0.94
13.20
17.60
4.60
—
—
—
—
1.05
2.90
Production (yg
determined by
i'*C-carbonate
tPh)
2
8
8
4
40
10
17
b
35
44
89
5
12
C/ (i 'day} , as
uptake of
•^C-pyruvate
(PE)
1.31
4.68
8.17
2.87
17.30
13.48
4.37
1.50
8.42
32.80
22.46
0.82
3.80
P£/Ph
0.65
0.58
1.02
0.71
0.43
1.35
0.26
0.26
0.24
0.75
0.25
0.16
0.32
Specific production
Pg/S,
yg C/( cells -day)
2.4 10~9
6.0 10~y
4.0 10~y
3.0 10~y
1.3 10^
0.76 10~9
0.95 10~y
—
—
—
—
0.78 10~y
1.30 10~y
-------
sutficientiy developed, which retlect the functional state ot aquatic eco-
systems. In our opinion, biomonitoring methods involving use ot parameters
of enzymatic activity of both individual hydrobionts and populations, and
even aquatic communities merit special attention. Proceeding from the key
theses of biology to the effect that a reaction to environmental changes
is manifested primarily by change in rate and nature of enzymatic proces-
ses, it can be assumed a priori that the level of enzyme activity could be
used to assess water quality. The response of an enzyme system character-
izes the state or aquatic organisms when there are fluctuations in physi-
cochemical parameters ot the environment, while the functional reserves
retlect the degree of adaptation to them.
This approach is not without some flaws. On the one hand, some en-
zymatic reactions are "universal" in a sense and provide for basic meta-
bolic -processes over a rather wide range of fluctuations ot conditions in
the aquatic environment. On the other hand, some biochemical parameters
are "specific," i.e., they reflect a response to only a specific type of
pollution.
Thus, use ot biochemical parameters, including enzymatic ones, re-
quires a validated choice of methods that are adequate to the purpose or
observation, as well as determination of the range ot changes in a given
enzymatic tunction in hydrobionts that are not exposed to a tactor, i.e.,
definition ot the reaction "norm." The difficulty of this problem as it
applies to biomonitoring ot the quality of natural waters has already
been stressed (Braginskiy 19bl).
We used the biorhythmological approach in studies of enzymatic
activity ot naturally occurring populations ot moliusks exposed to the
chronic eftect ot sewage from the paper and pulp industry. It was estab-
lished that, in gastropod moliusks taken from different biohydrocenoses,
the curves of circadian activity of enzymes—esterases, alkaline phospha-
tase, succinate, malate and lactate dehydrogenases—differed. Depending
on the role of the tested enzymatic systems, we observed prevalence ot
"diurnal" or "nocturnal" activity.
To compare these data in moliusks taken from different sections of
the reservoir, we calculated the "parameter of circadian adaptability"
(.PCAJ for each enzyme. We found that PGA was positive for virtually all
enzymes in moliusks inhabiting a conditionally pure zone. In moliusks
taken from a part ot the reservoir subject to the chronic effect ot sew-
age, PGA was negative tor ail enzymes. Moliusks living in zones differ-
ing in degree of pollution diltered not only in biorhythms ot enzymatic
activity, but response to an additional burden— brief change in tempera-
ture. Thus, with exposure to change in temperature, the moliusks ot the
"conditionally pure zone" showed some decline in enzyme activity, while
the nature or the circadian rhythm did not change appreciably. In mol-
iusks from the zone subject to the ettect ot sewage, the circadian rhythms
of all enzymes studied changed appreciably, and their activity dropped
drastically. Moreover, we observed much taster restoration ot initial
rhythms in moliusks 01 the control zone than those exposed to the influ-
ence ot sewage.
-------
Thus, it should be concluded that organization ot the circadian rhythm
ot enzyme activity in naturally occurring populations ot mollusks can be
used as a characteristic ot functional state of an organism, its capacity
for adaptation under altered physicochemical environmental conditions, as
well as extent ot effect of environmental factors. Apparently, consid-
ering the significant fluctuations of hydrobionts1 biochemical parameters,
which are a typical feature and reflect adaptation to the aquatic environ-
ment, the biorhythmological approach can be recommended tor studies on
ditterent levels ot organization—cellular, tissular, organic, organism,
population, etc.
At the present stage ot development of scientific methodological
procedures for evaluating the ecological status ot the naturally occurring
water environment, an integrated approach is needed for analysis of suit-
ability of physiological and biochemical parameters tor biomonitoring of
surface waters. It should be noted that most ot the methods based on
physiological and biochemical parameters are reterable to biological tests,
and they were developed primarily to assess the toxicity of sewage or
chemical compounds (when setting maximum permissible concentrations).
Trial ot these methods in naturally occurring waters requires concurrent
investigation not only ot hydrobiological parameters characterizing the
structure ot the water ecosystem, but ot hydrologicai-hydrochemical para-
meters. The latter is particularly necessary when investigating physio-
logical and biochemical parameters on the level of populations and aquatic
communities, due to the complexity and diversity ot links between both
individual trophic levels and between the biota and environment, as well
as the very dynamic nature of these links.
Viewing enzymatic activity as a function of interaction ot an organism
with the environment, which changes in the time parameter of development
of the ecosystem and trophodynamic processes, we analyzed the informative-
ness of enzymatic parameters ot naturally occurring seston communities in
waters subject to natural eutrophication and anthropogenic influences. In
these studies we used the following parameters: species composition, total
number and biomass ot bacterioplankton, phytopiankton and zooplankton,
intensity of production and destruction processes, activity ot alkaline
phosphatase and total esterases ot seston, levels ot biogenous elements
(total, nitrate, and nitrite nitrogen, ammonia nitrogen, total and mineral
phosphorus), and levels of oxygen, hydrocarbonates and pH.
it was established that the absolute levels ot enzyme activity are low
in an aquatic ecosystem that is not exposed to any signiticant anthopogenic
ractors, while potential activity (Ap) is 3 to 5 times higher than the
level observed (Ao). Distinct seasonal dynamics ot changes in enzyme activ-
ity can be observed. In situations related to drastic changes in environ-
mental conditions (pollution, algal "blooming," etc.), absolute enzyme
activity levels rise and there is change in seasonal cycle ot enzyme activi-
ty, as well as in Ap^. Determination was made of the range ot fluctua-
tions ol enzyme activity during the period of blooming ot blue-green algae
with exposure to sewage trom the paper and^pulp industry, as well as with
salt pollution. It was shown that the reaction of seston with regard to
enzyme activity during natural eutrophication ot the reservoir is described
1 n/l
-------
by a curve that is synchronized with the developmental phases of bacteria
and algae, as well as changes in proportion of biogenous elements contained
in the environment. With decline of mineral phosphorus level in water,
there is increase in esterase activity ana the enzyme activity ratio (es-
terase;alkaline phosphatase) increases by a factor of 10. In the case of
anthropogenic influences, enzyme activity of seston increases drastically,
which mak.es it possible to differentiate water zones in the region of waste
dumping according to extent of pollution.
In conclusion, we should like to emphasize that there are several
tasks that are of prime importance in the problem of biomonitoring »t
natural waters. As a result of hydrobiological observations pursued in
the system of state supervision, information for a period of many years
has been accumulated. This information is presently being systematized in
order to find the parameters that reflect the basic trends of change,
mainly in structural organization of aquatic communities as a function of
extent and nature of anthropogenic burden on a reservoir's ecosystem as a
whole.
Analysis of the accumulated information, with consideration of re-
sults of scientific studies dealing with biomonitoring of natural waters,
including functional parameters, will enable us to undertake development
of the methodology for assessing the ecological status of water systems
and developing the optimum variant of routine and efficient biomonitor-
ing.
Introduction of new methods, which permit evaluation of functional
changes in the ecosystem of a body of water, offers vast opportunities
for both efficient monitoring and forecasting the state of the ecosystem
and water quality. Unification and standardization of both approaches
and procedures, as well as methods as a whole, is a mandatory prerequisite
tor successful use of these methods.
BIBLIOGRAPHY
Anon. 19b3. Rukovodstvo po metodam gidrobioiogicheskogo analiza pover-
khnostnykh vod i donnykh otiozheniy (Manual or Methods for Hydrobiologi-
cal Analysis of Surface Waters and Bottom Deposits). Leningrad, USSR.
239 p.
Braginskiy, L.P. l9bl. Theoretical aspects of the problem of normal and
pathology in aquatic ecotoxicology. In Teoreticheskiye voprosy vodnoy
toksikologii (.Theoretical Problems of Aquatic Toxicology). Leningrad,
USSR. pp. 29-40.
Fedorov, V.D. 1979. 0 metodakh izucheniya fitoplanktona i yego aktivnosti
(.Methods of Studying Phytoplankton and Its Activity). Moscow, USSR.
Ibb p.
Israel", Yu.A. 1983. Ekologiya i kontrol" sostoyaniya prirodnoy sredy
(Ecology and Inspection of the Environment). Leningrad, USSR. 376 p.
-------
Odum, J. 1975. Fundamentals o± ecology. Moscow, USSR. 740 p.
Pearl, H.W. and C.R. Goldman. 1972. Heterotrophic assays in the detec-
tion of water masses of Lake Tahoe, California. Limnol. and Oceanogr.
I7(l):145-14tt.
Pursons, T.R., M. Tachakashi, and B. Chargrive. 19tt2. Biological ocean-
ography. Moscow, USSR. 432 p.
Sirenko, A.A. 19fc>3. Curve of vertical distribution of chriorophyll in
a body of water as an indicator tor integral evaluation of correlation
between production and destruction processes. In Obobshchennyye
pokazateli kachestva vod—83. Prakticheskiye voproshy biotestirovaniya
i bioindikatsii. Tezisy doki. (Summaries of Papers Delivered on
Overall Indicators of water Quality—1963. Practical Problems of
Biological Testing and Biological Monitoring). Chernogolovka, USSR.
pp. 124-126.
Spenser, M.J. 197b. The trophic status of 21 lakes of Mountain Province
in New Zealand. N.Z. Mar. and Freshwater Res. 12(4):415-427.
Vernichenko, A.A. and N.V. Starko. 19b2. Prospects of using functional
indicators in the system of monitoring aquatic ecosystems. In
Kontrol' kachestva prirodnykh i stochnykh vod (Monitoring Quality of
Natural and Run-Off Waters). Kharkov, USSR. pp. 14-20.
Wright, R.T. and J.G. Hobble. 1966. Use of glucose and acetate by bac-
teria and algae in aquatic ecosystems. Ecology. 47(3):447-464.
Zhukinskiy, V.N., O.P- Oksiyuk, G.N. Oleynik, and S.I. Kosheleva. 1978.
Plan tor a system of integral evaluation fo quality of surface waters,
Vodnyye Resursy. (.3):83~93.
106
-------
LONG RANGE TRANSPORT OF TOXIC ORGANIC CONTAMINANTS
TO THE NORTH AMERICAN GREAT LAKES
by
rain1
2
W.R. Swain1
M.D. Mullin
J.C. Filkins2
ABSTRACT
Studies of persistent organic contaminants made for the Upper Lakes
Reference Group of the International Joint Commission in 1974-1976 indicated
increased levels of organic contamination in the flesh of fish taken from the
vicinity of Isle Royale in Lake Superior. These findings led to a prelimin-
ary study of polychlorinated biphenyl (PCB) compounds in atmospheric precipi-
tation deposited in Siskiwit Lake on Isle Royale.
The subsequent studies reported here support the earlier findings but
show significant shifts in observed concentrations. Initially, these alter-
ed values were thought to be the result of increasing environmental concen-
trations of PCBs. The use of refined analytical procedures suggests, however,
that the apparent changes are a function of the co-occurrence of derivatives
of the pesticide Toxaphene within the analytical spectrum of the PCB frac-
tion. A comparison of the older data from Siskiwit Lake with those acquired
from contemporary analytical procedures suggests that it is likely that^de-
rivatives of Toxaphene may have been enumerated with PCB peaks, thus being
inadvertently reported as total PCB. Based on results obtained with high
resolution capillary chromatography, apparent increases in concentrations of
PCB compounds in 1980 packed column chromatography samples are undoubtedly a
function of historic analytical limitation, rather than an absolute change in
environmental concentrations of the PCB compounds.
This paper has been reviewed in accordance with the U.S. Environmental
Protection Agency's peer and administrative review policies and approved for
presentation and publication.
l-Vakgroep Aquatische Oecologie, Universiteit van Amsterdam, Kruislaan 320,
1098 SM Amsterdam, The Netherlands
2Large Lakes Research Station, U.S. Environmental Protection Agency, Grosse
He, Michigan 48138 USA
107
-------
INTRODUCTION
In 1974, studies of the accumulation of selected persistent organic
residues in the fish species of the Lake Superior ecosystem were initiated
under the auspices of the International Joint Commission's Upper Lakes Refer-
ence Group. The findings of these studies indicated substantially increased
levels of halogenated organic substances in Lake Superior fish near Isle Royal,
particularly in lean lake trout, Salvelinus namaycush, and in fat lake trout,
Salvelinus namaycush siscowet.
As a check on contaminant levels in Lake Superior fish, a control site,
Siskiwit Lake on Isle Royale, was established. This deep, cold, oligotrophic
lake contained indigenous species of fish similar to those in Lake Superior,
allowing direct comparison. The island, and hence the lake, was remote from
inhabited areas, thus being well removed from the industrial and cultural
influences of man's activities. Surprisingly, however, the values for several
organic contaminants in the flesh of fish from Siskiwit Lake were significant-
ly higher than corresponding values in fish from Lake Superior. Polychlorin-
ated biphenyl (PCS) compounds were nearly double the Lake Superior mean value,
and p,p-DDE showed a more than 10-fold increase in Siskiwit Lake. These find-
ings led to a preliminary study of the transport of PCB compounds in atmos-
pheric precipitation. For a more complete summary of this work, the reader
is referred to Swain (1978).
The present study was undertaken as a continuation of the earlier effort,
with particular interest in the changes in residue levels in ttje fish popula-
tion of this landlocked, remote island lake.
GEOGRAPHICAL SETTING
Isle Royale is a rock-strewn conglomerate of a central island mass with'
associated irregular reefs. It lies 27.4 km from the closest shoreward mar-
gin of Lake Superior at Thunder Cape, Ontario, but it is more than 50 km
to Thunder Bay, Ontario, the nearest major population center (Figure 1).
Isle Royale (Figure 2) has an extreme length of 70.8 km, lying on a
southwest-northeast axis. Its maximum width, 14.5 km, occurs near the southwest
end of the island. The highest point in the island is Mount Desor, 242 m
above Lake Superior. The entire island structure, including its associated
reefs and islets, constitutes Isle Royale National Park, established in 1940.
Efforts from that date have been made to preserve the island's native wilder-
ness character. As a result, the only means of access from one part of the
island to another is by means of foot trails. Siskiwit Lake, accessible only
in this fashion, provided a remote aquatic ecosystem relatively undisturbed
by human activity.
Siskiwit Lake lies 0.6 km inland from Lake Superior in a depression of
Precambrian Middle Keweenawan flow of the Portage Lakes Volcanics. The entire
watershed of the lake also lies in bedrock of this type. The long axis of
Siskiwit lake follows the same direction as the long axis of Isle Royale,
i.e., northeast-southwest.
10Q
-------
48* -
46°
_L
90-
88*
1
SAULTSTE. MARIE.
ONTARIO
10O KILOMETERS
SAULTSTE. MAHIE.
MICHIGAN
_1_
49»
47°
91° 89° 87°
Figure 1. Geographic area of study in Lake Superior.
85"
The length of Siskiwit Lake is 11.10 km, the maximum breadth is 2.25 km,
and the mean breadth is 1.40 km (Huber, 1975). The lake area is 15.6 km ,
exclusive of islands. Shoreline development, calculated from the expression
D = L/A , is typical of subcircular eliptiform lakes, having a value of 2.66.
Trie elevation of Siskiwit Lake above the mean Lake Superior level is 17.37 m,
indicating no direct input from Superior to Siskiwit Lake. Siskiwit Lake
drains by a single small stream to Lake Superior, the Siskiwit River. The
maximum depth of Siskiwit Lake is 36.58 m.
METHODOLOGY
Sample collection from the remote sites at Siskiwit Lake were made in 1980
from a canoe that was portaged to the site from the Superior shoreline. Fish
samples were taken by means of a 350-foot (106.68 m) section of 5.72-cm gill net
that was backpacked into Siskiwit Lake and set overnight at a depth of 24.4 to
46.5 m at the location shown in Figure 2. For all collections, the catch was
exclusively lake trout (Salvelinus namaycush), whitefish (Coregonus culpeafor-
mis), and bowfin (Amia calva). In each case, only lake trout and whitefish
were retained, carried out in a clean mesh bag, eviscerated, iced, and after
transport to a populated area, immediately frozen in dry ice for transport to
the laboratory.
Snow samples were taken from the undisturbed surface of the lake and were
transported to the laboratory in sealed teflon containers. Because of prevail-
ing ambient temperatures, further refrigeration was not necessary.
109
-------
— ta-20
LEGEND
o Fish Sample *
• Water Sample
Snow Sample
89° IS'
earoo-
88-301
Figure 2. Location of sampling sites for fish, water, and snow in the vicinity
Siskiwit Lake, Isle Royale.
of
-------
Samples of atmospheric precipitation in the Lake Huron area were taken
using both event samplers and bulk collectors lined with thin teflon bags.
These bags were suspended within the frame of the bulk collector in such a
manner as to minimize evaporation losses from the collected sample. In all but
one case, samples were taken from the Michigan side of Lake Huron. The single
exception was" an August sample taken from the Canadian shore line of Lake
Huron below the Bruce Peninsula.
ANALYTICAL PROCEDURES
In an effort to minimize analytical error, samples of fish were submitted
to three separate laboratories for analysis. These analytical efforts includ-
ed the Columbia National Fisheries Laboratory of the U.S. Fish and Wildlife
Service at Columbia, Missouri; the Cranbrook Institute of Sciences at Bloom-
field -Hills, Michigan, and the U.S. Environmental Protection Agency's Large
Lakes Research Station at Grosse He, Michigan.
Initial packed column gas chromatography studies followed accepted analy-
tical procedure. A detailed description is presented by Swain (1978). When it
became apparent that PCB congeners were co-occurring with compounds of Toxa-
phene, however, subsequent analytical efforts utilized high resolution capillary
column chromatography. The procedures associated with this type of analysis
are summarized below.
High resolution fused silica capillary gas chromatography was performed
on a VARIAN Model 3700 gas chromatograph equipped with a "%I electron capture
detector. A 50-m fused silica column (0.2 mm i.d.) coated with SE-54 (Hewlett-
Packard) was used to separate the PCB congeners and the Toxaphene-like com-
pounds. The oven temperature was programmed at a rate of 1.0 °C min from
100 to 240 °C. The injector and detector temperatures were 270 and 330 °C,
respectively. Sample volume, 6.0 yl, was injected by using an automatic sam-
pler with splitting in the injector (10:1 split ratio, vented from 0.75 to
1.75 min.) The hydrogen carrier gas was held at a constant pressure of 2.25
kg cm~2 to give the optimized velocity (y) at 100 °C of 45 cm s~ .
ANALYSIS
The interpretation .of the chromatograms was complicated by the presence,
in all samples, of PCBs as well as the polychlorinated camphene (Toxaphene-
like) components. The high resolution chromatography used permitted the eval-
uation of both groups of compounds together without additional separation pro-
cedures. There were more than 110 peaks in the PCB reference standard and 39
of the larger peaks in technical Toxaphene were used in the Toxaphene refer-
ence standard. Of these approximately 150 different compounds, only about a
dozen co-eluted and, of this number, fewer than six were significant. Where
co-eluting compounds existed, the following procedure was used: if other PCB
peaks in the same time portion of the chromatogram were present and other
Toxaphene-like peaks were absent, the peak in question was assumed to be a
PCB; conversely, if other PCB peaks in the same time portion of the chromato-
gram were absent and other Toxaphene-like peaks were present, then the peak in
question was assumed to be a Toxaphene-like compound; if both PCB and Toxa-
phene-like peaks were present, then the peak was excluded from both the PCB
and Toxaphene-like calculations.
Ill
-------
PCB data were determined by comparing the relative retention times (RRT)
of the individual peaks (octachloronaphthalene as internal standard = 1.000)
in the chromatogram to those generated by a mixture of Aroclors 1016, 1254 and
1260. If the RRT were within a narrow window of ca 0.0004 RRT units and the
height of the peak was proportionate to other PCB peaks in the same time por-
tion of the chromatogram, then the PCB assignment was made and the amount cal-
culated was based on a comparison of the height of the peak in the standard
for which the concentration was known) to the height of the peak in the sample.
Toxaphene-like data were calculated differently because of the lack of
individual compound standards against which to compare the sample. The 39
peaks in the Toxaphene reference standard all were assigned the same value—
the total concentration of the Toxaphene standard—and relative response
factors (RRF) for each peak were calculated. Summing these concentrations and
dividing by the number of peaks yields the concentration of the solution.
Toxaphene-like compounds from the samples were determined in a similar man-
ner. After making qualitative assignments based on the same +/- 0.0004 RRT
units exclusion window as used in the PCB determination and comparing the
peak height to other Toxaphene-like peaks in the same time portion of the
chromatogram to ensure that they were proportionate to the same peaks in the
standard, a concentration was calculated for each peak by comparing its height
to the height of the same peak in the standard. These concentrations were
summed and the sum divided by the number of Toxaphene-like peaks identified in
the sample. If fewer than 20 peaks were identified, then the qualitative iden-
tification of the peaks as Toxaphene-like was uncertain and reported as ques-
tionable.
It should be noted that at the time of preparation of this report, the
precise identification of Toxaphene in biologic tissue is still a matter of
some analytical discussion. While two laboratories participating in the ana-
lytical portion of this study unhesitatingly refer to the material observed as
"Toxaphene", a third group is concerned that all of the peaks observed are not
entirely co-incidental with pure standards of Toxaphene. This group observes
that the sample appears to be "environmentally weathered" as a result of passage
through biologic systems. It is their opinion that the substances should be
labeled as "chlorinated turpines/camphenes" or as "Toxaphene-like" compounds,
suggesting this analytical uncertainty. The latter, more conservative approach
has been used for this report.
RESULTS AND DISCUSSION
In 1976, a series of four fresh snow samples was collected, melted, ex-
tracted, and analyzed from the Isle Royale site (Figure 2). PCB compounds
were expressed as an aggregate of Aroclor 1254, the reference mixture that
Veith et al. (1977) reported as most resembling mixtures observed in Lake
Superior. This Aroclor mixture was used as an index of airborne movements of
persistent organic compounds, and the levels observed at Siskiwit Lake were
compared with values at Duluth, Minnesota. The mean value for PCB compounds
observed in precipitation samples from the Duluth, Minnesota, metropolitan
area was 50.0 ng/1, whereas the mean value from Siskiwit Lake was nearly five
times greater, 230.0 ng/1 (Swain, 1978). Similar results were reported by
Murphy (1976) and by Murphy and Rzeszutko (1978) in the Lake Michigan area,
112
-------
comparing precipitation in Chicago, Illinois, against that of Beaver Island in
Northern Lake, Michigan.
Strachan and Huneault (1979), working in the Province of Ontario adjacent
to the Great Lakes found 86 percent of 50 rain samples to contain an average of
21 ng/1 PCB, with an.observed maximum of 120 ng/1 PCB. These workers found an
average of 26 ng/1 in 14 rainfall samples adjacent to Lake Superior and a mean
of 38 ng/1 in 4 samples of snow from the Lake Superior area. In their later
summary of airborne contaminants to the Great Lakes ecosystem, Eisenreich et
al. (1981) report a range of 10 to 100 ng/1 PCB in precipitation over the
Great Lakes with an average content of 30 ng/1. These workers project total
annual PCB deposition of airborne trace organic compounds to Lake Superior at
9.8 metric tons per annum (MTA), 6.9 MTA for Lake Michigan, 7.2 MTA for Lake
Huron, 3.1 MTA for Lake Erie, and 2.3 MTA for Lake Ontario.
All of these studies suggested continuing inputs of PCBs to the Great
Lakes ecosystem via the atmosphere, even though usage of these materials in
North America was drastically curtailed in the early 1970s. For this reason,
it was thought that sampling of the remote Siskiwit Lake site should be con-
tinued because the only source of contaminants to this system is apparently
the atmosphere. Lake trout from Siskiwit Lake were selected for this study
because of their well known predisposition to accumulate contaminants and
because these fish occur both in Siskiwit Lake and in the upper Great Lakes,
thus enabling direct comparison. Further, continued use of this species would
permit extension of previous data from the lake and reflect changes in levels
of contamination with time.
In the 1976 data set (Swain, 1978), a single composite of two lake trout
was reported. These fish averaged 54.5 cm in body length, weighed an average
of 1.6 kg, and contained 3.5 percent lipid. Two additional unreported compos-
ites of two lake trout each—1.5 kg and 1.7 kg average weight, and.3.2 and 3.8
percent lipid, respectively—contained 0.9 and 1.3 mg/kg PCB on a wet weight
basis.
For comparison with these data, 18 lake trout were taken in 1980 and
assigned to five composites as indicated in Table 1. These composites were
forwarded to three participating laboratories with a request to run standard
packed column gas chromatography analysis for PCBs with methods analogous and
comparable with those used in the 1976 study. Laboratory 1 reported individu-
al values for each of the five composites. These values were 3.26, 2.27, 1.57,
1.22, and 2.17 mg/kg PCB on a wet weight basis for composites 1 through 5,
respectively. The reported mean value was 2.1 mg/kg PCB. Laboratories 2 and
3 homogenized the five individual composite samples and performed the analysis
on a single grand composite of 18 fish. Laboratory 2 reported an observed level
of 3.85 mg/kg PCB, and Laboratory 3 reported 4.3 mg/kg PCB. This yielded an
average value of 3.42 mg/kg PCB for the analysis of the three laboratories.
Although there was a rather large range between the three laboratories, both
the individual values reported and the mean of the analyses were substantially
higher than the average of 1.13 mg/kg observed in 1976.
Because of the individual variation among laboratories and because of
the availability of newer, more sensitive methodologies, a blind duplicate
113
-------
TABLE 1. CHARACTERISTICS OF LAKE TROUT FROM SISKIWIT LAKE,
ISLE ROYALE (LAKE SUPERIOR), 1980
Composite
sample
1
2
3
4
5
Sample
mean
Fish
per
composite
1
4
4
5
4
3.6
Length (cm)
Min.
71.0
54.0
54.0
35.0
52.0
53.2
Max.
71.0
58.0
58.0
53.0
57.0
59.4
Avg.
71.0
55.8
55.0
46.2
53.2
56.24
Avg.
weight
(kg)
4.682
1.784
1.598
1.045
1.412
2.104
Percent
lipid
10.41
9.82
5.85
9.06
7.20
8.486
sample was submitted to each of the three laboratories with a request for anal-
ysis of the identical material using high resolution capillary column gas
chromatography. The results, shown in Table 2, were surprising. Not only
were the analyses in excellent agreement, but the mean of the analyses was
well below the levels observed on Siskiwit Lake in 1976, and nearly five times
lower than the packed column analyses.
To verify these results, the five Siskiwit Lake lake trout were reanalyzed
as individual composites, and in addition, two fillets of lake trout were includ-
ed from outer Saginaw Bay in Lake Huron. A 1.16-kg fillet was submitted taken
from a 62-dm male lake trout containing 13.7 percent lipid, and the second was
a 1.9-kg fillet containing 12.5 percent lipid taken from a 75-cm female lake
trout. The results of the comparative analyses are shown in Table 3. In all
cases, the values determined by packed column chromatography exceeded those
observed with high resolution capillary column chromatography by an average
of more than three times.
TABLE 2. ANALYSIS OF PCB IN SISKIWIT LAKE COM-
POSITE LAKE TROUT SAMPLES BY PACKED
COLUMN AS COMPARED WITH HIGH RESOLU-
TION CAPILLARY COLUMN GAS CHROMATOG-
RAPHY
Sample means Sample means
total PCB total PCB
Laboratory packed column capillary column
identification (mg/kg) (mg/kg)
1 2.10 0.70
2 3.85 0.72
3 4.30 0.70
114
-------
TABLE 3. ANALYSIS OF PCB BY PACKED COLUMN AS COMPARED WITH
HIGH RESOLUTION CAPILLARY COLUMN GAS CHROMATOGRAPHY
Sample means Sample means
total PCB total PCB
packed column capillary column
Sample (mg/kg) (mg/kg)
Lake trout fillet, male
Saginaw Bay-Lake Huron
Lake trout fillet, female
Saginaw Bay-Lake Huron
Lake trout-Siskiwit Lake
composite number 1
Lake trout-Siskiwit Lake
composite number 2
Lake trout-Siskiwit Lake
composite number 3
Lake trout-Siskiwit Lake
composite number 4
Lake trout-Siskiwit Lake
composite number 5
11.08
5.16
3.26
2.27
1.57
1.22
2.17
5.70
2.30
1.35
0.91
0.57
0.35
0.41*
0.39*
Sample means 3.82 1.5
*Duplicate analysis.
To answer the obvious question of the reason for the differences, an
intense study of the individual chromatograms was made. It quickly became
apparent that a number of non-PCB peaks co-eluted in the same portion of the
gas chromatographic spectrum as the PCB congeners. Careful resolution and sep-
aration with capillary column chromatography and comparison with known stan-
dards revealed that these additional peaks were derived from chlorinated tur-
pine/camphene products characteristic of technical Toxaphene. Although, as
noted previously, it was not possible to match all peaks with pure standards
of Toxaphene because of biologic "weathering" of environmental samples, a con-
sistent match of more than 20 peaks was normally achieved.
Aliquots of the same samples reported in Table 3 were reanalyzed against
Toxaphene reference standards. The results of this effort are reported in
Table 4. In all cases, the absolute quantities of Toxaphene-like materials
exceeded the levels of PCBs observed by a considerable margin; an average dif-
ference of more than 5-fold was seen. A relatively consistent ratio of Toxa-
phene-like compounds to PCB congeners was observed. Only in one case, compos-
ite number 4, did this ratio become excessively large. In this case, the
ratio observed was nearly double the mean of the other samples (10.6 as compar-
ed with 5.42).
For comparison with the values observed in lake trout, two water samples
were taken at 1 meter of depth at the site on Siskiwit Lake indicated in Figure
2. Two additional samples were taken in Lake Superior form the 50-meter strata
of water approximately 5 km southwest of Menagerie Light on Siskiwit Bay of
115
-------
TABLE 4. PCB AND TOXAPHENE-LIKE COMPOUNDS (TOX) OBSERVED IN LAKE TROUT
Sample location
Fillet, Saginaw
Lake Huron
Fillet, Saginaw
Lake Huron
Siskiwit Lake
composite ho.
Siskiwit Lake
composite no.
Siskiwit Lake
composite no.
Siskiwit Lake
composite no.
Siskiwit Lake
composite no.
Bay
Bay
1
2
3
4
5
Total
PCB
(mg/kg)
5.7
2.3
1.35
0.91
0.57
0.35
0.41*
0.39*
Number
of peaks
utilized
for PCB
26
26
25
25
25
25
25
25
Total
TOX
(mg/kg)
25.84
12.05
6.37
5.14
3.01
3.71
2.57
2.45
Number
of peaks
utilized
for TOX
28
29
31
28
39
27
30
30
Ratio .
TOX: PCB
4.53
5.24
4.72
5.65
5.28
10.6
6.27
6.28
Sample means
1.49
25.25
7.64
29
6.07
*Duplicate analysis.
Isle Royale. The means of these samples are compared in Table 5 with the aver-
age values of samples collected in Lake Huron by staff of the Cranbrook Institute
of Science, and jointly analyzed by this group and the second author. Again, a
consistent ratio of Toxaphene-like material to PCB congeners was observed for
all samples except those of Siskiwit Lake. Apparently because of its relative
size, state of oligotrophy, and the relatively higher inputs of PCBs received,
the ratio in this lake was reduced to 1.72, as compared with an average of 3.45
for the other analyses.
To assess the quantity of PCB compounds in atmospheric precipitation, a
series of bulk collectors and event samplers had been set out along the Lake
Huron shoreline at Hoeft State Park, Michigan, Sturgeon Point, Michigan, and
Tawas State Park, Michigan. Additionally, one confirmatory station was estab-
lished on the Canadian shoreline of Lake Huron below the Bruce Peninsula, and
one station was established below Lake Huron at Grosse lie, Michigan. When
the results of the studies of lake water and lake trout became available, the
precipitation samples were also analyzed for PCB and Toxaphene-like compounds.
The results of these analyses are shown in Table 6. Again in every case,
the values observed for the Toxaphene-like substances exceeded those observed
for PCBs. The mean of the ratios observed for all bulk precipitation samples,
by far the majority of the data, was 3.24, which compares well with the mean
ratio found for open Lake Huron water, 3.51, and with that of Lake Superior,
3.16. As was previously noted, the data points that are derived from peaks
of Toxaphene-like substances of less than 20 in number should be regarded with
116
-------
caution. Although the data have been included for continuity, four such points
exist in the data set—one from Hoeft State Park, two from Sturgeon Point, and
one from the Tawas State Park sampling station. At best, the values for the
Toxaphene-like substances associated with these samples should be regarded as
a qualitative estimation of the amount of these substances present.
Although these data are currently being more completely Analyzed, it is
apparent that Toxaphene-like components make up a currently significant portion
of the contaminant burden of Upper Great Lakes water, fish, and precipitation
inputs. It may even be necessary to reconsider some of the historic data on
PCBs in the Great Lakes in light of this new information to evaluate' the pos-
sible positive interference of Toxaphene-like compounds in the analysis of
these samples.
In keeping with the original purpose of this sequence of studies, a com-
parison of several other 1980 pesticide levels in lake trout from Siskiwit Lake
also was made with the earlier data. The results of the associated analyses
are shown in Table 7. In all cases for which comparative data were available,
significant reductions in levels of pesticides were observed. Compounds show-
ing such declines included the alpha isomer of benzenehexachloride (BHC, also
known as hexachlorocyclohexane), heptachlor epoxide, p,p'-DDE, and p-p'-DDD. A
group of new materials also was observed among which were the gamma isomer of
BHC (also known as hexachlorocyclohexane, or Lindane), Toxaphene-like compounds,
and the family of substances related to chlordane, including oxychlordane, trans-
chlordane, cis-chlordane, and cis-nonachlor.
Finally, using a 1-kg homogenate of the 18 lake trout comprising composites
1 through 5 of Tables 1 through 4, negative ionization mass spectroscopy analy-
ses were made for polychlorinated dibenzo-_p_-dioxins (PCDDs) and polychlorinated
dibenzofurans (PCDFs). The results of these analyses are shown in Table 8.
TABLE 5. MEAN VALUES OF ANALYSES OF PCS AND TOXAPHENE-LIKE COM-
POUNDS (TOX) IN SELECTED WATER SAMPLES
Numbers of
Total Total peaks in
PCB TOX TOX Ratio
Location (ng/1) (ng/1) analysis PCB: TOX
Lake Huron
Southern Lake Huron 0.39 1.50 42 3.85
Middle Lake Huron 0.46 2.14 39 4.65
Northern Lake Huron 0.28 1.23 35 4.39
North Channel 0.77 1.60 25 2.08
Georgian Bay 0.57 1.48 36 2.60
Lake Superior 0.31 0.98 38 3.16
Adjacent to Isle
Royale
Siskiwit Lake, Isle 1.28 2.2 37 1.72
Royale
117
-------
TABLE 6. PCS AND TOXAPHENE-LIKE COMPOUNDS (TOX) OBSERVED IN SAMPLES-
EVENT AND BULK PRECIPITATION, LAKE HURON
Location
Hoeft State
Park
sampling
station
Sturgeon
Point
sampling
station
Canadian
shore
station
Grosse lie
sampling
station
Tawas State
Park
sampling
station
Mean of all
samples
Month Sample
sample volume
taken (1)
1
2
6
8
9
10
12
1
2
5
6
6*
7
8
8*
9
10
10*
11
8
6
8
10
6
8
9
10
11
12
12.65
5.86
11.76
15.25
36.05
9.85
12.10
5.53
7.83
31.05
30.20
5.65
30.44
30.91
5.79
36.75
9.00
1.65
18.30
45.72
28.36
83.73
10.44
24.72
24.29
42.15
12.26
24.24
7.40
21.39
Number of
Total peaks in
PCB PCB
(ng/1) analysis
7.7
11.1
19.9
1.8
2.7
22.9
108.0
87.0
6.9
4.3
7.3
8.6
1.4
2.2
21.0
7.6
19.9
31.6
9.2
2.6
5.0
8.1
22.9
8.5
3.5
5.3
30.0
4.9
46.0
17.9
28
60
40
40
60
49
71
65
54
40
47
59
36
37
42
40
54
65
41
35
43
56
46
42
42
47
51
38
62
47.9
Number of
Total peaks in
TOX TOX Ratio
(ng/1) analysis TOX: PCB
36.3
21.0
55.9
14.5
5.0
51.4
112.0
99.0
16.7
17.7
13.1
17.8
5.58
14.6
37.0
16.3
51.5
40.4
21.3
26.7
16.2
18.9
74.2
15.6
19.9
11.7
52.1
9.7
80.0
33.5
15
31
26
38
31
32
26
22
25
23
19
38
33
34
41
34
15
29
28
22
39
39
32
37
49
36
17
30
21
29.7
4.73
1.88
2.80
8.13
1.89
2.24
1.03
1.14
2.42
4.11
1.79
2.07
3.96
6.59
1.75
2.15
2.59
1.28
2.32
10.38
3.26
2.35
3.24
1.84
5.73
2.21
1.74
1.99
1.74
3.08
Mean of bulk
precip. samp,
Mean of event
precip. samp.
23.36 17.5 47.1
4.36 20.4 55.3
33.7
29.0
31.73 36.0
3.24
1.7
*Event samples,
118
-------
TABLE 7- COMPARISON OF RESIDUE-FORMING ORGANIC SUBSTANCES
OBSERVED IN LAKE TROUT (EVISCERATED WHOLE FISH)
FROM SISKIWIT LAKE, 1974-1976 and 1980
Concentration (ng/g, ppb)
Compound
aBHC
yBHC
HCB
Heptachlor epoxide
Oxychlordane
trans- chlordane
cis-chlordane
p,p'-DDE
p,p'-DDD
cis-Nonachlor
Toxaphene-like compounds
1974-76
11.0
-
5.0
8.0
-
-
-
2370.0
36.0
-
-
1980
3.5
1.5
-
5.2
5.3
8.8
24.4
318.0
25.0
25.0
3200.0
Trace amounts of the 7 and 8 chlorine substituted PCDDs were found, but the
more toxic 4 chlorine substituted isomers were not observed, nor were the 5
and 6 chlorine substituted groups. Four chlorine substituted PCDFs made up
two-thirds of the total PCDFs found, the remaining one-third consisting of the
5 chlorine substituted compounds. The concentration ratio (PCDFs/PCBs) x 10
yields a value of 21.4 when the high resolution capillary column PCB values
are used. This ratio may be of particular significance when potential impacts
of PCBs on human health are considered.
CONCLUSION
The occurrence and distribution of selected persistent organochlorine
compounds has been studied in the Siskiwit Lake, Lake Superior, and Lake Huron
ecosystems. Although the origins of the more ubiquitous PCB compounds are not
clear, historic use patterns of the pesticide Toxaphene suggest major mechan-
isms for long range transport of derivatives of this material to the Great
Lakes. Traditionally, the vast majority of Toxaphene compounds have been used
in the deep South of the United States as an agricultural insecticide against
pests of cotton crops, specifically the cotton boll weevil Anthonomus grandis
(Coleoptera: Curculioninae). Utilization patterns of Toxaphene in other areas
closer to the Great Lakes suggest that only in recent years has application of
this material been made as a herbicide, or as an insecticide in conjunction
with the cultivation of sunflower crops in the mid-western states. Toxaphene
has never been registered or approved for use in Canada. Because existing data
indicate Toxaphene-like components in fish'of the upper Great Lakes as early
as 1974 (Swain and Glass, 1984), and because a loading rate of more than a metric
ton per year would be required to observe the burdens of these Toxaphene-like
materials in each of the Upper Great Lakes, it is highly likely that the major-
ity of these components are derived from extremely long range transport (thous-
ands of kilometers) by way of the atmosphere from areas of major utilization
in the southern United States. Recent research (Rice et al., 1984), utilizing
air and precipitation samples, and meteorological data, reached the same conclu-
119
-------
sion, i.e., that the probable source of the Toxaphene-like compounds observed
is from the central-southern United States.
ACKNOWLEDGMENTS
The authors wish to express their appreciation to the men and women of
Isle Royale Division of the National Park Service, U.S. Department of the
Interior for their continued cooperation and assistance over the several years
of the project period. The cooperative analytical assistance of Drs. Jim Petty,
David Stallings, Mike Ribick and Dick Schoettger of the Columbia National Fish-
eries Research Laboratory, U.S. Fish and Wildlife Service, and Dr. Eliott Smith
of the Cranbrook Institute of Science is acknowledged with gratitude.
The contributions of A.J."Rusty" Davis to the sampling program are re-
membered with appreciation.
Ook aan Carolien Schamhardt, secretaresse Vakgroep Aquatische Oecologie,
Universiteit van Amsterdam, die het manuscript heeft geprapareerd, hartelijk
bedankt.
TABLE 8. NEGATIVE IONIZATION MASS SPECTROSCOPY ANALYSIS OF EVISCERATED
WHOLE LAKE TROUT FROM SISKIWIT LAKE ( 1 kg composite of 18
fish taken in 1980)
Concentration (ng/kg, ppt) by number
of chlorine substitutions
Compound
Total
Chlorinated dibenzo-p-dioxins
(Detection limit)
Chlorinated dibenzofurans
(Detection limit)
ND ND ND Tr. Tr. Trace
(1.0) (1.0) (2.0) (2.0) (3.0)
10 5 ND ND Tr.
(0.5) (0.5) (0.5) (1.0) (1.0)
15
Ratio of four chlorine PCDFs to total 10:15 = 0.67
Concentration ratio (PCDFs/PCBs) X
Packed column PCBs
High resolution capillary
column PCBs
(15/4.3) X 106 = 3.5
(15/0.7) X 106 = 21.4
126
-------
REFERENCES
Eisenreich, S.J., B.B. Looney, and J.D. Thornton. 1981. Airborne organic
contaminants in the Great Lakes ecosystem. Environmental Science and
Technology 15:30-38.
Huber, T.J. 1975. The geologic story of Isle Royale National Park. Geological
Survey Bulletin 1309:1-66.
Murphy, T.J. 1976. Polychlorobiphenyls in the Atmosphere and in Precipitation
in the Lake Michigan Basin. Interim Report to the U.S. Environmental Pro-
tection Agency, March 17, 1976. U.S. Environmental Protection Agency,
Grosse lie, Michigan. pp. 1-4 (Typescript).
Murphy, T.J., and C.P. Rzeszutko. 1977. Precipitation inputs of PCBs to Lake
Michigan. J. Great Lakes Res. 3:305-312.
Rice, C.P., P.J. Samson, and G. Noguchi. 1984. Atmospheric Transport of Toxa-
phene to Lake Michigan. Report to the U.S. Environmental Protection Agency,
February 1984. U.S. Environmental Protection Agency, Grosse lie, Michigan.
Strachan, W.M.J. and H. Huneault. 1979. Polychlorinated biphenyls and organo-
chlorine pesticides in Great Lakes precipitation. J. Great Lakes Res. 5:
61-68.
Swain, W.R. 1978. Chlorinated organic residues in fish, water, and precipita-
tion form the vicinity of Isle Royale, Lake Superior. J. Great Lakes Res.
4:398-407.
Swain, W.R., and G.E. Glass. 1984. Potential for Pollutant Alteration of the
Hydrocycle. International Association for Great Lakes Research. Abstracts
of the 27th Conference on Great Lakes Research, pp. 69.
Veith, G.D., D.W. Khehl, F.A. Puglisi, G.E. Glass, and J.G. Eaton. 1977. Resi-
dues of PCBs and DDT in the western Lake Superior ecosystem. Archives of
Environmental Contamination and Toxicology. 5:487-499.
-121
-------
BIOLOGICAL TESTING OF INDUSTRIAL EFFLUENT
by
A.M. Beym1
ABSTRACT
Examined are procedures for biotesting of industrial effluents. These
procedures include toxicological criteria, field observations, biochemical
diagnosis, and toxicogenetic research.
INTRODUCTION
Planning and organizational measures, as well as sizable capital in-
vestments, for environmental protection that have been implemented in re-
cent years in the Soviet Union have appreciably slowed pollution of inland
and continental bodies of water while significant growth has occurred in
industrial and agricultural output (Federov 1977). However, development of
industry, increasing industrialization, and use of chemicals in agriculture
still lead to movement into the environment of a certain amount of chemicals,
including xenobiotics.
The paper and pulp industry still uses much water. A trend in all
countries in the 1950s has carried through to a full transition from the
sulfite method to the more progressive and ecologically beneficial sulfate
method of recovering pulp.
Use of new technological procedures permits maximum utilization of use-
ful timber products for the national economy. In addition to the main
product, cellulose, modern enterprises are recovering turpentine and other
terpenes, tall oil, methanol, hydrogen polysulfide odorizer, furfurol,
feed yeast and other products. All these engineering developments have made
it possible to prevent passage of excessive carbohydrates (pentoses and hex-
oses), terpene hydrocarbons, resin and fatty acids, alcohols, sulfur-contain-
ing agents of the methylmercaptan class, and other chemical compounds into
liquid sewage.
'institute of Ecological Toxicology,_
122
-------
Treatment plants serve as barriers to prevent pollution of water by sew-
age. Within a short period of time, 85% of the enterprises of the paper and
pulp industry were supplied with biological treatment installations. Two
combines, the Baykal Paper and Pulp Combine (BTsBK) and the Selenginsk Pulp
and Cardboard Combine, have integrated purification processes consisting of
three steps: biological, chemical, and mechanical. Other plants are equipped
with mechanical purification methods and systems of aerating ponds. The
treatment installations that have been started up recently were designed on
the basis of progressive water-consumption standards and existence of an effi-
cient intraplant system of purifying liquid sewage.
Work dealing with environmental protection and optimum use of natural
resources is acquiring increasing importance with each passing year. In the
USSR, sanitary-hygienic, water-management, and other environmental protection
requirements imposed on construction and operation of industrial enterprises
are growing increasingly strict. This requires additional work on problems of
environmental protection at all organizational stages: scientific research,
designing and building enterprises, and routine industrial work (Tipisev 1984).
Liquid effluent from different types of plants contain diverse chemical
compounds, the number of which may exceed several tens or hundreds. It is
possible to identify the chemical composition of this multicomponent flow,
but it is unrealistic to implement thorough and regular monitoring of these
substances. For this reason, inspection of quality of industrial sewage is
effected in our country and abroad mainly by performing hydrochemical analy-
ses of basic overall indicators and so-called priority pollutants. For exam-
ple, the following mandatory parameters for hydrochemical monitoring have
been defined for treated liquid sewage of the Baykal and Selenginsk combines:
active medium reaction (pH), temperature, color index, biological oxygen de-
mand (BOD5 and BODfull), chemical oxygen minimum (COM), total mineralization,
volatile phenols, and total organic sulfides. In a number of cases, when
treated sewage is discharged into reservoirs of a special category, which are
used for fisheries and have recreational value, additional parameters are
monitored: levels of iron and turpentine, methanol, volatile fatty acids,
carbohydrates, and others. In spite of expansion of the list of monitored
parameters, however, it is not feasible to efficiently determine the quanti-
ties of the entire range of chemical compounds, which makes it difficult to
provide a biological assessment of the quality of sewage dumped into bodies
of water.
BIOLOGICAL TESTING OF INDUSTRIAL SEWAGE
In recent times, methods of direct evaluation of toxicity of the water
environment are gaining increasing importance, i.e., biotesting of water qual-
ity by means of sensitive hydrobionts. Biotesting amounts primarily to check-
ing toxicity of industrial waste. In addition, it is equally important to
use biotests for determination of toxicity of polluted natural waters, and
this is already a matter of biological monitoring of their quality. In par-
ticular, this makes it possible to use biotesting in the system of monitoring
the quality of natural waters, as is being done in our country and abroad
(Braginskiy 1978, Izrael1 et al. 1978).
123
-------
Biotesting opens up vast opportunities for monitoring water quality on a
quantitative basis, since it yields concrete figures characterizing toxicity
of the water environment for hydrobionts. The results of biotesting are of
considerable interest, not only in the ecological, but hygienic aspects. On
the one hand, biotesting is used in hygienic studies as a rapid method of
assessing toxicity of the water environment and determining the nature of
change in it under the influence of different factors (Krasovskiy et al.
1983). On the other hand, hydrobionts are representatives of a complicated
aquatic biocenosis, which is actively involved in processes of spontaneous
self-purification of water to remove pollutants. Consequently, the toxic
effect of chemicals on them could lead to decline of self-purifying capacity
of a reservoir and worsening of its sanitary conditions, which is already of
interest from the purely sanitary and hygienic point of view (Beym et al.
1984).
TOXICOLOGICAL CRITERIA FOR ASSESSING
QUALITY OF TREATED LIQUID SEWAGE
Studies dealing with biotesting of sewage, in particular the industrial
waste from the BTsBK (Izrael1 et al. 1981), hold an important place in the
set of state measures to protect the flora and fauna of Lake Baykal against
anthropogenic pollution. Biological studies of liquid waste preceded devel-
opment of a basic industrial plan for integrated purification of industrial
waste at the BTsBK [Baykal Paper and Pulp Combine]. On the basis of model
and semiproduction experiments, marine toxicologists demonstrated that intro-
duction of methods of biological and chemical treatment of liquid waste leads
to significant decline of water pollution in reservoirs.
From the very first days of operation of the BTsBK (1967), a biological
service for monitoring liquid waste quality was organized. The purpose of
the studies included:
o Multilevel investigation of toxicity of untreated and treated liquid
sewage from sulfate and cellulose plants to aquatic organisms.
o Biological assessment of qualitative composition of liquid sewage.
o Determination of vital (inactive) concentrations of a number of organ-
ic and inorganic substances contained in runoff.
Typical representatives of plankton and benthos (algae, protozoans, hel-
minths, mollusks, crustaceans), as well as fish at different stages of devel-
opment (roe, larvae, young and adult specimens) were the objects of marine
toxicological studies. The distinction of the toxicological work done in
Lake Baykal was the difficulty of choice of test objects, since the lake's
ecosystem is unique with respect to presence of endemic species that play
the leading role in processes of transforming matter into energy.
The following served as indicators of the reaction of test organisms to
presence in water of toxic impurities: survival rate, poisoning symptoms, *
changes in reproduction and fertility, a set of pathophysiological and
124
-------
biochemical changes, and others. The vital properties of liquid waste were
evaluated in acute (24, 48 and 96 h), subacute (30 days) and chronic (6-9
months) experiments. The results of scientific research pursued for more
than 15 years revealed that, after combined purification, liquid waste has
mild toxicity (even if not diluted) for most endemic and palearctic aquatic
organisms. The degree of toxicity of industrial waste to hydrobionts dimin-
ishes substantially if it is diluted unadulterated water (gradually reaching
zero toxicity) (Table 1 )'.
TABLE 1. RESULTS OF BIOTESTING TREATED LIQUID WASTE FROM THE BTsBK (1967-
1984) '
Test Objects Vital Degree of Dilution
Algae 1:1 - 1:10
Chlorella pyrenoidosa
Chara sp.
Scenedesmus quadricauda
Protozoa 1:1
Paramaecium caudatum
Turbellaria 1:4 - 1:10
Baicalobia guttata
Annelides 1:1 - 1:4
Oligohaeta
Tubifex tubifex
Lumbriculus variegatus
Lamprodrilus nigrescens
Mollusca 1:2 - 1:5
Benedictia baicalensis
Crustacea 1:1 - 1=10
Cladocera
Daphnia magna
J). pulex
I), longispina
Chydorus sphaericus
Symocephalus vetulus
Copepoda
Cyclops kolensis
Epischura baicalensis 1 : 50
125
-------
TABLE 1 (continued)
Test Objects Vital Degree of Dilution
Amphipoda 1:1 ~ ^ : 8
Gammarus lacustris
Eulimnogammarus verrucosus
Eulimnogammarus cyaneus
Acanthogammarus victorii
Pisces 1 : 6 ~ 1:20
Coregonus autumnalis migratorius
Thymallus arcticus baicalensis
Paracottus kessleri
P. kneri
Cottocomephorus grewingki
Rutilus rutilus lacustris
Lenciscus lenciscus
Phoxinus phoxinus
Special tracer studies using the gold isotope Au-198 revealed that
treated liquid waste is diluted by a deep-lying dispersing device to 1/20th
or more at the dumping site and to more than 1/100 at a distance of 500 m
or more (Vetrov and Dekin, 1977). These data indicate that diluted sewage
in water that is being mixed does not have a devastating effect on repre-
sentatives of the aquatic biocenosis, including such particularly sensitive
species as the endemic little crayfish, Epischura baicalensis Sars.
FIELD STUDIES AND OBSERVATIONS: ICHTHYOLOGICAL MAPPING
In all periods of observation, fishing-crib experiments were performed
with representatives of the ichthyofauna (Baykal cisco—Coregonus autumnalis
migratorius Gorgi, grayling—Thymallus arcticus baicalensis Dybowski, bull-
head—Cottus Knei Dybowski, Cottus Kessleri Dybowski, minnow—Phoxinus phoxi-
nus Linne) in the immediate zones of dilution of industrial waste and along
the cone of their distribution in a north-easterly direction along the shore-
line. The fish in the cribs, which were placed at different depths, survived
in these zones without being fed for 48 days or more. Control catches in
this region with special fishing gear revealed presence of diverse represen-
tatives of ichthyofauna of Lake Baykal. Organoleptic studies showed that the
fish had no odor whatsoever. Concurrently, cytologists, geneticists, and
physiologists investigated cytogenetic changes in the cells of some fish
tissues, behavioral reactions of fish and gammarids exposed to low doses of
liquid waste. Electro-physiological monitoring methods were used to observe
changes in the chemo-receptor system of fish and crustaceans, as well as
biochemical criteria for evaluating stress reactions of fish and the basic
parameters of carbohydrate, fat and protein metabolism.
126
-------
BIOCHEMICAL EVALUATION OF FUNCTIONAL STATE OF FISH IN
BIOTESTING OF INDUSTRIAL LIQUID WASTE
One of the most important indicators that determine the resistance of
living systems to adverse (toxic) factors is their reactivity.
In biology, reactivity usually refers to the organism's capacity for an
adequate reaction in response to altered endogenous and exogenous conditions.
In other words, the highest levels of reactivity enable a living system to
respond in the most advantageous way to diverse factors and preserve homeo-
static functional parameters of vital systems.
It should be emphasized in particular that optimum forms of adaptation
to adverse factors do not necessarily provide for maintaining stability of
the organism's endogenous environment as a whole. Moreover, optimum level
of reactivity also refers to the capacity of higher integrative regulatory
elements of the organism to select the particular systems (which differ when
exposed to essentially different factors), the activity of which must be in
homeostasis. Deviations in their functional parameters must not exceed the
physiological range. At the same time, the functional activity of other
systems can change over the widest range without detriment to the organism
as a whole.
In the light of these conceptions, when conducting toxicological studies
of water, it is necessary to make a more distinct differentiation between the
results of observations of so-called target systems, i.e., those to which the
effect of a given toxicant is addressed, and purely regulatory changes, which
are necessary to compensate for functional deviations in organs or target
systems.
When performing biotests on fish, we made an attempt at biochemical de-
termination of the functional state of systems both directly related to de-
toxification of compounds contained in waste from the BTsBK and reactions of
the organism that are instrumental in maintaining optimum levels of reactiv-
ity.
In this regard, it is of special interest to investigate the system of
enzymes that catalyze oxidative metabolism of hydrophobic compounds (Kozlov
et al. 1983a, 1983b), localized in membranes of the liver's endoplasmic
reticulum. This system, which consists of two membrane proteins of NADP'H
cytochrome P-450 reductase and cytochrome P-450, is capable of utilizing
virtually any hydrophobic compounds in toxic agents. Moreover, these oxy-
genases participate in the metabolism of steroid hormones and fatty acids.
An interesting distinction of this oxygenase system is that a number of
organic compounds (barbiturates, polycyclic hydrocarbons, etc.) are capable
of inducing synthesis of new molecules of cytochrome P-450 in hepatic micro-
somes.
As a result of the experiments, the following results were obtained for
concentrations and activity of microsomal oxygenases (Table 2). Lake Baykal
is one of the cleanest bodies of water, and for this reason the demonstrated
low concentrations of cytochrome P-450 in the liver of local fish is due to
127
-------
the exceptionally low levels of NADP inductors, on the one hand, and on the
other hand it enables us to consider the demonstrated level of development of
systems of oxygenases with mixed function to be constitutive, used primarily
for metabolism of endogenous substrates with synthesis of steroid hormones
and bile acids.
This hypothesis is also confirmed by the fact that we demonstrated ex-
ceptionally low activity of benz-oi-pyrene hydroxylase, which does not ex-
ceed 0.5 pmol phenol metabolic products of benz-a-pyrene/mg protein/min,
which is 1/50th the activity inherent in warm-blooded animals and 1/1Oth the
activity of the liver of some salt-water fish.
TABLE 2. LEVELS AND ACTIVITY OF CYTOCHROME P450 IN LIVER OF LAKE BAYKAL
FISH
Fish
Cytochrome P-450,
nmol/mg microsomal
protein
Activity of NADP'H
cytochrome C
reductase, nmol
cytochrome C mg/min
Phoxinus Phoxinus Phoxinus 0.8 +^ 0.13
(Linne)
Perca fluviatilisl(Linne) 0.11 ^0.06
Rutilus rutilus (Linne) 0.31 _+ 0.12
Thymallus arcticus baicalensis 0.04 + 0.02
(Dybowski)
Coregonus autumnalis migratorius 0.04 +_ 0.01
(Georgi)
Procottus jeitelesi 0.21 + 0.05
2.49 +_ 0.99
0.60 +_ 0.20
1.17 _+ 0.31
0.61 +_ 0.01
0.58 + 0.07
It was of special interest to investigate the effect of waste from paper
and pulp plants on the system of microsomal oxygenases of fish in Lake Baykal.
For these experiments, fish (perch, grayling and Cisco) caught both in the
immediate vicinity of dumping from the BTsBK and other parts of Lake Baykal
were adapted to living in tanks with Baykal water for 5-7 days. After this,
the adapted fish were exposed to treated sewage in a concentration of 1:20.
We analyzed levels of cytochrome P-450, activity of NADP'H cytochrome P-450
reductase, as well as kinetics of enzymatic peroxidation of lipids induced
by NADP-H cytochrome P-450 reductase and cytochrome P-450, in microsomal
fractions of che liver of these fish. The levels of cytochrome P-450 were
the same in microsomal fractions from the liver of fish exposed to treated
sewage and in the liver of control fish. This indicates that, under these
128
-------
experimental conditions, the liquid sewage apparently contained no polycylic
hydrocarbons in concentrations sufficient for induction of microsomal oxy-
reductases. This also is confirmed by the fact that neither the kinetics
of cytochrome C reduction nor rate of lipid peroxidation in microsomal frac-
tions of the liver of fish kept in the run-off from the BTsBK in a dilution
of 1:20 differed from control values (Kotelevtsev et al. 1984).
It also is known that mechanisms of nonspecific adaptation are well-
represented in cold-blooded animals. The system of hormonal regulation
of physiological functions and hemoglobin system are the most labile in
reactions to changes in environmental conditions. The nature of physiologi-
cal changes in response to chemical stress indicates that the general adap-
tation syndrome develops in fish (Luk'yanenko 1967). There are no clearcut
and validated explanations, however, for the mechanisms of this phenomenon
due to lack of studies of different aspects of hormonal physiology of fish.
This makes it difficult to furnish a biochemical diagnosis of functional
state of fish under stress in aquatic toxicological studies. The role of
adrenocortical hormones in regulation of systemic adaptive reactions is
known. The data on glucocorticoid levels in fish blood are contradictory,
and this is attributable to the flaws in analytical methods (Wedemeyer et
al.1981).
The radioimmunological method we use permits very accurate evaluation
of the system involved in nonspecific resistance and choice of optimum forms
and levels of behavior in teleost fish under normal conditions and with ex-
posure to chemical anthropogenic factors. In control experiments, corticos-
terone was demonstrated in blood plasma of examined fish in the following
concentrations, ng/ml: grayling—3.3+0.4 (n = 16), cicso—6.2+JD.5 (n = 37),
perch—5.5+0.8 (n = 4) and burbot—2.1jf0.3 (n = 3). A preliminary series of
studies failed to demonstrate sex-related differences in levels of this hor-
mone. The condition of the system that provides for nonspecific resistance
was studied with exposure to the set of chemical factors contained in treated
liquid waste from sulfate-cellulose plants (BTsBK). Corticosterone level
constituted 2.4+_0.3 ng/ml (n ~ 14) in experimental grayling, and it did not
differ with statistical reliability from normal physiological values.
At the present time, it has become possible to develop special systems
adapted for selective detection of steroid compounds in fish with mineralo-
corticoid and glucocorticoid activity. Assays were made of concentrations
of aldosterone, 11-deoxycorticosterone and cortisol in samples of blood
from intact fish—grayling and cisco. It was demonstrated, in principle,
that factors differing in nature and duration of exposure (impairment of
abiotic environmental factors) elicit changes in relative concentrations of
glucocorticoid and mineralocorticoid hormones with an adaptive effect. Thus,
the radioimmunological method of assaying adrenocortical hormones permits
more accurate evaluation of the system of nonspecific resistance in teleost
fish under normal conditions and in the presence of chemical pathology caused
by anthropogenic factors, and it can be used with success for biological
testing of industrial liquid waste.
Studies of different fractions of hemoglobin (Hb) were conducted in the
Baykal cisco caught in a net in the southeastern part of Lake Baykal. During
129
-------
the adaptation period, the fish were kept in tanks with running water from
the lake (t = 8-10°C). It was ultimately found that the levels of alkali-
resistant Hb differed significantly in different species of fish (ranging
from 11.9% in the grayling to 25.5% in the cisco). There was similar varia-
tion in concentration of carboxyhemoglobin (from 0.86% in the grayling to
1.38% in the burbot) and methemoglobin (MetHb) (from 3.6% in the cisco to
6.7% in the perch and burbot). Methemoglobin content depends on blood levels
of oxidants that change Hb iron to a trivalent state. In the course of oxi-
dation of Hb to MetHb there is formation of oxygen superoxide anion with high
reactivity.
One of the most important enzymes of antioxidant protection of the
organism is superoxide dismutase (SOD), which causes dismutation of super-
oxide anion. High SOD activity is demonstrable in the blood of the examined
fish: from 1.18 to 10.93 arbitary units per mg Hb. On the whole, analysis
revealed that SOD activity is higher in cold-blooded animals than in warm-
blooded.
Investigation of the effect of liquid sewage on parameters of Hb compo-
sition in perch and grayling revealed the same direction of changes in met-
hemoglobin and carboxyhemoglobin content: increase in concentration of oxi-
dized Hb (methemoglobin), with decrease in CO-bound Hb (carboxyhemoglobin).
Evidently, there are substances in liquid waste that are instrumental in
converting hemoglobin to an irreversibly oxidized form (MetHb).
As a rule, the amount of alkali-resistant Hb increases under the effect
of adverse environmental factors. Under physiological conditions, higher
resistance to the denaturing effect of alkali is inherent in Hb at the early
developmental stages—embryonic and fetal. It was also found that diluted
liquid sewage from the aerating pond of the BTsBK could not alter signifi-
cantly either the Hb fractions studied or the activity of one of the enzymes
of the antioxidant system of red blood cells—superoxide dismutase.
Thus, development of modern biotesting methods requires use of the re-
sults of basic research in different branches of biology, including biochemi-
cal studies on the membrane and molecular levels. Indeed, identification of
the intimate mechanisms that protect functional systems of the cell against
the toxic effects of poisons, pollutants and their naturally occurring metab-
olites permits not only development of highly sensitive and specific tests
for biological monitoring of the environment, but prediction of mechanisms
of subsequent effects of xenobiotics and their metabolic products on aquatic
animals.
TOXICOGENETIC STUDIES
A modification of the semiquantitative Ames test—salmonella/microsomes
with use of a system of metabolic activation from the liver of the Baykal
cisco induced by methylcholanthrene—was developed for investigation of the
toxic genetic effect of constituents of waste from the paper and pulp indus-
try. Preparations of microsomal fractions from the cisco liver are highly
active, as are fractions from the liver of induced rats.
130
-------
Investigation of a series of samples of sewage from the Baykal Paper
and Pulp Combine collected at different times from different stages of pro-
duction and treatment using the Ames test with a system of metabolic acti-
vation from the liver of fish and rats, revealed that the composition of
liquid waste was heterogeneous, as manifested by the existence of direct
mutagenic activity in some of the samples taken from the first and inter-
mediate stages of production and treatment. No genetic activity was found
in virtually all samples taken from the last stage of sewage treatment
(Glazer et al. 1984).
BIOTESTING OF CONSTITUENTS OF LIQUID SEWAGE
The waste from paper and pulp enterprises is characterized by specific
organic substances that make up the wood, as well as those formed in the
course of the technological operation for production of cellulose and treat-
ment measures. For this reason, investigations to set the scientifically
validated maximum allowable concentrations (MAC) of substances that are a
potential hazard to aquatic organisms acquire importance.
Experimentation is the principal method of establishing the threshold
of toxicity of chemical compounds. MAC is determined for the weakest bio-
logical element that is damaged by the lowest concentrations of pollutant
and it becomes the minimum factor in assessing reservoirs. In addition to
the known method of fish samples and survival of other aquatic organisms,
a wide assortment of physiological, biochemical and biological tests is
used as a criterion of toxicity, and together they made it possible to de-
termine the concentrations of reagents at which the disturbances in vital
functions of hydrobionts are at the stage of prepathological development.
Special attention was devoted to long-term experiments to determine the
chronic adverse effects of low doses of toxicants and to experiments on
organisms with a short cycle of development in order to assess the genetic
sequelae of poisoning.
Many years of toxicological research established the vital permissible
concentrations of a number of deleterious elements in waste from the paper
and pulp industry. These substances are referable to different classes of
chemical compounds: aliphatic and terpene hydrocarbons, organic sulfides,
aromatic hydrocarbons of the phenol class and others. Taking into con-
sideration the toxicological indicator of deleteriousness, biologically
validated MAC were set for 30 components, including such agents as car-
bolic acid (phenol), ortho-cresol, meta- and para-cresol, guaiacol and
others.
For the first time in biological practice, studies were made of degree
of harm to hydrobionts of suspended substances (residue of cellulose fibers
and lignin).
The levels of all these substances in liquid sewage are strictly limited
and monitored on the basis of chemical analyses made by laboratories of the
State Committee for Hydrology and Metallurgy [Goskomgidromet].
131
-------
Toxicometric analysis for hydrobionts is made not only of individual
components, but different combinations (total monatomic phenols, total
multiatomic phenols, combinations thereof mixed with sulfur-containing com-
pounds of the methylmercaptan class and others). The MAC are used to assess
the degree of toxicity of formed industrial liquid waste. Ultimately, these
biological standards determine the extent of capital investments in planning
and building treatment plants, and they assure purity of lake water.
BIOMONITORING AS A COMPONENT AND INTEGRAL PART
OF BIOTESTING INDUSTRIAL WASTE
In order to solve problems of biotesting, it is important to select and
develop "key" or model territories that are exposed to industrial liquid sew-
age. The area of scattered purified liquid waste from the Baykal Paper and
Pulp Combine is one of these testing areas of permanent monitoring in South
Baykal (Kozhova 1971, Kozhova et al. 1965). Biomonitoring is performed with
particular thoroughness in a section that is adjacent to the site of dis-
charge of sewage. Material is collected in sections at distances of 0.05,
0.1, 0.5 and 1 km in the dumping region. Each section consists of several
stations in different depth zones, i.e., different biotopes.
A control testing area is situated outside the area affected by dumped
sewage. In addition to chemical composition of the water, studies are made
of both plankton and benthos communities, including water and soil bacteria.
The following are determined in plankton communities: composition and
quantity of phytoplankton; primary production by the oxygen and radioactive
carbon methods; total quantity of bacterioplankton by the method of ultra-
membrane filtration; basic physiological groups of bacteria; population of
bacterioplankton by the method of separate tests for total number and dark
assimilation of ^C; and composition, number, biomass and production of zoo-
plankton determined by indicators of growth and reproduction. Samples are
taken along the entire column of water from standard hydrological levels
using a (Dzhedi) net and bathymeter. Benthos is studied with consideration
of vertical zonality of communities and composition of soil at different
depths, with use of scuba divers who determine the projective cover of the
bottom and organisms that are difficult to detect with the usual collecting
gear (Kozhova 1974).
A comparison of all the communities studied revealed that an anthropo-
genic effect is manifested in the zone of a small spot of polluted soil
(0.1 km2) at the lake's bottom (beyond the littoral zone), where there is
accumulation of anthropogenic sediment. There, microbiological indicators,
total biomass of zoobenthos and proportion of dominant animals in the bio-
mass—gammarids, mollusks and Oligochaetae—undergo substantial change.
For example, at a depth of 5-20 m, the mollusk and Oligochaetae community
of such soil changes to a gammarid community. In the zone at a depth of
20-50 m, an Oligochaetae community changes into a gammarid one. The same
happens in the zone at a depth of 50-100 m. Gammarid biomass undergoes the
least change. However, even in gammarids, as in the other animals studied,
there is a reduction in species composition.
132
-------
Alteration of communities also occurs in less polluted soil, where the
anthropogenic sediment is less marked. Interestingly, at some depths the
macrozoobenthos on such soil is even larger than in control sections. At a
depth of 5-20 m, this happens due to increased profusion of mollusks and at
20-50 m, Oligochaetae. But in the 50- to 100-m zone, zoobenthos biomass also
decreases on less polluted ground. This is particularly noticeable for
Oligochaetae biomass.
A comparison of the parameters used for ecological mapping enables us to
rank them in the following order according to significance to water quality
in Lake Baykal: bacteria at the bottom and zoobenthos > bacterioplahkton >
primary production > zooplankton > phytoplankton > littoral phytobenthos.
Let us note that the zone of impact of the Baykal Paper and Pulp Combine
remained at virtually the same level in the course of a 10-year observation
period (Izrael1 et al. 1978).
Thus, biotesting of industrial sewage, unlike traditional analytical
methods of monitoring or in addition to these methods, makes it possible to
determine water quality on a quantitative basis and characterizes the extent
of toxicity of the aquatic environment for hydrobionts. It was demonstrated
with laboraory biotests that 24-h water from the sulfate-pulp industry has
virtually no acute toxicity after integrated three-step purification (bio-
logical, chemical, mechanical) for most of the endemic and palearctic hydro-
biont species examined. The results of chronic experiments revealed that
the degree of toxicity of liquid waste for aquatic organisms diminishes when
diluted in natural water (from 2- to 50-fold). Vital dilution of industrial
waste in the lake is obtained over a radius of 100 m from the heads of the
deep scattering dumping devices.
Modern methods of physiocochemical and molecular biology hold an impor-
tant place in biotesting, with regard to diagnosing pathophysiological states.
Studies have been made of mixed function monoxygenase systems of the liver
of Baykal fish, which affect oxidative metabolism of xenobiotics. It was
shown that the level of microsomal fraction cytochrome P-450 in fish, as
well as level of induction in membranes of the hepatic endoplasmic reticulum
of phospholipid peroxidation,can serve as a test to evaluate pollution of
the water environment. Using spectrophotometry and disk electrophoresis, it
was established that the waste from the BTsBK in a ratio of 1:20 does not
affect the level of cytochrome P-450 and, unlike the tested exogenous xeno-
biotics (3-methylcholanthrene), does not elicit synthesis of cytochrome P-450
isoforms.
All of the results of integrated studies are used in routine work for
experimental evaluation of toxicity of treated liquid waste from the Baykal
Paper and Pulp Combine, and they serve as the basis for determining condi-
tions under which waste is dumped to preclude any possible harm to the lake's
biocenosis.
Biotesting problems are multifaceted, and they are far from being
solved. One of the tasks for the next few years is to develop automated
biotesting systems, which would facilitate considerably the work of biologist-
133
-------
toxicologists and yield results faster. It would be useful and desirable to
specially discuss these aspects of biotesting within the framework of Soviet-
American collaboration.
BIBLIOGRAPHY
Beym, A.M., G.N. Krasovskiy, I.V. Sutokskaya, and L.Ya. Vasyukovich. 1984.
Hygienic-ecological approaches to biomonitoring of water quality. In:
Samoochishcheniye vody i migratsiya zagryazneniy po troficheskoy tsepi
(Water Self-Purification and Migration of Pollutants Over the Trophic
Chain). Moscow, USSR. pp. 16-22.
Braginskiy, L.P. 1978. New system for biomonitoring the aquatic environment
in the United States. Gidrobiol. Zhurn. 14(1):77-83.
Fedorov, Ye.K. 1977. Ekologicheskiy krizis i sotsial'nyy progress (The Eco-
logical Crisis and Social Progress). Leningrad, USSR, 176 pages.
Glazer, V.M. , V.M. Savov, S.K. Abilev, et al. 1984. A test system for bio-
monitoring based on membrane-bound enzyme complexes. 4: Evaluation of
toxicogenetic effects in Ames test system with metabolic activation by
microsomal monoxygenases from fish liver. Biol. Nauki, 5:85-89.
Izrael1, Yu.A., N.K. Gasilina, F.Ya. Rovinskiy, and L.M. Filippova. 1978.
Osushchestvleniye v SSSR sistemy monitoringa zagryazneniya prirodnoy
sredy (Implemention of System for Monitoring the Environment in the
USSR). Leningrad, USSR. 67 pages.
Izrael1, Yu.A., Yu.A. Anokhin, B.B. Chebanenko, et al. 1981. Integrated
analysis of environmental and validation of monitoring in the region
of Lake Baykal. In: Vsestoronniy analiz okruzhayushchey prirodnoy
sredy (trudy IV Sovetsko-amerikanskogo simpoziuma, Dzhekson, Vayoming,
22-27 oktyabrya 1979 g.) (Comprehensive Analysis of the Environment,
Transactions of 4th Soviet-American Symposium, Jackson, Wyoming, 22-27
October 1979), Leningrad, USSR. pp. 43-59.
Kotelevtsev, S.V., V.M. Glazer, V.B. Ritov, et al. 1984. Methods of
physicochemical biology for ecological and toxicological evaluation of
waste dumped by' enterprises of the paper and pulp industry. In: Prob-
lemy okhrany prirody. Tezisy dokl (Problems of Environmental Protection.
Summaries of Papers). Baykalsk, USSR. pp. 86-87.
Kozhova, O.M., L.A. Izhboldina, and G.S. Kaplina. 1965. Benthos of littoral
and sublittoral region along eastern shores of Lake Baykal. Gidrobiol.
Zhurn. 1(4):14-21.
Kozhova, O.M. 1971. Current status of fauna and flora of Lake Baykal in the
region of dumping industrial waste by Baykal plant. In: Issledovaniya
gidrobiologicheskogo rezhima vodoyemov Vostochnoy Sibiri (Investigations
of Hydrobiological Conditions in Waters of East Siberia). Irkutsk, USSR.
pp. 3-9
134
-------
Kozhova, O.M. 1974. Yearly changes in biocenoses of the Utulik-Murin region
of South Baykal. In: Produktivnost' Baykala i antropogennyye izmeneniya
yego prirody (Productivity of Lake Baykal and Anthropogenic Changes in
It). Irkutsk, USSR. pp. 160-171.
Kozlov, Yu.P., V.Ye. Kagan, A.M. Beym, et al. 1983. Test systems for bio-
monitoring based on membrane-bound enzymatic complexes. 1: Investigation
of oxygenases with mixed functions in liver microsomes of endemic fish of
Lake Baykal. Biol. Nauki. 1:20-24.
Kozlov, Yu.P., V.Ye. Kagan, A.M. Beym, et al. 1983b. Test systems for bio-
monitoring based on membrane-bound enzymatic complexes. 2: Investiga-
tion of enzymatic and nonenzymatic systems of lipid peroxidation in
liver microsomes of endemic fish of Lake Baykal. Biol. Nauki. 5:18-23.
Krasovskiy, G.N., I.P. Pletnikova, and I.V. Sutokskaya. 1983. Criterion of
deleteriousness in evaluation of toxic effect of chemical pollution of
water. In: Teoreticheskiye problemy vodnoy toksikologii (Theoretical
Problems of Water Toxicology), Moscow, USSR. pp. '42-48.
Luk'yanenko, V.I. 1967. Toksikologiya ryb (Toxicology of Fish), Moscow,
USSR. 216 pages.
Tipisev, A.Ya. 1984. Environmental protection—a priority task. Bumazhnaya
Promyshlennost'. 5:1-3.
Vetrov, V.A. and S.A. Dekin. 1977. Investigation of distribution of impuri-
ties using radioactive tracer. In Techeniya v Baykale [Currents of Lake
Baykal], Novosibirsk, USSR. pp. 133-143.
Wedemeyer, G.A., P.P. Meyer, and L. Smith. 1981. Stress and diseases of
fish. Moscow, USSR. 128 pages.
135
-------
COMPLEX EFFLUENTS TOXICITY INFORMATION SYSTEM
by
R.C. Russo1, A. Pilli2, and J. Crane2
ABSTRACT
The Complex Effluent Toxicity Information System (CETIS) is a com-
puterized data management system designed to assemble the results of efflu-
ent toxicity tests so that toxicity characteristics of complex effluents
can be determined on an industry-by-industry basis. Data are obtained
through literature searches of published reports and from individuals who
provide unpublished bioassay data from their state or regional biomonitor-
ing programs. Before entry into the computer file, the data are evaluated
by reviewers who are experienced in bioassay methods and trained in CETIS
procedures. In June 1984, data from 500 references for 1500 bioassay tests
were in the CETIS data base.
INTRODUCTION
Toxicity data from bioassays with freshwater and saltwater organisms
are used in assessing the effects of complex effluents on the aquatic
environment. Such information can be used in setting pollutant discharge
limits for industries and municipalities for wastewaters being discharged
into natural aquatic systems, in determining where to use toxicity testing,
and in interpreting toxicity test results.
Much of the toxicity data on complex effluents is available in un-
published form, although some information has been included in reports and
journal articles dealing with biomonitoring or with environmental impacts
of specific kinds of wastewaters. Bioassay methods used for complex
1 Environmental Research Laboratory, U.S. Environmental Protection Agency,
Athens, GA 30613.
Environmental Research Laboratory, U.S. Environmental Protection Agency,
Duluth, MN 55804
-------
ettluent testing are available in the literature (U.S. EPA 1971, Peltier
1978, APHA et al. 1981, Weber and Peltier 1981, Peltier and Weber 1984).
To make such data widely available to researchers and to those
involved in setting water quality standards, all obtainable published and
unpublished data have been compiled in the Complex Effluents Toxicity
Information System (CET1S).
The CET1S data base is a computerized data management system de-
signed to assemble the results of effluent toxicity tests so that toxicity
characteristics of complex effluents can be determined on an industry-by-
industry basis. The information contained in the data base is available
from the U.S. Environmental Protection Agency, Environmental Research
Laboratory in Duluth, Minnesota, or through the National Computer Center
in Raleigh, North Carolina.
DESCRIPTION AND METHODS
Data to be entered in the data base are obtained in two major ways.
First, literature searches identifying toxicity tests of complex effluents
are performed by computer bibliographic retrieval. Published reports are
screened tor suitability for CET1S, and those containing useful information
are incorporated into the system. Second, a coordination network consist-
ing of regional contact persons throughout the United States is in opera-
tion. These individuals are involved with state or regional biomonitoring
programs and provide unpublished bioassay data for inclusion in the system.
Published reports and unpublished data are evaluated by reviewers
who are experienced in bioassay methods and trained in the procedures of
the CETIS system. The bioassay information in a published or unpublished
report or on unpublished data sheets is carefully examined by a technical
reviewer to identify, extract., and record those data suitable for entry
into the data base. Appropriate information is entered onto data record
forms and entered into computer data files on a PDF 11/70 computer.
Information extracted from the published and unpublished data reports
is grouped into 12 general categories: Facility Information, Effluent/
Receiving Water Information, Reviewer Name/Coding Date Information, Sam-
pling Information, Toxicity Test Information, Test Water Information, Other
Water Profile Information, Test Organism Information, Dilution Water Infor-
mation, Test Method Information, Test Result Information, and Remarks.
Data encoded consist of both data elements unique to CETIS and data ele-
ments obtained through cross-reference from other National Computer Center
data systems.
Information included under "Facility Information" includes the fa-
cility name (.direct or indirect discharger), discharge number, pipe num-
ber, facility type, facility address, receiving water, and basin.
The types of wastewaters included in the data base are organized
according to the general categories of industry type, such as: ore mining,
-------
coal mining, textile mill operations, timber products processing,
pulp and paper board mills and converted paper product operations, paint
and ink formulating and printing, inorganic chemicals manufacturing,
plastics and synthetic materials manufacturing, miscellaneous chemical
manufacturing, organic chemical manufacturing, soap and detergent manu-
facturing, petroleum refining, paving and roofing material manufacturing,
rubber processing, leather tanning and finishing, iron and steel manufac-
turing, nonferrous metals manufacturing, machinery and mechanical products
manufacturing, electroplating, electric services, laundry operations, and
sewerage system operations.
Discharge types are identified as to whether a wastewater is a cool-
ing water, process water, or both; or whether it is some other type of
of discharge such as storm water runoff. Receiving waters are identified
as to name of water body, name of major and minor basin, and river reach.
Other information regarding the receiving water is also included, if
available, such as mean annual flow. Treatment processes to which a
tested wastewater was subjected are also noted in the encoding process.
These treatment processes are grouped into five categories: Physical
Treatment (lor example, ammonia stripping, floatation, multimedia filtra-
tion), Chemical Treatment (for example, carbon adsorption, chlorine
disinfection, ion exchange), Biological Treatment (for example, activated
sludge, nitrification-denitrification), Sludge Treatment and Disposal
(for example, aerobic or anaerobic digestion), and Other Processes (for
example, rueuse or recycle of treated effluent).
Sampling information on the bioassay test sample is recorded, such
as collection date and time, how the sample was obtained (continuous
sampling, grab sample, or composite sample), and whether the sample was
an actual field sample or a spiked or synthetically prepared sample.
If available, the flow from the discharge pipe at time of sampling is
entered.
Identifying information is provided about the bioassay, such as
test date and time, testing organization, test duration, effluent con-
centrations tested, exposure type, residue analysis, and bioassay type.
Exposure types include static, renewal, flow-through, and diet tests.
Bioassay types include screening tests, acute (short-term) tests, and
partial or full life cycle tests. Effluent concentrations tested may
be expressed as grams, milligrams, or micrograms per liter, or as per-
cent effluent in the test solution. Characteristics of the test water
are then entered. This includes such information as the concentration
ranges for the test water, dissolved oxygen, pH, temperature, alkalinity,
and hardness. The mean for each of these parameters is recorded if
available. The range includes any duplications or replications of the
dilution series concentrations. Test water chemical data for screening
tests are reported for the 0 percent and 100 percent effluent concen-
trations. If test water data are not reported, diluent or effluent values
may be used.
If measurements of additional water chemical characteristics have
been reported by the experimenter for the bioassay test water, effluent,
1,38
-------
or diluent (such as, for example, anions, metals, nonmetal cations,
organics), these are briefly identified as to their availability in the
original report.
Aquatic species used in bioassays included in the CETIS are fish,
macroinvertebrates, amphibians, and aquatic plants. Table 1 provides a
list of test species for which toxicity data are available in the CETIS.
The lifestage of test organisms is identified as one of the following:
alevin, adult, egg, eyed egg, fertilized egg, fingerling, fry, instars,
juvenile, larval, neonate, nymph, underyearling, young, yearling, or zoea.
The ages, weights, and lengths of test organisms are recorded, and the
source (cultured, field collected, hatchery, etc.) is identified. Infor-
mation is recorded on acclimation of the test organism to the test
dilution water, and source and pretreatment of the dilution water. In-
formation is included on use of control test specimens in the bioassay
and on statistical data treatment of bioassay results.
Bioassay data included in the CETIS are evaluated and assigned a
review code value tor quality, based on the experimenter's description of
bioassay test methods used and on the thoroughness ot the experimenter's
documentation of procedures and results. Published reports meeting the
following criteria are rated highest:
- Effluent collected less than 24 hours prior to testing. The method
of collection was reported. Experimental procedures followed pub-
lished methods.
- Standard test water chemistry data (that is, D.O., pH, temperature,
alkalinity, and hardness) were reported.
- Control mortality (or other adverse effect) was satisfactory (that
is, equal to or less than 10%).
- Statistical methodology used to determine the endpoint was reported.
TABLE . COMPLEX EFFLUENTS TOXICITY INFORMATION SYSTEM: TEST SPECIES
Species
No.
Latin Name
Common Name
1 PIMEPHALES PROMELAS
2 LEPOMIS MACROCHIRUS
3 SALVELINUS FONTINALIS
4 SALMO GAIRDNERI
5 DAPHNIA MAGNA
6 GAMMARUS LACUSTRIS
7 GAMMARUS FASCIATUS
8 DAPHNIA PULEX
10 CARCINUS MAENAS
11 CRANGON CRANGON
16 GAMBUSIA AFFINIS
FATHEAD MINNOW
BLUEGILL
BROOK TROUT
RAINBOW TROUT, DONALDSON TROUT
WATER FLEA
SCUD
SCUD
WATER FLEA
SHORE OR GREEN CRAB
COMMON SHRIMP
MOSQUITOFISH
139
-------
TABLE 1 (cont'd) COMPLEX EFFLUENTS TOXICITY INFORMATION SYSTEM: TEST SPECIES
Species
No.
Latin Name
Common Name
18 SEMOTILUS ATROMACULATUS
19 LAGODON RHOMBOIDES
20 ICTALURUS PUNCTATUS
21 CYPRINUS CARPIO
22 ONCORHYNCHUS TSHAWYTSCHA
23 ONCORHYNCHUS KISUTCH
25 CARASSIUS AURATUS
27 GAMMARUS PSEUDOLIMNAEUS
28 POECILIA RETICULATA
30 LEPOMIS CYANELLUS
33 FISH
38 PERCA FLAVESCENS
39 PALAEMONETES KADIAKENSIS
49 SALMO TRUTTA
52 HYALELLA AZTECA
54 SIMOCEPHALUS SERRULATUS
56 ASTERIAS RUBENS
67 CRASSOSTREA VIRGINICA
68 SALMO SALAR
69 LABIDESTHES SICCULUS
70 DAPHNIA SP
74 PENAEUS AZTECUS
75 PENAEUS DUORARUM
77 PENAEUS SETIFERUS
82 SALVELINUS NAMAYCUSH
83 STIZOSTEDION VITREUM VITREUM
85 CHIRONOMUS TENTANS
86 CYPRINODON VARIEGATUS
89 MICROPTERUS SALMOIDES
90 ONCORHYNCHUS NERKA
94 BARBUS TICTO
96 LEPOMIS MICROLOPHUS
102 PROCAMBARUS CLARKII
105 UMBRA PYGMAEA
106 MICROPTERUS DOLOMIEUI
110 LEIOSTOMUS XANTHURUS
112 NOTEMIGONUS CRYSOLEUCAS
113 CLARIAS BATRACHUS
133 ICTALURUS MELAS
140 CYMATOGASTER AGGREGATA
142 MYSIDOPSIS BAHIA
202 LABEO ROHITA
205 OSCILLATORIA LIMNETICA
208 MYTILUS EDULIS
211 ALBURNUS ALBURNUS
219 ONCORHYNCHUS GORBUSCHA
224 NEREIS ARENACEODENTATA
CREEK CHUB
PINFISH
CHANNEL CATFISH
COMMON, MIRROR, COLORED, CARP
CHINOOK SALMON
COHO SALMON
GOLDFISH
SCUD
GUPPY
GREEN SUNFISH
FISH
YELLOW PERCH
GRASS SHRIMP, FRESHWATER PRAWN
BROWN TROUT
SCUD
WATER FLEA
STARFISH
AMERICAN OR VIRGINIA OYSTER
ATLANTIC SALMON
BROOK SILVERSIDE
WATER FLEA
BROWN SHRIMP
PINK SHRIMP
WHITE SHRIMP (AMERICA)
LAKE TROUT, SISCOWET
WALLEYE
MIDGE
SHEEPSHEAD MINNOW
LARGEMOUTH BASS
SOCKEYE SALMON
TWO SPOTTED, TIC TAG TOE BARB
REDEAR SUNFISH
RED SWAMP CRAYFISH
EASTERN MUDMINNOW
SMALLMOUTH BASS
SPOT
GOLDEN SHINER
WALKING CATFISH
BLACK BULLHEAD
SHINER PERCH
OPOSSUM SHRIMP
ROHU
BLUE-GREEN ALGAE
COMMON BAY MUSSEL, BLUE MUSSEL
BLEAK
PINK SALMON
POLYCHAETE
140
-------
TABLE 1 (cont'd) COMPLEX EFFLUENTS TOXICITY INFORMATION SYSTEM: TEST SPECIES
Species
No.
Latin Name
Common Name
235 CAMPELOMA DECISUM
239 PHYSA INTEGRA
263 COREGONUS CLUPEAFORMIS
283 CATOSTOMUS COMMERSONI
285 ICTALURUS NEBULOSUS
288 CULAEA INCONSTANS
290 LEPOMIS MEGALOTIS
293 ETHEOSTOMA SPECTABILE
302 PALAEMONETES PUGIO
308 MYSTUS VITTATUS
309 SKELETONEMA COSTATUM
346 HOMARUS AMERICANUS
352 HETEROPNEUSTES FOSSILUS
361 ORCONECTES VIRILIS
366 ARTEMIA SALINA
371 CYPRINODONTIDAE
375 MENIDIA MENIDIA
376 ACARTIA TONSA
407 NOTOPTERUS NOTOPTERUS
418 CHANNA PUNCTATUS
422 CIRRHINUS MRIGALA
423 BREVOORTIA TYRANNUS
436 BOWMANIELLA DISSIMILIS
466 NOTROPIS CORNUTUS
482 ALGAE
483 CHAOBORUS PUNCTIPENNIS
486 SELENASTRUM CAPRICORNUTUM
488 INVERTEBRATES
508 CERATOPHYLLUM DEMERSUM
522 RANGIA CUNEATA
540 HYDROPSYCHE SP
549 HIPPOLYTE SP
572 CATLA CATLA
574 ANGUILLA ANGUILLA
575 CRASSOSTREA GIGAS
578 ALOSA AESTIVALIS
602 ORCONECTES PROPINQUUS
639 OPHIOCEPHALUS PUNCTATUS
677 PTERONARCYS SP
713 MYSTUS SEENGHALA
736 ONCORHYNCHUS KETA
892 ICTALURUS NATALIS
894 CLUPEA HARENGUS PALLASI
964 CHIRONOMUS RIPARIUS
970 AMBYSTOMA OPACUM
988 BUCCINUM UNDATUM
BROWN MYSTERY SNAIL
POUCH SNAIL
LAKE WHITEFISH
WHITE SUCKER
BROWN BULLHEAD
BROOK STICKLEBACK
LONGEAR SUNFISH
ORANGETHROAT DARTER
GRASS SHRIMP, FRESHWATER PRAWN
CATFISH
DIATOM
AMERICAN LOBSTER
INDIAN CATFISH
CRAYFISH
BRINE SHRIMP
KILLIFISH, TOPMINNOW FAMILY
ATLANTIC SILVERSIDE
CALANOID COPEPOD
FEATHERBACK
SNAKE-HEAD CATFISH
CARP, HAWKFISH
ATLANTIC MENHADEN
MYSID, OPOSSUM SHRIMP
COMMON SHINER
ALGAE, PHYTOPLANKTON, ALGAL MAT
PHANTOM MIDGE
GREEN ALGAE
INVERTEBRATES
COON-TAIL
COMMON RANGIA OR CLAM
CADDISFLY
SHRIMP OR PRAWN
CATLA
COMMON EEL
PACIFIC OYSTER
BLUEBACK HERRING
CRAYFISH
SNAKEHEAD
STONEFLY
CATFISH
CHUM SALMON
YELLOW BULLHEAD
PACIFIC HERRING
MIDGE
MARBLED SALAMANDER
LARGE WHELK
141
-------
TABLE 1 (cont'd) COMPLEX EFFLUENTS TOXICITY INFORMATION SYSTEM: TEST SPECIES
Species
No.
Latin Name
Common Name
1022 CTENODRILUS SERRATUS
1074 NOTROPIS HETEROLEPIS
1132 NOTROPIS ANOGENUS
1133 NOTROPIS EMILIAE
1137 TUBIFEX SP
1140 CENTRARCHIDAE
1178 PAROPHRYS VETULUS
1186 BRANCHIURA SOWERBYI
1295 MYSIDOPSIS ALMYRA
1321 DAPHNIA SCHODLERI
1329 HEMIGRAPSUS SP
1369 BALANUS GLANDULA
1407 LEMNA SP
1527 HETERANDRIA FORMOSA
1570 OPHRYTROCHA LABRONICA
1584 BARBUS SOPHORE
1585 CHANNA MARULIUS
1586 ANISOGAMMARUS PUGETTENSIS
1587 LEPOMIS SP
1593 LEMNA PERPUSILLA
1619 POMOXIS SP
1620 HELIODIAPTOMUS VIDUUS
1738 NOTROPIS PERCOBROMUS
1739 NOTROPIS ZONATUS
1905 MACROBRACHIUM KISTNENSIS
1984 MYSIDACEA
1990 HYBOGNATHUS PLACITUS
1991 GLYPTOTENDIPES SP
2052 MERCENARIA CAMPECHIENSIS
2053 DONAX VARIABILIS TEXASIANA
2054 DINOPHILUS SP
2065 PHOXINUS SP
2107 PAGURUS BERNHARDUS
2108 GAMMARUS DAIBERI
2109 NEOMYSIS AMERICANA
2123 ONCHIDORIS FUSCA
2124 CANCER PAGURUS
2143 MENIDIA PENINSULAE
2190 MUDFISH
2191 SALVELINUS ALPINUS
2203 PSAMMECHINUS MILIARIS
POLYCHAETE
BLACKNOSE SHINER
PUGNOSE SHINER
PUGNOSE MINNOW
TUBIFICID WORM
SUNFISH FAMILY
ENGLISH SOLE
OLIGOCHAETE
OPOSSUM SHRIMP
WATER FLEA
SHORE CRAB
ROCK BARNACLE
DUCKWEED
LEAST KILLIFISH
POLYCHAETE
TWO SPOTTED BARB, DOTTED BARB
SNAKE-HEAD CATFISH
SCUD
SUNFISH
DUCKWEED
CRAPPIE
CALANOID COPEPOD
FISH
BLEEDING SHINER
SHRIMP
MYSID OR OPOSSUM SHRIMP ORDER
PLAINS MINNOW
MIDGE
SOUTHERN QUAHOG
COQUINA
BRISTLE WORM
DACE
HERMIT CRAB
SCUD
OPOSSUM SHRIMP
SEA SLUG, NUDIBRANCH
EDIBLE OR ROCK CRAB
TIDEWATER SILVERSIDE
MUDFISH
ARCTIC CHAR
SEA URCHIN
The second highest data quality rating is given to published reports that
contain satisfactory etfluent collection methods and experimental proce-
dures but:
142
-------
- Standard test water chemistry data (D.O., pH, temperature, alkalinity,
and hardness) were either only partially reported, or not reported
at all.
- Control mortality or other adverse effect was not reported, was high
(greater than 10 percent) or was high but accounted for statistically.
- Statistical methodology, used to determine the endpoint was not re-
ported.
The third highest data quality rating is given to unpublished reports (such as
state and Federal agency studies) that document experimental procedures and
results. The lowest quality rating is given to unpublished data sheets that
contain the raw test data and include minimal information about experimental
procedures.
Bioassay test results may have different endpoints, and all of these are
accommodated within the CETTS. For example, test results may be expressed as
median lethal concentration (LC50), median lethal time (LT50) values, or
median effective concentration (EC50). EC50 values may be based on such
effects as abnormalities, growth, immobilization, or reproduction. Effect
endpoints used are listed in Table 2.
Any additional pertinent and useful information provided by the experi-
menter may be included in the data base as special observations. This often
includes such items as use of an effluent carrier or solvent, whether test
organisms were fed or unfed, or any observations of unusual behavior or evi-
dence of stress.
TABLE 2. EFFECT ENDPOINTS USED IN COMPLEX EFFLUENTS BIOASSAYS
ABD - Abundance: number of organisms of the same species has changed within
a population
ABN - Abnormalities: physical deviations observed from normal control organ-
isms
AVO - Avoidance: organism avoids or is attracted to certain effluent concen-
trations.
BEH - Behavior: quantifiable change in activity that arose from exposure to
internal or external stimuli
BIO - Biochemical Effect: physiochemical reactions (e.g., change in glycogen
levels) occurring within the organism on a cellular level
CYT - Cytogenetic Effect: genetic mutation on a cellular level
DIS - Disease: impairment of vital functions observed as a result of efflu-
ent concentrations, specific infective agents, inherent organism de-
tects, or a combination of these factors
ENZ - Enzyme Effect: deviations in enzyme activity
FCR - Food Consumption Rate: quantifiable change in rate of food consumed by
test animal
GRO - Growth: measured increase of animal size in length and/or weight
HAT - Hatchability: percent hatch
143
-------
HEM - Hematological Effect: changes in the blood parameters observed
HIS - Histological Effect: indicated by the presence of lesions or other
damage to tissues
MOR - Mortality - percentage of dead organisms
MOT - Motility: change in locomotor behavior
OC - Oxygen Consumption: change in 0-^ uptake in animals
PGR - Population Growth: increase or decrease in growth of an algal popula-
tion (e.g., change in cell number)
POP - Population: change in the species composition or diversity
PSE - Photosynthesis Effect: change in plant productivity
RES - Respiratory Rate: change in respiratory rate of vertebrates, inverte-
brates
RSD - Residue: toxicant uptake by tissues of test organism
SS - Swimming Speed: change in swimming speed
STR - Stress: observed physiological tension in animals or plants
TMR - Tumor Occurrence: presence of a mass of abnormal tissue
RESULTS
Data from 500 references for 1500 bioassay tests are in the CETIS data
base as of June 1S)84. These comprise data from 140 published papers (2050
tests) and from 900 state data sheets (2600 tests). The data currently in-
clude intormation from 16 states in the United States and from Canada.
Entered data also are transmitted to a larger computer system at the
National Computer Center in Raleigh, North Carolina. These data are cross-
reierenced with corresponding data from other data bases at the National
Computer Center. These are the Industrial Facilities Discharge (IFD) file,
which is linked to the United States Geological Survey river gage level file
(GAGE) by way of a hydrologic network file (REACH). By these linkages,
cross-reterenced information from the Industrial Facilities Discharge file,
river gage level file, and hydrologic network file are included in the CETIS
data base. Figures 1 and 2 provide schematic diagrams of the overall data
base management system, showing the ERL-Duluth and National Computer Center
portions.
Computer programs have been prepared to search and sort through the
data base to retrieve compiled information. These template retrievals are
available to provide data selected by industry, area or receiving water,
test, test species, and effluent treatment. A total data listing for up to
eight references can also be obtained. Table 3 provides a summary of re-
trieval options. Figure 3 is an example of a template retrieval output
sorted by test. Users can also design and implement specific retrievals of
their own design to obtain data from the system. A user's guide and pro-
grammer's manual describe these programs in detail (Gueldner et al. 1984).
Development of programs is underway to analyze the effluent toxicity data to
provide comparisons between organisms, effluents, and bioassay endpoints.
144
-------
REVIEWERS ENCODE
DATA & ASSIGN A
QUALITY REVIEW
CODE
\
/
DATA ENTRY
OPERATORS ENTER
DATA INTO
PDF 11/70 AT ERL-D
HARD COPY
IS PRINTED
ERL-D
POP
11/70
COMPUTER OUTPUT
IS DOUBLE PROOFED
AGAINST THE CODING
FORMS FOR ERRORS
CORRECTIONS
OR APPROVAL
CODE IS
ENTERED INTO
THE DATA FILES
\L
DATA IS FORMATTED FOR
TRANSMISSION TO THE
IBM COMPUTER AT NCC
Figure 1. CETIS at ERL-Duluth
-------
STORET
THE "PRE-SCAN" PROGRAM ON THE IBM COMPUTER
AT NCC COMPARES INCOMING CETIS DATA WITH
EXISTING DATA ON THE GAGE, IFD, AND REACH
DATA BASES. BOTH SETS OF DATA ARE STORED
IN A TEMPORARY FILE.
CTi
TEMPORARY IHS
FILE
AT
NCC
RECORDS ARE SELECTED
FOR INCLUSION IN THE
PERMANENT CETIS DATA
BASE
HARD COPY REPORTS
ARE PRINTED AND
PROOFED
PROGRAM OFFICES,
REGIONS, STATES
AND OTHER AGENCIES
ACCESS THE DATA BASE
THE TEMPLATE SYSTEM IS
USED FOR PACKAGED
RETRIEVALS AND REPORTS
V
SPECIAL PURPOSE PROGRAMS CAN
BE DESIGNED AND WRITTEN TO
PERFORM SPECIFIC RETRIEVALS
RETRIEVALS CAN
BE INTERFACED
WITH SAS
Figure 2. CETIS at National Computer Center
-------
TABLE 3. CETIS REPORT SPECIFICATION SUMMARY RETRIEVAL OPTIONS
Report Options
Select Options
Sort Options
Format Options
Industry Report
Industrial
Category
SIC Codes*
SIC Codes b State
State & SIC Codes
Brief Format
Expanded Format
SAS Format Disk
File
Area or Receiving
Water Report
Eight-digit
Catalog Unit
Eight-digit
Catalog Unit &
three-digit
Segment Number
Basin Code
State
Catalog Unit &
Segment Number
NPDES Number!
Brief Format
Expanded Format
SAS Format Disk
Test Report
Bioassay Type 1
Bioassay Type 2
Exposure Type
Test Duration
NPDES Number
SIC Code
Brief Format
Expanded Format
SAS Format Disk
File
Test Species Report
Test Species
Litestage
NPDES Number
SIC Code
Brief Format
Expanded Format
SAS Format Disk
File
Effluent Treatment
Report
NPDES Number
SIC Code
Effluent
Treatment
NPDES Number
SIC Code
Low Flow
Discharge Flow
Brief Format
Expanded Format
SAS Format Disk
File
Total Data Listing
Report
Reference
Numbers
Not applicable
Full Data Listing
Format
*SIC code = Standard Industrial Classification Code
TNPDES number = National Pollutant Discharge Elimination System Number
(discharge number)
147
-------
00
Th.SI1 DAfF
EKI.007JB5GOO
/ /
/ /
/ /
/ /
GAFl>00009oou
BM/nh/^b
M0/0b/2b
M1EOOOOI luou
74/1 i/ia
OHEOOOOI7UCJU
7b/0b/30
7S/08/21
73/11/21
/ /
ClHEIHlOO IHHOO
7b/u3/03
OHE000054DOU
73/11/12
73/11/12
OHEDOOObhOOU
73/10/lU
73/IO/ld
.
OlieuOOObToO"
73/10/23
73/1 1 /id
OHEUCuObHuOu
73/10/24
73/1 I/O 1
OHED0005900U
73/lu/ Jo
73/10/30
FXP. Hill.
•*9b2
1 F A
405?
ISA
2 S it
<«4b2
1 F A
4Sb2
1 A
i A
i S A
4 S A
49b2
1 A
4952
1 S A
2 S A
49b,2
1 S A
2 S A
49b2
1 S A
1 S A
4152
1 S A
i S *
49b2
1 S A
2 S A
Bid.
TYC-2
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
nsi
nun.
H4
M
1 1 /
11?
1
1
4
4
4
2
4
4
2
4
2
4
2
4
?
4
2
SHtriF.S
23
1
b
1
I
1
1
a
i
i
7
1
1
1
B
1
H
1
7
CHNC.
1.1
J.b
1 .1
J.b
Jb
47
1.7
45
100
100
100
100
13
2b
100
8b
100
H3
100
100
42
32
CDNC. RANGE; EFFECT
OU FH TEMP ALK
.7 13.7 106.7
LC50
b.4 125
LiCSO
b.4 22 125
LCSO
LC&O
7 21
21
7 21 315
7 21 31S
7.6 21
LCSO
7.5 21
LC50
7.t> 21
7.2 21
LCbO
7.2 21
7.2 21
LC50
7.2 21
7.3 21
7.3 21
LC5U
/ . b 21
LC50
7.b 21
HAR
100
too
260
262
262
242
242
82
82
238
238
200
200
384
384
Figure 3. Example of CETIS retrieval system test report—expanded format
-------
ACKNOWLEDGMENT
Many individuals have been involved with the Complex Effluents Toxicity
Information System, and their contributions are gratefully acknowledged.
Bruce Newton has provided invaluable suggestions on data base development.
Kenneth Carlson, Doretfce Gueldner, Charles Marks, Jeanne Rondeau, Daniel
Sivertson, and Phillip Taylor contributed to programming tasks. Cay Moriarity,
Beth Nordling, and Eve Katich assisted with reviewing and encoding of data.
Dorette Gueldner and Judy Veith have worked as data entry operators. PEDCO
Environmental, Inc., assisted with reviewing, encoding, and entering data.
REFERENCES
American Public Health Association, American Water Works Association, and
Water Pollution Control Federation. 1981. Standard methods for the
examination of water and wastewater, 15th ed. Am. Public Health Assoc.,
New York. 1134 p.
American Society for Testing and Materials. 1980. Standard practice for
conducting acute toxicity tests with fishes, macroinvertebrates, and
amphibians. pp. 1-25 in Annual Book of ASTM Standards.
Gueldner, D.R., A. Pilli, J.L. Crane, and D.J. Sivertson. 1984. CETIS:
Complex Effluents Toxicity Information System. CETIS retrieval system
user's manual. EPA/600/8-84-030, U.S. Environmental Protection Agency,
Duluth, Minnesota. (In press). 10 p.
Gueldner, D.R., D.J. Sivertson, and A. Pilli. 1984., CETIS: Complex
Effluents Toxicity Information System. Programmer's manual. U.S. Envi-
ronmental Protection Agency, Duluth, Minnesota. (Unpublished report).
92 p.
Peltier, W. 1978. Methods -for measuring the acute toxicity of effluents
to aquatic organisms. U.S. Environmental Protection Agency. Cincinnati,
Ohio. EPA-600/4-78-012.
Peltier, W., and C.I. Weber. 1984. Methods for measuring the acute toxicity
of effluents to freshwater and marine organisms. U.S. Environmental
Protection Agency, Cincinnati, Ohio. EPA/600/4-85/013.
U.S. Environmental Protection Agency. 1971. Algal assay procedure bottle
test. Nat. Eutrophication Res. Pro. 82 p.
Weber, C.I., and W. Peltier. 1981 Effluent toxicity screening test using
daphnia and mysid shrimp. U.S. Environmental Protection Agency,
Cincinnati, Ohio. (Unpublished report.)
149
-------
OPTIMIZING A PROGRAM FOR THE DEVELOPMENT OF WATER CONSERVATION
IN A SYSTEM OF WATER QUALITY MANAGEMENT CONSIDERING
POINT AND NON-POINT POLLUTION SOURCES
by
G. A. Sukhorukov^
ABSTRACT
A mathematical model is examined for optimizing a program ot gradual
and coordinated development of water conservation measures for achieving
standard water quality. The model includes a unit for the development of
new production methods, a unit for the development of procedures for pro-
cessing and recovering waste waters, including the establishing of inter-
sectorial and regional reusable systems, and a unit for forming water
quality.
For selecting the optimum program tor gradually achieving standard
water quality, a quadratic criterion has been proposed tor the accuracy of
achieving the maximum tolerable effluents. For considering the impact of
non-point pollution sources and tor calculating the maximum acceptable
etfiuents, two approaches have been examined—determined and probability.
A system and model also have been proposed for calculating the optimum water
conservation measures according to the criterion ot the minimum mathematical
expectation of the loss functions.
INTRODUCTION
Long-term and current forecasting and planning of water-protection
measures are tasks of utmost importance in the control of water quality.
Unlike ongoing control, where such parameters as capacity of industrial and
water sources are considered to be relatively stable, long-term and current
planning includes such parameters as the basic controllable variables.
Thus, long-term and current planning tasks determine the general task of
controlling development of water protection measures.
The complexity ot developing water protection measures is the reason
for the urgency and necessity of developing mathematical models for optimizing
Ail-Union Scientific Research Institute of Water Protection, USSR Ministry
of Land Reclamation and Water Resources, Kharkov, USSR
150
-------
development ot industrial and water protection complexes, and adopting wide
use of computers.
Analysis ot the survey literature (Vavilin and Tsitkin 1977) and other
sources shows that formulation ot the tasks of optimum development of water
protection was based on optimization ot parameters of treatment plants as
the basic water protection measure. In a number of instances, such tasks
included optimization of water drawdown. As a rule, such tasks were static
(single step) and did not include units for optimization of production
technology, which is the main factor that influences the quantity and
composition of diverted water prior to purification.
The urgency of the problem ot coordinated development of production
technology and treatment (purification) technology and the use of diverted
water had been repeatedly discussed in the literature; several models with
a high unitization level determining the correlation between development of
industry and environmental protection were proposed (Anon. 19ttZ). At the
same time, there is still the pressing problem ot developing mathematical
models and methods of optimizing coordinated development ot technology 01
production, processing, and utilization ot diverted water and water-use
tehnology on the level of an individual enterprise or a set ot enterprises
united by intersector water management systems.
It is expressly on this level, for enterprises, that the key parameters
are determined: maximum permissible discharge ot impurities into water
projects, plans tor development ot industry, and water protection measures
consistent with the plans for construction of installations. On the basis
ot the foregoing, we are considering models for optimization of a set of
water protection measures as a set of models for optimization ot industry
and its development (Industry unit), models for optimization of treatment
and use ot diverted water (Treatment and utilization unit), and models ot
processes within water reservoirs (Water system).
The dynamic nature ot development of industry and water protection, the
multistage nature ot development ot industry and water protection, and the
multistage planning ot such development (in 5-year periods) determine the
need to develop multistep, dynamic models of optimizaton ot programs tor
coordinated development of industrial and water protection complexes. The
inherent distinction of such models is that there are intermediate and end
goals ot development and criteria ot optimality for effectiveness of
achieving these goals stage by stage.
Bearing in mind the above-mentioned distinctions ot optimization of
the set of water protection measures, we developed a system of optimization
models tor coordinated development ot industry and water protection, which
is based on the method of building dynamic models of developing systems
(bukhorukov 1977).
Such models are designed for use in developing current and future
plans and torecasts. In such developments, the choice ot methods ot
considering nonpoint sources ot pollution—surface runoff from cities and
farms—is ot substantial significance. This must also be reflected in the
151
-------
models of optimization of programs for development of water protection, and
the problem arises of considering the random nature of processes of removal
of. superficial runolt.
MODELS FOR OPTIMIZATION OF PROGRAMS OF DEVELOPMENT
OF WATER PROTECTION
Let us examine moaels tor optimization of developing systems ot tech-
nological complexes. We shall use the term, technological complex (TK) , to
reter to the interrelated aggregate of units (.elements ot complex) that
affect physical, energetic, and informational conversion in order to obtain
a specified end result 01 production.
A developing system will be considered given (.specified) if the follow-
ing has been determined:
A. Finite set of points in time or steps, T = (t) limited by interval
of development 10, TJ , aggregate of sets, elements ot which are the following
non-negative phase variables (.vectors): Y(.t) — end results ot development,
Y(t=U) = Y(0) — original state Y(t), V(t) --- current control; Z(.t) — status of
system determining maximum production capacity and its initial state Z(.t=U)
= Z(U), u(t) — control of system development, i.e., control of development
ot production capacities, $t — external resources, o)t — assignment beyond
purpose of development.
B. Model of object of control that determines its "physical" properties,
as a set ot equations such as:
Criterion ot optimaiity ot development
Set of restrictions on phasic variables:
- resources t € j" (.4)
- goals ot development V C"t} G t^C*
(3)
In these expressions, we use the term, program ot development, to
refer to the aggregate of vectors of phasic variables X(.t) = LY(t) VU)
Z(t), UU)J.
152
-------
nln essence, optimization ot development programs consists of selecting
set x(t) on te T, with which the system changes from base state Z(U) to
states Z(i)...Z(t)...Z(T) as a result of controlling development—u(l)...
u(t)...u(T)—so that current equations—v(l)...v(t)...v(T)—lead to achieve-
ment of end results—y(l)...y(t)...y(T) , which meet externally given goals
of development (5), provided the extreme is reached lor the given criterion
of optimality ot development (2). The set ot restrictions (i, 3, 4) that
determine the correlation between phasic variables and restrictions tor
each of them (3), as well as restrictions on resources (4) allocated for
development (capital investments, etc.)—31...3t...3T—that are considered
given, are also important to choice of optimum program.
Proceeding from the general model of optimization of development (1-5)
and models of special-purpose programmed planning, let us formulate a typi-
cal problem of optimization of program of development ot technology of the
complex in the following rather simple form:
1
mm .. (6)
IB)
G U-
yC-fc)e
* •• t .* ' — - -)
where fv and fu are operating and capital expenditures with consideration
of adduction, B is matrix ot direct expenses, D is matrix of production
capacity expenses, y is vector of end product, v is vector of total product
(current production control), z is vector of production capacity, u is
vector of control ot development ot production capacity, fi is the set ot
given end goals, V and U are sets detemining restrictions on v(t), u(t) and
teT, z(0) is the initial state of production capacity with consideration
ot equipment no longer used.
Solution of problems (b-lU) yields the optimum program for ongoing
control v(t) and development u(t) ot the technological state with consideration
of its base state z(U) and need to achieve intermediate and end goals wc in
interval [I, TJ.
Let us turn to construction of a model for optimization ol development
of a set of water protection measures for a basin (region). We shall pro-
ceed from the following premises.
153
-------
1. The model defines Che set of measures directed toward improvement
and development of technology of production, technology of treatment and
utilization of diverted water with consideration of development of intersector
water-management systems, and technology of water use related to increase
in assimilating capacity of the water system in question *.as a result of
drawdown from reservoirs, diversion of run-off, etc.). Accordingly, three
units are distinguished in the model — "Industry," "Treatment and utilization,"
"Water system."
2. The tasks of releasing the end product and locating the plant are
considered to be given for planning intervals, since their estimation is
made on a higher level in the hierarachy of national economic planning.
3. Standards for water quality considered set and tied in with given
specified check: section lines, which is determined by the type of water use
on a river section.
4. The water-consuming enterprise, which is characterized by the
following elements, is an elementary subsystem of the model: discharge of
liquid waste, unit with given technology of production or treatment of
diverted water.
5. The overall task of optimizing development is separated into two
tasks, which are, in general terms, as follows:
Optimization of development of set of water protection
measures for the entire planning period 10, TJ , consisting
of more than one 5-year interval according to criterion of
minimal adduced expenditures in the absence of restrictions
on external resources.
Optimization of development of set of water protection measures
tor the first and subsequent 5-year periods in the presence of
restrictions on overall capital investments, according to cri-
terion of minimal A between given values for maximum permissible
discharge (MPD) of substances for which standards are set (.in
each enterprise) and level of obtained discharge of substances,
values of MPD are found by solving the first problem.
Each pair of problems is solved every 5 years with a shift made so
that MPD is recalculated in every 5-year period.
Let us consider the task of optimizing development of the set of water
protection measures for the period 10, TJ for a set of enterprises
which contains the following components.
Criterion of optimality — overall adduced expenses:
Ui)
H» vnin,
">v
154
-------
I. "Industry" unit—Kth enterprise,
(12)
(13)
where B^l' is the technological matrix of production balance^ D(^) is matrix
of coefficients of specific expenditures of unit output, V^1' = (vS1') is
total output of the ith type of product on the S^th unit or by the 0-j_th tech-
nology. u(U = (U^l') is the vector of buildup of production capacity of the
6-|_th unit, Z^Ho, I) = (z£J )) is the vector of initial status of production
capacity of the 6j_th unit with consideration of removal of capacities from
use, U^D is the set determining specific restrictions on control u^).
II. "Treatment and utilization" unit, xth enterprise, KeK equations
of balance of consumption of fresh, recycled and reused water, balance of
water diversion
(15)
oo
K -
where MK is the matrix of standards of water consumption according to cate-
gory (fresh water, industrial), G^>^' is matrix of standards of water diver-
sion (according to categories of run-off), e1, e11, e111, eIV are zero-unit
[?] structural matrices, V^^ is vector of outlay of diverted water treated
via different technological routes seS. Set SK is broken down into the fol-
lowing nonintersecting subsets: S]^ treatment and discharge into water sys-
tem, Sj recycled and reused for own use, S^ treatment and transfer to other
enterprises i PK = (P^, P^) is outlay of consumed fresh wate_r, including P£
from superficial sources and P1^ from subterranean sources, PK is limit on
water consumption and m^ is outlay of water coming from other enterprises.
where Kn is set of enterprises situated in the same industrial region or
city where it is feasible to organize intersector water-management systems,
155
-------
system of inequalities that take into consideration development of production
capacity of the "Treatment and utilization" unit.
where D is the matrix of coefficients of specific expenditures of production
capacity of units, u(^) = (U^)) is vector of buildup of production capacity,
Z^)(0, T) = Z^?)) is the vector of initial status of production capacity of
th unit with consideration of capacity no longer used.
III. "Water system1" unit.
System of recurrent equations of transfer and transformation of impurities:
(20)
where K+l, K are numbers of estimated section lines in stream, A7 and Av are
matrices of coefficients that consider dilution, self purification and trans-
formation of impurities, which also depend on expenditure increment in the
section (ic+l, K), YK+i is vector of concentrations of impurities for which
standards are set in the water system, G is the matrix of concentrations
of impurities in sewage after treatment, b is vector of uncontrolled dis-
charge of impurities,
system of restrictions on water quality in control section lines
(21)
where Z is goal matrix, w(3) is vector of water quality standards (.with con-
sideration of LPV [expansion unknown]).
System of recurrent equations of water balance.
~* ' (22)
156
-------
where YOK+1 and YOK are estimated outlay of water in section lines <+l and
K; XK is influx in the section (K+! , K) with consideration of water-manage-
ment balance, 6y.Oi< is increment of expenses due to drawdown from reservoirs,
diversion of run-off, etc., YO|C+^ is minimal environment-protective outlay,
E^ is zero-unit structural matrix.
System of equations determining feasiblity of increasing diluting
capacity of water system:
=y fcfe (23)
where vy3' is additional outlay from Jith water-management system (reservior,
canal) in the &th section line, aQ' is coefficient of influence ((Ka(3)o.) ?
u|3) is increment of indicators of production capacity of Jlth water-manage-
ment system, d(3) is coefficient of specific expenditures of production capac-
ity of &th system, V^) and u'^) are sets defining specific restrictions.
o
As a result of solving problems (11-25), the optimum values for v(T)
and u(T) are calculated, which determine all the necessary water-management
characteristics:
Input of capacities into "Industry," "Treatment and utilization" and
"Water system" units.
Volumes (expenditures) of fresh, recycled, reused water, volumes of
water discharged according to type of treatment (purification).
Capital [fixed asset] and operating expenses for implementation of
integrated measures.
Maximum permitted dumping of pollutants into water systems of
different enterprises.
* ^ /•> \ A / * V »
(26)
where E, is the type of substance for which standard is being set, g is
concentration of substance in sewage after treatment via sth route ^element
of matrix
157
-------
Calculation of w^ is probably one of the most important results of
the problem, which permit in the next step solving the set of local problems
for the first 5-year planning period in order to define the priority water-
protection measures. Let us consider, in order to describe formulation of
such problems, a hierarchically organized system with the following elements:
a) restrictions on overall capital expenditures in the basin (region) and
global criterion of optimality, b) set of local systems for individual enter-
prises, or set of several enterprises (industrial centers, cities) — neN.
We shall use the previously introduced definitions and designations to
form the optimization problem.
Global criterion of optimality is the minimum sum of squares of discrep-
ancies determining a mismatch between values and MPD — $(2) ancj obtained dis-
posal of impurities for which standards are set.
where C K is the vector of discrepancies [ errors] and R is a positively
difined matrix (in particular, R = E) .
Global restrictions on total capital expenditures for the basin (region
are.
(28)
where yn is scalar coordinating parameter.
Restrictions on overall capital expenditures allocated for development
of the industry and water-protection measures of the nth subsystem are:
'Industry" unit of Kth subsystem, KeK:
.iV""'* \
(30)
(31)
(32)
*- K *"'
158
-------
where U^'U) is the set of controls of development, elements of which are
defined by solving the preceding problem for the period [0, Tj.
"Processing and utilization" unit:
.CO. . £?!
a)
c
o
(35)
(37)
where G is the matrix, the elements of which are found b^ dividing concen-
trations of impurities (elements of matrix G^2)) by MPD — WP, U2^!) is
set of equations of controls of development, elements of which are found from
solving the preceding problem for the period [0, Tj , and e^ is a unit vector.
As can be seen from (27-37), the model of the water is not considered,
since restrictions on impurity discharge with sewage is determined by MPD.
The problem we have discussed (27-37) contains neN local subsystems,
each of which corresponds to an industrial center, city or agglomeration and
urban centers and industrial enterprises, agricultural enterprises where
there is a potential possibility of establishing intersector water-management
systems, including general treatment plants, local and intersector recycling.
Expressly such a structure of systems of treatment and utilization is in-
herent in all industrially developed regions, so that planning tasks should
solve the actually ocurring problems of intersector cooperation. Of course,
in the special case, each subsystem can contain only one enterprise.
The above tasks were defined in continuous (nonintegral) variables,
which makes it possible to obtain the "absolutely best" result, as compared
to integral formulation. Problems (27-37) can be solved in integral formu-
lation, if this is necessary, by different methods. One method involves
introduction of integral variables in the following manner, variables UQ
are replaced with the following expression:
159
-------
where 6 j is the optimizing variable, UQJ is the given numerical series of
unified values for production capacities.
Another, simpler method with regard to calculations, involves finding
the closest integral solution UQJ _> Ug.
On the whole, choice of problem solving method depends on the type of
functional and functions contained in the system of restrictions. In the
case of several simplifications, the problems can be reduced to problems of
mathematical linear (LP) and convex programming with use of standard packages
of applied LP programs in automated control systems. Problems (27-37) are
solved on two levels using a parametric noninteractive procedure (Sukhorukov
1977).
It should be noted that, along with use of systems (11-26, 27-37) in
the optimization mode, such systems can be well-used in the direct reading
mode for making forecasts for the period [.0, Tj or any intervals It, t-1).
In addition to obtaining a practical result—for example, forecasting water
quality—such a mode makes it possible to determine some initial approxima-
tion for optimization problems.
CONSIDERATION OF NONPOINT SOURCES OF POLLUTION WITH DETERMINISTIC
AND PROBABILISTIC FORMULATION OF PROBLEMS
Because of the specific nature of nonpoint sources of pollution of sur-
face run-off in cities and farms, they can be taken into consideration in
optimizing water-protective measures in two ways: a) removal of impurities
from non-point sources is considered uncontrollable (unoptimizable) in the
general model, depending only on the characteristics of the water catchment
area and precipitation, as components of hydrological calculations, b) re-
moval of impurities also depends on the set of water protection measures for
decontamination of superficial run-off, which are characterized by optimiz-
able variables. These variables are analogous to those discussed above, and
they are included in the "Industry" unit (for example, change in technology
of agriculture) and "Treatment and utilization" unit (for example, treatment
of a city's surface run-off).
In both the first and second instances, there can be both deterministic
and probabilistic formulation of the problem.
With deterministic formulation of the problem, the system of models
described above remains unchanged in structure, merely new elements (opti-
mizable and unoptimizable) are added to the "Water system" unit (2022). The
difficulty lies in estimating the unoptimizable removal of impurities with
superficial run-off. In addition to other methods, this can be calculated
by separating the superficial component of run-off in a river section per
month from the aggregate of surface and underground runoff under specified
nominal conditions, for example, in a year of 95% supply. Then, on the basis
of the charateristics of the water catchment area, determination is made of
mean concentration of impurities in superficial run-off and removal of impuri-
ties in water system averaged for a month. The advantages of such an approach
160
-------
is that removal of impurities with superficial run-off is tied in to nominal
hydrological conditions, and it is possible to consider the joint effect of
point and nonpoint sources on water quality under the same estimated condi-
tions, as well as to build a deterministic model of optimization of level
of decontamination of superficial run-of. A comprehensive method was devel-
oped on the basis of such an approach (Pasyuga 1985) and has been used in
practice.
At the same time, the physical nature of such processes as removal of
impurities with surface run-off in cities and outlay of water in the stream
is in the nature of random processes with considerable deviation frW mean
values, which is why probabilistic approaches are relevant. In this case,
to build models of optimization with consideration of random processes, an
approach is proposed, in which the criterion of optimum (minimum or maximum)
mathematical expectation of loss function is used as the optimality criterion,
which take into consideration the probabilistic characteristics of the
random process.
Let us discuss in greater detail the method of building optimization
models with consideration of the probabilistic factor. Let us add to our con-
sideration the vector of set of optimizing variables w = (v, u), vector of
set of parameters of water quality y = CyK) to Kelt of control section lines,
as well as vector of set of nominal parameters aeQ, which determines removal
of impurities with surface run-off. The same vector may include other nomi-
nal parameters: nominal debit of river, self-purification coefficients, etc.
In these terms, the single-step problem of optimizing water-protection
measures with given nominal parameters has the following appearance
(38)
(39)
where f is a function defining expenses for water-protection measures and is
the set of permissible concentrations of impurities (maximum permissible
concentration, MFC), for example, ft = {y^-.y MPC,
161
-------
MFC, when there is infraction of water quality standards and loss could be
defined as detriment to water consumers with use of polluted water, in the
fish industry and other sectors of the national economy (Anono 1982). On
the other hand, these losses could equal losses related to the basic industry,
either expenses for compensatory measures (drawdown from reservoirs, addi-
tional treatment before using water, and finally, enterprise shut-down).
If the values of z; are such that y(ct)>y(c)
-------
o Determination of function of losses of two types: loss due to in-
adequate actual level of water-protection measures for the actual
discharge of water in the channel [or water course] and loss related
to implementation of water-protection measures, the efficacy of
which exceeds the minimum required level for actual discharge of
water in channel.
o Determination of mathematical expectation of overall losses at given
planned (nominal)levels of water discharge in channel, as well as
possible variations of water quality.
o Determination of optimum nominal discharge of water in channel in
order to plan water-protection measures on the basis of minimizing
mathematical expectation of total loss at given densities of
probability of random values.
o Determination is made of final solutions to problem of selection of
water-protection measures with optimum nominal discharge.
Water discharge in a channel (river) was chosen as the sole nominal
parameter for the following reasons. In the first place, such a parameter
is the basic factor in forming water quality and it is also a random varia-
ble. The methods of building probabilistic characteristics of discharge
and run-off have been well-studied. In the second place, surface run-off
and river discharge are related by balance functions, so that one can con-
struct the characteristics of a one-dimensional random process determining
removal of an impurity with surface run-off and variations in river discharge.
For this reason, one can select a certain minimal discharge as r^m±n in (40),
and then a is the sought nominal discharge.
SOME PRACTICAL RESULTS AND CONCLUSIONS
Practical studies of the above-discussed models of optimization and fore-
casting of coordinated development of industry and water-protection complexes
remain to be done in order to develpo rather complicated systems of software
for computers, as well as a series of digital experiments, including those
for actual systems. For this purpose, a problem-oriented package of variable
programs (PPP) for forecasting water quality and optimizing waterprotection
measures, and some program complexes forming as a whole the OKVOPLAN (optimum
integrated water-protection planning and forecasting) were developed at the
Ail-Union Institute of Water Protection.
Calculations on a computer with use of the above-mentioned models yielded
the following results:
o Determination was made of MPD that are optimum with respect to minimal
expenditures for water-protection measures for several river basins.
The "Treatment and utilization" and "Water system" units, and surface
runoff in the deterministic variant were taken into consideration.
The variants of development of industrial technology were set on the
basis of the results of multivariant forecasting (direct reading).
163
-------
o A study was made of the etticacy of developing intersector water-
management systems with optimization of technology for utilization
of heat of water from heat and electric power plants (TES) to meet
the needs of the fish industry and agriculture. The calculations
revealed that there was substantial reduction of expenditures for
development of industry and water protection in the intersector
complex (TES, hothouse, irrigation system, fishery). The decline
of expenses constituted about 100% with 1 MW increment in TES
capacity.
o A study was made of the efficacy of introduction ot new technolo-
gies for steel production with out blast furnace in ferrous metal-
lurgy, and effect on expenditure indicators for the water-protection
complex.
o A study was made of probabilistic run-off on choice of optimum
water-protection measures, with determination of the need to estab-
lish different nominal river discharge, depending on the concrete
situation. Standard nominal discharge at the 95% level is usually
not the optimum according to the meaning of (40).
On the whole, our studies lead to the conclusion that coordinated devel-
opment of industrial technology and water-protection technology makes it
possible to reach the production goals and those of environmental protection
at substantially less cost than if they are developed separately. Further-
more, an optimum combination of rates of development in the "Industry and
"Treatment and utilization" units leads to substantial improvement of pro-
duction efficiency indicators with concurrent solution of problems of water
protection.
The difficulty of calculating optimum programs of coordinated develop-
ment of industrial and water-protection technological complexes makes it
necessary to develop special methods to design efficient routines tor the
computer. One method of constructing such routines is described in the
Appendix.
BIBLIOGRAPHY
Anon. 1982. Okhrana okruzhayushchey: model! sotsiai'no-ekonomicheskogo
prognoza (Environmental Protection; Models of Socioeconomic Forecasting).
Korn, G. and T. Korn 1973. Spravochnik po matematike dlya nauchnykh
rabotnikov i inzhenerov (Guide of Mathematics for Scientific Workers and
Engineers). Moscow, USSR. 831.
Pasyuga, N. P- 1985. Hydorlogical support of development of set of
water-protection measures in: Regulirovaniye kachestva prirodnyich vod
(Control of Quality of Natural Waters). Kharkov, USSR, pp 79-87.
Pervozvanskiy, A. A. 1975. Matematicheskiye modeli v upravienii proizvodstvom
(Mathematical Models in Control of Industry). Moscow, USSR. bib.
164
-------
Pospelov, G. S. and V. A. Irikov 1976. Programmno-tselevoye planirovaniye
i upravleniye. Vvedniye (Special-Purpose programmed Planning and Control.
Introduction), Moscow, USSR. 440.
Suk.norukov, G. A. 1977. Planning production and development in hierarchic
control systems in: Modeli i metody analiza ekonomicheskikh
tselenapravlennykh sistem (.Models and Methods of Analysis of Economic
Goal-Oriented Systems). Novosibirsk, USSR, pp 58-71.
Vavilin, V. A. and M. Yu. Tsitkin 1977. Mathematical modeling and
control of water environment. (5): 114-132.
Volgin, L. N. 1968. Problema optimal"nostey v teoreticheskoy kibernetike
(Problem of Optimums in Theoretical Cybernetics). Moscow, USSR. 160.
APPENDIX
Current methods of theory of hierarchic systems make it possible to
reduce the complicated task of optimizing the program for development of
industrial and water-protection complexes to solving a set of simple local
problems united by the coordination procedure. The body of this paper
described the structure of one of these methods, in which one first solves
problem (11-25) for the period LO.TJ, after which one solves problem (27-
37), which has an hierarchic structure for the first stage, t = 1. We
propose below a method of hierarchic optimization that permits determination
of the program of development for all t eT.
The method is based on aggregation in time of controls of development
u(t) in intervals of T, T-l , etc., and analysis of the problem in these
intervals with use of aggregation of variables (Vie) = V and methoda of theory
of duality. Let the initial problem of optimization of development program
have the following appearance:
The aggregated restriction variables have the following appearance:
, ,V S ".«),„. ,T- ft,
165
-------
Let us substitute the solution to the initial problem (41) with the
step- by-step solution of sequence of problems at the following steps:
U, v) = C L iXT H- cC V C V J \ -*• w
# r
-< }
- (43)
where u— is the result of solving the problem at preceding step T. The
solution continues analogously (43) to step t = 1.
It was shown that, if solution (41) exists an step-by-step routine has
admissible solutions, solutions (42, 43) coincide with the solutions to the
initial problem (41).
Let us turn to solving a problem such as (42) or (43) at each step.
In scalar form, (42) has the following appearance:
(44)
Z!vie - EX',
OL *L eeQ
(45)
. (46)
Decompositon of (44-46) leads to local problems such as:
(48)
166
-------
(49)
^e VLUu) ,
(50)
where y and n. are coordinating parameters and h is the number of iteration
with formation of y and ft.
The values of y and /z. are calculated using the formulas,
where v-j_ is a variable aggregated for 6 e 0j_ , B is aggregated square matrix
for 6 e 9j_ . The values of aggregated coefficients of matrix B are calculated
using the following formulas:
9=6
where n is length of steg^ at hth iteration, which is determined by the
"golden section" method, elO,l], i* is the number of the "leading"
subsystem in the hth interation, 8 * is the number of the "leading" column,
which is identified with the formula,
I
(51)
where A* y • are dual evaluations of (48,
Some details referable to logic of choosing i* , Q * were omitted in
formula (51). It was shown in practice that iterative solution of (47-51)
amounts to solution of (44-4b).
167
-------
GREAT LAKES AGRICULTURAL POLLUTION CONTROL IN PERSPECTIVE
by
V.J. Saulys1
ABSTRACT
Major efforts are underway in the Great Lakes area to control environ-
mental pollution, particularly with respect to nonpoint agricultural sources.
Farmers in the Lake Erie Basin have joined with environmental and soil con-
servation experts to institute programs to control these sources. Their work
has demonstrated the potential of no-till or reduced tillage practices to
reduce phosphorus loadings to streams and lakes. The potential for increased
pesticide pollution is of growing concern, however.
This paper has been reviewed in accordance with the U.S. Environmental
Protection Agency's peer and administrative review policies and approved for
presentation and publication.
INTRODUCTION
The Great Lakes, the "sweet water seas" of North America, have long
served to focus United States' energies on pollution control. President
Lyndon Johnson, when urging the Congress to pass the Federal Water Pollution
Control Act of 1965, said that Lake Erie, tor all intents and purposes, was
"dead." Some years later, Mr. V/illiam Ruckelshaus, the first Administrator of
the U.S. Environmental Protection Agency (EPA), noted that he once met with
officials from the City of Cleveland (Ohio). They assured him that the pollu-
tion problems of the Cuyahoga River, a major tributary of Lake Erie, had been
solved. "Next week," Ruckelshaus said, "the river caught tire and burned down
two bridges and a houseboat" (Lake and Morrison 1977a). Such incidents,
together with beaches closed by raw sewage and open beaches where rotting mats
of algae and floating rows of dead fish competed with people for sand and
water, stirred the people and their elected officials to action.
-U.S. Environmental Protection Agency, Region V, Great Lakes National Program
Office, Chicago, IL 60605 USA
168
-------
The pollution sources in the Great Lakes Basin were quickly identified
and targeted. These sources could be grouped into classes and categories
(Table 1).
TABLE 1. POLLUTION SOURCES
Location
Source
Urban
Rural
Point
Nonpoint
Municipal discharges
Industrial discharges
Atmospheric releases
Hazardous waste sites
Municipal waste landfills
Stormwater runoff
Human waste systems
Animal feed lots
Agricultural chemical and
waste spills
Erosion
Improper land use
Pesticide misuse
Fertilizer management
A decade ago the problems were highly visible and priorities apparently
simple. The point sources of pollution, particularly, industrial, were highly
visible. Thus, they were the first targets.
Today, many of these pollution sources are under control or have control
programs to address them. Municipal and industrial discharges must meet
strict federal technology-based requirements and state water quality stand-
ards. Phosphorus controls were imposed on sewage treatment plants and large
industrial sources in the Great Lakes Basin to slow the rate of eutrophica-
tion. As a result, the phosphorus loadings to Lake Erie from these sources
decreased from 15,260 metric tons in 1972 to 2,541 metric tons in 1982 (Figure
1).
Similar progress was reported for the industrial section. For example,
the U.S. paper ana pulp industry, once one of the "dirtiest of polluters," is
now one of the exemplary industries (Anon. 1981) (Figure 2).
Less progress has been achieved with urban nonpoint runoff. Many large
cities with serious problems in this area in the Great Lakes Basin, however,
have programs to capture and treat storm runoff or separate storm runoff from
domestic waste. The $3.2 billion surface and subsurface capture programs in
Chicago, Illinois, the $1.6 billion project in Milwaukee, Wisconsin, and
smaller projects in Rochester, New York, and elsewhere are examples of such
programs. Many open municipal and industrial dumps were buried during the
1960s and early 1970s to eliminate surface water pollution. These pro-
grams turned out to be, in many instances, an example of sweeping a problem
169
-------
1972 U.S. Load Estimate 13,870 Ml/year
8000 i-
_ 7000
ra
o>
o 6000
— 5000
c
T3
CO
JO
^ 4000
b
JZ
Q.
C/)
O
|- 3000
"5
2000
1000
UNITED STATES
U.S. Load at 1 mg/L
Cdn. Load at 1mg/L
J I
1972
1975
1980
1983
Figure 1. Lake Erie municipal phosphorus loads
under a carpet. Today, a massive state and federal effort is underway to
correct the resultant groundwater and surface water pollution created by past
expedient actions.
Major efforts are underway to address large atmospheric sources of con-
taminants. While much of the public's attention is on "acid rain," U.S. and
European research studies are gathering evidence that the rain may not be only
an acidic, but also a toxic, cocktail. Although much remains to be done in
the atmospheric control area, the basic technology to reduce sulphur and ni-
trogen oxides has been identified. Many necessary controls have been put in
place on mobile (vehicles) and stationary sources of pollution. As of April
1983, 106 industrial facilities and electric power plants in the United States
have sulphur dioxide control technology in place. An additional 35 have sul-
phur dioxide scrubbers under construction; and 70 more facilities are under
consideration for sulphur removal technology (.Personal communication, EPA
Region V, Chicago, IL. July 1984).
170
-------
1200
1000
800
C
o
I 600
400
200
Total Combined Loading to the Great Lakes
1967 1973
1977
1980
Figure 2. BOD5 Discharge trends of paper and pulp mills discharging directly
into the Great Lakes
Much progress has been achieved in cleaning up industrial and municipal
sources. Beaches closed for decades and ignored have been reopened as health
warning signs are removed one by one. Lake Erie, once pronounced dead, now
supports a renewed walleye sport fishery bringing $100 million per year to the
economy of the State of Ohio alone. The Cuyahoga River shoreline, once a fire
hazard, is now the site of new restaurants and luxury condominiums.
Rural point sources also have seen progress in pollution control. Reg-
ulations have been passed to eliminate large feedlot runoff, to prevent pesti-
cide misuse, and to contain agricultural chemical and waste spills. Violators
face large monetary fines. Small rural communities are encouraged to install
water treatment systems designed tor low capital costs and low maintenance.
Monetary assistance is available from state and federal agencies for design
ana installation of municipal treatment systems. Regulations have been
tightened by local government bodies to ensure the proper design, location and
installation of individual household treatment systems. The engineering solu-
tions, tor the most part, have been identified. Their implementation, in many
instances, is nearly complete.
The major remaining pollutant impact upon the Great Lakes ecosystem
comes not from point sources but from nonpoint agricultural sources. It has
been the farmer — not the city dweller or the industrial baron — who has
shaped the Great Lakes ecosystem. True, the cities have done their share to
damage the ecosystem, and, industry has contributed much more than its share
171
-------
of pollutants, particularly those that are toxic. The farmer, by changing the
face of the land, changed the ecology of the Great Lakes forever. Unless
agricultural nonpoint sources of pollution are controlled, however, the in-
vestment made in cleaning up urban sources of pollution will not produce the
full benefits. We have gone past the point where only the engineers will
provide the answers. We have reached the point of resource management. To-
day, the agronomists and the ecologists will have to provide the leadership
to complete the cleanup of the Great Lakes and to ensure their proper manage-
ment .
This situation was foreseen on January 7, 1972, during a conference on
Maumee River pollution problems. The participants of this conference were
told by the EPA representative that the Maumee River was the major contribu-
tor of silt and associated nutrients to Lake Erie. The participants were
advised that the initiatives necessary to control industrial and municipal
pollution were underway. The EPA representative also forecast that "when
these problems are solved, the problems of agricultural pollution will re-
main" (Lake and morrison 1977a). Among those conference participants were
farmers from the Allen County (Indiana) Soil and Waste Conservation District.
They did not like what they heard and resolved to do something to change
the prophecy. They held a series of meetings involving agricultural and
environmental agencies at the local, state, and federal levels. Their
timing was right and, within a few months, agreement was reached to under-
take a research program involving agricultural and environmental special-
ists to answer three questions (Lake and Morrison 1977c).
Could a conventional program of land treatment, undertaken in
the Maumee Basin, result in a measurable improvement in the
water quality in Lake Erie?
What would the cost be of such a project and could the dollars
spent be correlated with an improvement in water quality?
What programs, incentives, and administrative techniques would
result in the most nearly complete participation by landowners?
The research program was to take place in Allen County with the full
support of the Allen County Soil and Conservation District. Thus, the
Black Creek Project was born in 1972. It was the first detailed look at
the contributions of agriculture to the degradation of water quality in the
United States.
The Black Creek watershed was chosen because it was representative of
the Maumee Basin's land use and soil types (Table 2). It was large enough
(4872 hectares) to allow for sufficient studies of the sub-watershed and
the differing land uses. The Maumee Basin extends into the states of Ohio,
Indiana and Michigan. It is one of the last of the areas in the Lake Erie
Basin to be settled (mid-19th century) because it was the site of the "Great
Black Swamp," the remains of the glacial Lake Maumee. Until the arrival of
German settlers with experience in farming heavy clay soils, it was not
considered fit for serious agriculture. Today, it is one of the most pro-
ductive agricultural areas of the United States.
172
-------
X x"— —RIVER BASIN BOUND
Figure 3. Maumee Basin—Black. Creek Location
TABLE 2. BASIN AND SUB-BASIN LAND USE*
Maumee River Basin
Black Creek Basin
1,711,493 hectares
73% cropland
4% pasture
80% woodland
9% urban
4,872 hectares
80% cropland
4% pasture
4% woodland
9% urban
*Lake and Morrison (1977a).
173
-------
The Black Creek scudy consisted of & research program to define the
water quality and biological parameters of Black Creek and an administrative
program to test the pollution control benefits of known soil conservation
technologies. During a 5-year study period, it was found that the tra-
ditional grab sampling methods were inadequate to accurately show the true
load of sediments and related pollutants discharged by Black Creek to the
Maumee River. Automated samplers had to be installed at several sampling
locations. This equipment allowed the scientists to collect more frequent
and, thus more accurate, information on water quality and sediment load fol-
lowing a rainfall event. The information provided by these samplers showed
that a single rainfall event could deliver as much as &6% of the total annual
sediment load.
The Black Creek chemical parameters were typical of such agricultural
watersheds: nitrate-nitrogen runoff of 2 to 20 kg/ha, phosphorus levels
significantly above Lake Erie objectives (0.05 - 0.16 mg/1 versus 0.01 mg/1
target), and an aquatic community dominated by pollutant-tolerant species
(Lake and Morrison 1977a).
The agricultural portion of this study planned to investiage the cost
and benefits of 30 agricultural soil control practices and combinations.
These were grouped into three categories:
Structural practices
Diversion structures to reduce length of slope
Slope stabilization structures
Grassed waterways to convey storm water to streams
Holding ponds and tanks for animal wastes
Land smoothing
Fences to keep livestock away from streams
Special livestock watering facilities
Ponds
Sediment control basins
Stream channel stabilizers
Stream bank protectors
Surface drains
Gradient terraces
Paralleled terraces
Tile drains
Cultivation practices
Contour planting
Crop residue management
Minimum tillage
Pasture and hayland planting
Strip cropping
Improved harvesting of woodlands
Woodland improvement
Woodland pruning
174
-------
Combined structural and cultivation practices
Farmstead and feedlot windbreaks
Field border plantings (hedgerows, etc.)
Field windbreaks
Reforestation
Sediment control during construction
Recreation area improvements via plantings
Wildlife habitat management
The study yielded answers in a number of areas. Of the above listed
structural and cultural practices, nine were chosen as constituting Best
Management Practices tor the Black Creek:
Field borders
Grade stabilization structures
Grassed waterways
Fences for • livestock
Pasture planting
Sediment control basins
Terraces
Limited channel protection
Tillage methods that increase crop residue and
surface roughness.
A mathematical model called ANSWERS (Aerial Nonpoint Source Watershed
Evaluation Simulator) was developed as part of the project. It was shown to
predict the effect of an actual storm event loading within 15% of that mea-
sured in Black Creek. The model also was used to identify the most criti-
cal areas within the watershed and simulate the effect of changing from
conventional tillage (using the molboard plow) to minimum tillage (using the
chisel plow). This simulation showed that 4U% of the reduction in soil loss
could have been achieved by treating 32 hectares in the most "critical"
areas rather than the 728 hectares affected by the project (Lake and
Morrison 1977a).
Work with rainfall simultors on small plots confirmed that crop resi-
dues and tillage practices, which prevent soil mobilization, were more sig-
nificant in preventing soil loss than those practices aimed at intercepting
mobilized soil. Once a rain drop's impact mobilized the soil particles, it
was an uphill struggle to keep the soil from the stream.
Furthermore, the biological studies showed that several of the struc-
tural changes, insisted upon by the local farmers, such as stream channeli-
zation and stream bank protection, did little to reduce total loss of soil
but significantly damaged the stream's biological community (Lake and
Morrison 1977b).
At the conclusion of the study, the U.S. Army Corps of Engineers,
which had been required by the U.S. Congress to "develop a demonstration
wastewater program for the rehabilitation and environmental repair ot Lake
Erie," took the findings of the Black Creek project and applied them to a
175
-------
nearby watershed in Ohio. The Honey Creek Project focused on no-till and
minimum tillage practices. As seen in Table 3, the project confirmed that
they were the cheapest and most effective measures to reduce phosphorus and
to rehabilitate Lake Erie.
The Maumee River and other Lake Erie tributaries are estimated to
deliver an average of 8400 metric tons of agricultural nonpoint phosphorus
per year to Lake Erie (U.S. Army, COE 1983). Since in 1982 the total phosphorus
input to Lake Erie was estimated at 12,349 metric tons, a reduction in agri-
TABLE 3. COSTS OF METHODS FOR REDUCING PHOSPHORUS RUNOFF*
Method
Rural nonpoint sources
No-till and conservation
tillage
Cover crops
Critical area seeding
Contour stripcropping
Diversions
Waterways
Vegetative filters
Runoff control structures
Terraces
Tile drains
Manure storage and
spreading
Barnyard runoff control
Fertility management
Fertilizer placement
Total
phosphorus
$/kg
0
276.00
326.00
82.00
2,640.00
97.00
30.00
368.00
73.00
9, 180. '00
4.40
2.20
0
44.00
Biologically
available
phosphorus , %
25
25
25
25
25
25
25
25
25
25
75
75
25
25
$/kg of
Biologically
available
phosphorus
0
1,100.00
1,300.00
330.00
10,600.00
390.00
120.00
1,470.00
290.00
36,720.00
5.90-36.3
2.90
0
176.00
*U.S. Army, COE (1983).
176
-------
cultural contribution is necessary to meet the 11,000 metric ton phosphorus
loading target set by the 1978 Great Lakes Water Quality Agreement between
the United States and Canada (Great Lakes Water Quality Board 1983). The
results of the Lake Erie Wastewater Management Study indicates that, with
use of no-till and reduced tillage practices, the achievement of this
target is feasible (Table 4).
TABLE 4. EXISTING AND POTENTIAL ACHIEVABLE REDUCTION OF SOIL EROSION IN
THE UNITED STATES LAKE ERIE DRAINAGE BASIN* (million metric tons
per year)
Conservation
practices
Western
basin
Central
basin
Eastern
basin
Total
Lake Erie
basin
Base year (1975)
Reduced tillage only
(percent reduction)
No-till and reduced
tillage (percent
reduction)
16.6
8.3
(50)
4.5
(73)
6.7
3.9
(42)
2.4
(64)
1.8
1.3
(28)
0.9
(50)
25.1
13.5
(46)
7.8
(69)
*U.S. Army, COE (1983).
In late 1980, the information available from the Lake Erie Wastewater
Management Study and the reanalysis of the Black Creek data using the ANSWERS
model made it evident that the no-till and minimum tillage practices would
significantly reduce the amount of phosphorus reaching Lake Erie. The in-
formation indicated that there would be little, if any, additional cost to
the farmer, if he was trained in the use of new equipment and practices.
Therefore, EPA Region V, U.S. Army Corps of Engineers, and Department of
Agriculture representatives signed an agreement in 1981 to jointly work
toward the acceptance of these practices. It was agreed that the most effec-
tive way to introduce a change in farming practices would be to fund the
local soil conservation districts to demonstrate the use of the new tech-
niques and equipment by trained local members of the farming community.
This proposal for conservation tillage demonstration funding was en-
thusiastically received by the local soil conservation districts. Projects
have been funded in 31 counties in the Lake Erie Basin. The projects
targeted "critical areas" where erosion would result in the greatest phos-
phorus control. During 1983, the projects tracked the performance of 1854
demonstration farm plots covering a total of 9472 hectares. These plots
were analyzed for costs of production (including fertilizer, pesticide,
and machinery use), crop yields, and estimated soil loss. The preliminary
conclusions from this demonstration program indicate that:
177
-------
1. Yields with notill and ridgetill were competitive with yields
produced under conventional tillage systems.
2. Costs of production for conservation tillage systems were less
than or equal to costs of producing the same crops using con-
ventional systems.
3. Conservation tillage systems reduced phosphorus loadings from
the project area and did not significantly increase herbicide
use (4 to 12%).
Although only 9400 hectares are being closely monitored by the program,
about 22% of the 3,237,500 hectares of farmland was in some form of conserva-
tion tillage in 1983. The demonstration projects should significantly in-
crease this percentage in the coming years (U.S. Environmental Protection
Agency 1984).
One of the principal concerns about encouraging the farmers to switch
to no-till or reduced tillage farming practices is the potential for in-
creased use of pesticides. There are two conflicting schools of thought on
the subject:
1. The chemical usage will increase because the conservation tillage
practices may rely upon pesticides to control weeds and insects.
2. The farmers currently make heavy use of these chemicals, thus,
conversion to conservation tillage practices with close attention
to chemical use should result in similar or lower pesticide usage.
The existing literature reveals little information on current pesticide
usage in the Lake Erie Basin. The information, which is currently available,
is significantly out of date (1978 and earlier). Recent work (Baker, 1983)
shows that the amount of pesticides present in runoff from Lake Erie agri-
cultural areas can be significant.
These pesticide loadings can be contrasted with the organic contami-
nants discharged by industrial and municipal sources to the Niagara River,
which drains Lake Erie into Lake Ontario (Table 6).
The latter loadings are deemed to be of enough significance as to call
for special cooperative measures between U.S. and Canadian pollution control
agencies to identify and reduce the sources. Granted, the formulations of the
pesticides measured in Maumee and other rivers are such that they should de-
grade readily in the environment. However, recent findings by Baker (1983)
of concentrations of atrazine, alachlor, and metolachlor in public water
supplies similar to the stream concentrations indicate that the degradation
mechanisms are not as quick as they were assumed to be. Sampling of Lake
Erie waters has not indicated a build-up to date of these compounds. No
attempt has been made, however, to assess the impact of pulsed loadings to
sensitive fishery areas during the annual reproductive cycle. Nor have
these compounds been assessed as to their synergistic, lifetime exposure
impacts on human health.
178
-------
TABLE 5. ESTIMATED PESTICIDE LOADS (kg) FOR THE PERIOD BETWEEN MAY 1, 1982,
AND JULY 31, 1982*
Total loading
Pesticide
Metolachlor
Atrazine
Alachlor
Metribuzin
Cyanazine
Simazine
Linuron
DEA
DIA
Chlorpyrif os
Penoxalin
Ethoprop
Totals
Honey
Creek
241
223
89.2
27.5
24.7
19.7
62.4
29.4
52.0
9.47
2.38
1.06
781.8
Sand.
River
1750
1600
1290
518
226
179
264
168
130
135
19.3
3.86
6,283.2
Maumee
River
2920
4240
2820
1370
1590
1280
571
490
629
226
95.4
34.7
16,266.1
*Baker (1983).
TABLE 6. CONTAMINANT GROUPS DISCHARGED INTO THE NIAGARA RIVER*
Contaminant group
Total loading
to river, kg/day (Ib/day)
Acid extractabies
Poiynuclear aromatics
PCBs
Pesticides
Other base/neutral extractabies
Purgeables
46 (100)
17 (38)
0.11 (0.24)
1.6 (3.5)
30 (66)
140 (310)
*Anon. (1984).
179
-------
In summary, the Great Lakes have long served as an early warning system
tor the world about environmental problems. They also have served to demon-
strate that close international cooperation can reverse trends in existence
for decades and even solve environmental problems. The control of phosphorus
and eutrophication in Lake Erie is an excellent example.
The Lakes have shown us clearly that soil and its attendant chemicals
is a major pollutant, destroying the ecological balance of one of the streams
and aging the Lakes. The consequences of land runoff control strategies are
only now being assessed. The new challenge is to keep the soil on the farm
as a resource for future generations while feeding the increasing population
of the world.
REFERENCES
Anon. 1981. The response of the pulp and paper industry in the Great Lakes
Basin to pollution abatement programs. Report to the Great Lakes Water
Quality Board: Pulp and Paper Point Sources Task Force of the Water
Quality Programs Committee. 68 pp.
Anon. 1984. Niagara River Toxics Committee Draft Report. Unpublished.
Baker, D.B. 1983. Studies of sediment, nutrient and pesticide loading in
selected Lake Erie and Lake Ontario tributaries - draft final report.
Unpublished.
Great Lakes Water Quality Board. 1983. 1983 report on Great Lakes Water
quality. Great Lakes Water Quality Board Report to the International
Joint Commission. 97 pp.
Lake, J., and J. Morrison. 1977a. Environmental impact of land use on
water quality: Final report on the Black Creek Project - Summary.
Great Lakes National Program Office, Chicago, Illinois. EPA-905/
9-77-007-A, United States Environmental Protection Agency. 94 pp.
Lake, J., and J. Morrison. 1977b. Environmental impact of land use on water
quality: Final report on the Black Creek Project - Supplemental Comments
Great Lakes National Program Office, Chicago, Illinois EPA-905/9-77-007-D,
United States Environmental Protection Agency. 107 pp.
Lake, J., and J. Morrison. 1977c. Environmental impact of land use on
water quality - Executive Summary. 12 pp.
U.S. Army, Corps of Engineers. 1983. Summary report of the Lake Erie waste-
water management study. U.S. Army, Corps of Engineers, Buffalo District.
31 pp.
U.S. Environmental Protection Agency. 1984. Lake Erie Demonstration Pro-
jects: Evaluating impacts of conservation tillage on cost, yield and
environment. U.S. EPA, Great Lakes National Program Office, Chicago,
Illinois. 17 pp.
180
-------
MODELING THE FORMATION OF WATER QUALITY IN WATER CHANNELS
RECEIVING SURFACE RUN-OFF FROM FARM LAND
by
Ye. V. Yeremenko, V. Z. Kolpak and N. I. Selyu1
ABSTRACT
A multicomponent mathematical model is examined for the terming of water
quality in a river considering surface run-off from farmlands entering the
water course in the form of a continuous-in-length lateral influx of water
and matter formed in the watershed. In using the equations tor the balance
of matter, a system of linear equations has been derived tor a mathematical
model of transforming the nitrogen compounds and the content of dissolved
oxygen in the water. Proceeding from the supposition that low lateral in-
flux does not substantially change the amount of the average rate of the
water course and using coefficients ot nonconservati,veness within the limits
of the sections into which the water course is divided, a solution has been
obtained for the model's equations in an analytical form that can be used
in optimizing water conservation measures.
The necessary information on the amount ot lateral influx and the con-
centration of the designated matter in the surface run-oft has been obtained
on the basis ol the elaborated mathematical model tor the water quality ot
surface run-oft. Here the watershed is represented in the form ot a sys-
tematized aggregate, ol flat slopes.
Also discussed are the questions ol determining the calculation con-
ditions and requirements that should be satisfied by the surface run-off
models. As an example, results are given for calculating water quality in
the water course considering the lateral influx from the watershed.
INTRODUCTION AND DISCUSSION
Control of water quality in river basins and related setting ot maximum
permissible run-oft (.MPRJ ot drainage water depend largely on conditions ot
•'-All-Union Scientific Research Institute of Water Protection, USSR Ministry
of Water Management, Kharkov, USSR
181
-------
surface drainage and levels of substances in it for which standards are set.
In the most general case, there is migration into water channels of surface
run-off from farmland that occupies large territories, in the form of a
continuous line of influent water and mass of substances for which standards
are to be set.
Evaluation of organized dumping of drainage water and level of surface
run-off on water quality in a river can be made by means of appropriate
models of water quality. One of these numerous models is described by
leremenko et al. (1981). We shall discuss a multicomponent mathematical
model of formation of water quality, which takes into consideration the
role of surface run-off and is aimed at solving water-protection problems.
It permits determination of transformation of substances to be standardized
and development of elements of a transitional matrix to solve optimization
problems of controlling the quality of surface waters. In this model, the
substances to be standardized are the basic forms of nitrogen.
Without imposing restrictions on the law of change in water discharge
Q in a stream (a linear law of change in water discharge is discussed by
Aytsom et al. 196b) and on concentration of substances in surface water
NP , it is considered by Koval'cyuk et al. (1979) that NP corresponds to
background concentration in the stream and using the equation of balance of
matter in the section x, x+Ax, we shall have:
which, considering expansions to values on the order of O(Ax^) and Ax-H),
leads us to the equation:
Orf,- Q
This equation is the simplest one for consideration of the effect of surface
run-off on concentration of matter N. Its right part characterizes a
continuous positive (NN?) source of matter coming in
with surface run-off. Considering the nonconservative nature of matter,
the initial equation is presented as follows:
which yields the following system, as applied to the successive transforma-
tion of nitrogen compounds within a section of the stream;
182
-------
dx Q
U dQ ' .
where Mj^ , N2 » N3 > N4 and S are concentrations ot organic, ammonia, nitrite,
nitrate nitrogen and dissolved oxygen, 6, Y^, Y^ and Y^ are stoichiometric
coef iicients, Kj^ , K2 , Kj , ana K^ are coetticients ot nonconservation.
On the assumption that lateral intlow does not alter appreciably the
mean fiow rate and that the coetficients of nonconservation remain constant
within a section, the solution ol system 3 can be expressed as follows.
= G33 Nfl3 •»• G32 ^02 •*• &31 A/01 * frj
j •»• G-^jA/Qs •**G'«Mu f GV^OI +
S =Gss^>o ^GssA/fli ^GszA/oz +&Sf A/of *Gs ,
183
-------
where U shows values ot concentrations o± substances at the beginning ol
the section, Gij and Gsj are elements of the transitional matrix, values o±
which are listed in Appendix 1.
To determine concentrations of Ni and S, we need information about
lateral intlow Qp(x) and concentration of substances NP, which is provided
by the model of surface run-off.
Surface run-oti is termed in the water catchment area as a result.of
rain and it usually contains substances in both dissolved form with liquid
run-otr and on particles ot washed away soil, with solid discharge. Thus,
the model ot surface run-oft should provide information about both lateral
intiow ot water ana levels in it of soluble agents, as well as agents
carried in with solid discharge.
At the present time, there are methods in the USSR tor calculating
maximum discharge, rain flood (in water catchments less than 5U km^ in
size) and soil wash-off (Anon. 1975, Anon. 1979). Maximum discharge ot
rain floods is calculated for water catchment areas up to 5U km^ in size,
while determination ot soil wash-otf with rain run-oft from slopes is made
on areas up to / km^ in size (Anon. 1979). The restriction on water
catchment area makes it necessary to develop an approach without such a
restriction, and to view a water catchment area ot arbitrary size in the
torm ot a certain set of slopes in order to be able to use system 3 in
estimating soil wash-off. Hence, one of the requirements tor a model ot
formation of surface run-oft in a water catchment area is to take into
consideration the structure of the catchment area, gully and ravine network
over which the surlace run-ott tormed on slopes travels without run-oft.
As a rule, precipitation falling in a catchment area has spatial and
time variability. For this reason, the second requirement imposed on the
model is that it must take into consideration the spatial and time varia-
bility of precipitation (as a special case, they could be stationary).
finally, application ot fertilizers in soil is also performed repeat-
edly on farm tields. Hence the third requirement of models, that of consid-
eration ot spatial heterogeneity ot distribution ot fertilizers in the water
catchment area. It should be noted that the requirements imposed on a
model ol tormation ot rain-caused tlood regions and removal ol suspended
dritt trom water catchment areas do not impose restrictions on applicability
ot the model tor the case where these characteristics are constant in time
and space.
It these requirements are met, it will be possible to get an answer
to the questions of how and where torrential rain water and solid run-ott
enter the water system (distributed in length or concentrated) when making
estimates ol tormation ot run-otf in a water catchment area.
By calculating the quality ot surface waters tormed on slopes and tht
gully-ravine network, one can iearn how the concentrations of substances
will change in the run-oft as a function ot meteorological and agrotechnical
tactors, as well as environmental protection measures. The latter is
184
-------
particularly important in our opinion: to assess the efticacy ot different
environmental protection measures involving the entire water catchment area
or some segment of it according to volume o± substances carried out.
The moael of surface run-off consists of several submodels. Let us
discuss them.
SUBMODEL OF RUNOFF FORMATION
It is known that the equation of Nash (1957;, equation for kinematic
wave (Kuchment et al. 1983) and equations of Saint Venant can be used to
describe flow ot water down a slope (thaiweg). In our opinion, Nash's
equation is the most suitable for solving practical problems (.ordinary
first-order differential equation):
where QP is efflux ot water from slope, 1 is influx of water to slope
(including intensity of precipitations) and K is a coef licient .
However, the equation includes coefficient K, which is difficult to
determine, and although we do know from literature sources its expression
by hydroiic parameters and slope characteristics, its use with reference to
specific water catchment areas requires special validation. For this
reason, for a model or movement of torrential rain water on a slope, we
obtained an ordinary first-order differential equation that contains no
uncertain parameters. It is based on the equation of continuity with
consideration of lateral influx:
r. •>
where w is flow cross section, mz, QP is water discharge, m-Ys, I(t) and
r(t) are intensity or rain and inriltration, mm/min, & is width of free
surtace of flow, m, q(5oK is lateral intlux per unit current length (slope
or stream), m2/s, x is a coordinate read downstream, m, t is time, s.
By integrating equation b for stream length x from U to £c we oDtain.
did'
(7)
where
185
-------
q is water discharge in first section of stream with x = U, m Is, Q^b is
water discharge in last section of stream with x = c, m-Vs,
0
w is cross section of stream.
Since equation 7 has two unknowns, Qpb and w(.t), let us express Q^ b
as w(,t), using the lormuias of Chezy and Manning, analogously to derivation
of the kinematic wave equation from the equation of continuity:
where R(w) is hydraulic radius averaged for length of current, i is mean
slope of current and n is roughness coefficient.
Substituing equation b in equation "/ , we obtain an ordinary nonlinear
first-oraer difterentiai equation with reference to w (.the line over the w
was omitted for the sake of simplicity) ;
rr
The following initial condition is added to this equation;
Equation ^ is solved mathematically using well-known methods. As a
result of digital integration, we obtain discrete values for current cross
section w^ = jT, j = 1, 2..., T is the time gradation of integration. The
value of discharge is calculated using formula 8 with w-,. The integration
step T [gradation] in equation 9 is chosen from the condition of stability
of the computing process .
186
-------
In the case ol level run-oft trom a slope,
using the maximum (stationary) value tor w, on the basis ol established
water run-on ,
we obtain trom equation 11
f^~ f\
""•"" ^ ,
(12;
Thus, by integration ot equation y one can obtain a hydrograph ot run-
ort trom a slope. for the calculations i, n, qbx, 1, r and ic must be known.
SUbMUUbiL OJe WATER QUALITY
Determination ot volume (hydrograph) ot solid run-ott, which is deter-
mined entirely by liquid run-olf which, in turn, carries ott dissolved sub-
stances, is an important element in a model for quality ot water of surtace
run-ott .
In the model developed at the All-Union Scientific Research Institute
of Water Protection for formation ol water quality ol surface run-oil, use
is made ol the basic thesis in the literature (Anon. ly7y). Suspended
sediment is estimated using the formula,
n _
LCH-
CH
where Mch is the modulus ot run-off ot sediments with ¥7, coverage, tons/ha,
*'. is size ol water catchment areas, tun7, W is volume 01 rain run-ofl, m ,
C . is concentration ol suspended sediment in run-ofl , g/m . Parameters We
and M,,^ are expressed as follows.
where h is the layer ot flood run-olt at P/i. coverage, mm, a^ is parameter
of channeling, 3 is a coefficient which considers the inlluence ot the
agrotechnicai background for the preceding year, kj^ is the coefticient of
slope steepness. Parameters «^ and g are calculated using tables submitted
187
-------
in .the literature (.Anon. iy7b»). Inserting equation 14 in equation 13, we
shall obtain.
Tne data on liquid and solid run-ott permit determination or removal
ot suspended particles, ammonium nitrogen, and nitrate ol nitrogen. It is
assumed that nitrate is removed with liquid run-ott and ammonium nitrogen,
with solid run-oil.
Removal ot nitrogen in run-oll rrom larmland was measured by a method
developed at VNIIVO (.Ail-Union Scientilic Research Institute ol water
Protection) (.Anon. 19bi;. Nitrate concentration in surface run-orl per
area V was calculated using the iormula;
where Kno is the factor for scaling N to M03 , Kn°3 = 4. 4267,
Dc W*
r =~=r
where pc is removal ot nitrates in liquid run-oft, kg, m is amount of
nitrates in plowed layer betore rain, kg/ha, hm is depth of plowed layer,
m, II is porosity ot soil and H is layer of precipitations with PX coverage,
mm.
Ammonium nitrogen concentration in surface run-oft trom the water
catchment area is calculated using the following tormula:
U»)
mf — 1
where Kn 4 is a factor ror scaling M to NIL and is equal to 1.2b^7. PT is
removal of nitrogen with solid run-oft, kg.
where na is the quantity of metabolic-absorbed forms of nitrogen in the
plowed layer by the time rains begin, kg/ha, Mch is the modulus ot sediment
run-ott (.calculated using formula 14) and y is volumetric weight of soil in
tilled layer, tons/in-*- Thus, the model ot formation of quality ot surface
run-off from farmed land consists of a set ot submodels. the run-ott
188
-------
submodels permits determination ol the hydrograph of torrential water run-
ott, while submodels ot water catchment area) ot suspended sediments,
ammonium nitrogen and nitrates.
The appropriate programs for a computer were developed tor use 01 the
above described submodels.
Let us discuss brietly determination ot nominal conditions. Accord-
ing to the "Rules tor protection of surtace waters against sewage pollution,"
which are in effect in the USSR, calculation ot volumes of diversiop of
sewage into water systems should be made for adverse hydrologicai condi-
tions, to which average monthly water discharge with b»5/i coverage is usually
related. however, the question of layer of rain run-off corresponding to
this water flow rate in the river remained unanswered.
Analysis ol tormation ot water quality in surface run-otf from tarm
land revealed that the most hazardous concentrations ot pollutants in the
annual cycle are observed during the period ol maximum rain activity. For
this reason, in developing the nominal hydrologicai conditions tor removal
of pollutants from rarmland, it was necessary to determine the probability
ol occurrence of two random processes together—maximum layer of raintlood
run-off and mean summer-tall low-water level of discharge in rivers situated
in zones with different moisture factors.
Studies of concurrent appearance of maximum layer of rain flood and
minimum discharge ot water with ^5% coverage in the spring and fall low-
water period revealed that a layer of rain run-off with 25% coverage
(.Kovalenko and Karabash 1983) should be taken as the nominal value.
As an example, let us estimate removal of ammonium nitrogen, nitrates
and suspended particles in surface run-off trom a hypothetical water catch-
ment area (.Figure 1). This water catchment has the following dimensions:
iU km on the y-axis and 2U km on the x-axis. The length or the river within
the limits ot the part of the catchment area in question is 27 km, and the
entire river is 5U km long. Estimates of surface run-off were made using
the tollowing base data. H—layer of precipitations was considered to be
3D mm, n—coefficient of roughness tor catchment slopes U.2 and for thalwegs
U.U2, r—rate ol absorption was considered to be U.2U mm/mm and constant
in time, II—porosity ot soil, U.iU, m—nitrate content in tilled layer ttU
kg/ha, hm—depth ot tilled layer U.2 m, na—ammonium nitrogen content in
tilled layer i> kg/ha, F—size ot water catchment area 2UU km^ .
On the basis of tne developed programs, a geometrical model ot the
catchment area and run-off forming chains ot triangles on the left and
right banks of the stream, as Weil as run-off of torrential rain on the
basis ol the geometric model of a water catchment area, were constructed.
To simplify input of information in calculating the concentraton ol substances
in the stream, several separate catchments, the run-off hydrographs ot which
are illustrated in Figure 2, were singled out in the model of the water
catchment area. Run-oft from the separate catchments 2 and 6 is given as
being concentrated (over girder system) and trom the others, deconcentrated
(.along stream) (.Figure 3).
189
-------
RUN-OFF
RUN-OFF
figure 1. Fragment of hypothetical water catchment area (.horizontals, m),
190
-------
-7=CATCHMENT
NUMBERS
234
Time, hours
iigure 2. Calculated hyarographs ol rain water runoll trom separate catch-
ments.
1
1
v >
4
k v
f A
i i
1
i
}
5 6
i
k
'
^
7
^
i
)
~\
i
2
E
1
r
3 1-7 = CATCHMENT NUMBERS
0 5 10 15 20 25 30 35 40 45 50
Distance, km
Figure J. Line diagram ot river section.
191
-------
Averaged water discharge oi surtace run-oti and concentrations ot
suspended substances, ammonium nitrogen and nitrates are listed in Table 1.
TABLE 1. DISCHARGE OF RAIN RUN-OiF AND CONCENTRATIONS OF AMMONIUM NITROGEN,
NITRATES AND SUSPENDED SUBSTANCES ENTERING STREAM.
Separate
catchment
number
1
2
3
4
5
b
7
Flow rate,
m-Vs
0.26
0.40
U.3j
u.25
0.64
U.37
U.Ob
Ammonium
nitrogen
0.15
0.17
0.13
U.lb
0.13
O.i3
0.14
Concentrations , mg/
Nitrates
74
bl
67
79
65
65
7U
Suspended
substances
1200
14UO
ibOO
1350
1560
1350
1000
formulas 3 were used to calculate concentrations oi substances in
river within the limits ot the section studied, we used the tormulas similar
to the one ror N^ (3) to estimate the concentrations ot suspended substances
(Nu,) under the conditions listed in Table 1, as well as with the toilowing
base data.
background characteristics:
Q = 40 m3/s, N! = 2 mg/i, NZ = 0.25 mg/£, N^ = 0.01 mg/£, N4 = 5 mg/ ,
S = 10 mg/£ and N^ = 13 mg/£.
Characteristics ot concentrated release E:
3
Q = 0.3 m/s, Nx = 3.2 ng/A , N2 = 0.5 mg/£, N3 = 0.25
S = U mg/i and Nu,=100 mm/ 1.
Nonconservation coelticients .
= 13 mg/£,
,-b
= 1 ' 1U ul/s, K2 = 1.2b'10 'l/s, K3 = 1 .25 'l
= 1 .15 '
Ks =
s and
2'10 7i/s.
192
-------
Calculations were made in two variants. The first corresponds to
lateral inllow or water as indicated in Table 1, and the second tripled
lateral intlow, which also failed to lead to substantial change in concen-
tration of substances in the flow of water. The results of calculations
made in the first variant are illustrated in Figure 4, whereas Table 2
lists comparative'concentrations of substances in typical sites of the
river calculated by the two variants.
I. BODs (N|), mg/l 4.Suspended Substances (N5), mg/l
2.02(S), mg/l 5. Nitrate Nitrogen (Nil, mg/l '
3. Nitrite Nitrogen (N3), I03-mg/l 6. Amonium Nitrogen (N2),mg/l
N= SNL N, No N,
2.0
173
153
133
1 13
93
73
53
33
13
'lOO
9.0
10
9
8
7
6
5
100
8.0
6.0
4.0
2.0
n?o
.027
0.26
025
I 9
0 2 4 6 8 10 12 14 16 18202224262830323436384042444648
Distance, k'm
Figure 4. Calculated concentrations of substances in river current.
It must be noted that this approach permits not only consideration of
the effect on water quality in the river of surface run-off, but allows an
approach to optimization of water protection measures in a catchment provided
the required amount of water is furnished in the river due to presence of
elements of transitional matrix, if the link between intensity of water
protection measures and cost indicators is Known.
TABLE 2. CONCENTRATIONS OF SUBSTANCES IN TYPICAL SITES OF RIVER CURRENT
WITH DIFFERENT LATERAL INFLOW
Calcul .
variant
No.
Distance
from 1st
sect. , km N-,
Concentration, mg/ &
N^'IO N..J-100 N4 N3'10 1
S
10 1.94 2.52 1.27 b.bl 2.b7 9.i>b
17 i.b7 2.5b 0.42 b.97 10.03 9.4
193
-------
TAB'I.H, 7, Continued
24
2.1
37
50
•L 10
17
24
27
37
5U
1.82
l.bU
1.77
1.73
i.»y
1.7b
l.bb
1 .b2
i.5y
1.57
2.bU
2.5y
2.b2
2.b7
2.50
2.5b
2.b3
2.bU
2.b2
2.b7
o.2y
0.26
U.27
0.28
i.U2
0.40
0.29
0.27
0.2b
0.28
11.26
11.88
11. bb
11. b4
7.b3
ib.iy
22.30
23. b4
23. bO
23.47
15.70
17.07
17.01
16. yu
5.y3
25. by
41.06
43. y8
43.tt2
43.42
y.2b
y.i4
9.12
y .05
y.4b
8.yi
8.67
tt.44
b.54
8.61
Key. N^) BODs LDiological oxygen demand] N^) nitrate nitrogen
No) ammonium nitrogen N^) suspended substances
N3) nitrite nitrogen S) dissolved oxygen
Note; Typical sites in the river were selected at the ends of the catchment
sections before concentrated releases ana at the end of the nominal
section.
BIBLIOGRAPHY
Anon. 1975. Metodicheskiye rekomendatsii po uchetu poverkhnostnogo stoka i
smyva pochv pri izuchenn vodnoy erozii (.Methodological Recommendations
on how to Keep a Record of Surface Run-Oft and Soil Wash-Off in Studies
of Aquatic Krosion;. Leningrad, USSR, bb.
Anon. iy79. Instruktsiya po opredeleniye raschetnykh gidrologicheskikh
kharatctenstik pri proyektirovanii protivoerozionnykh meropriyatiy
na Yevropeyskoy territorii SSSR (.Instructions for Determining Nominal
Hydrological Characteristics in Planning Anti-Lrosion Measures in
European USSR). Leningrad, USSR. 63 p.
Anon. lybl. Metodika otsenki vynosa pestitsidov i biogennykh veshchestv
neorganizovannym poverkhnostnym stokom sel'skokhozyaystvennykh ugodiy
Dogarnogo zemleaeliya (.Methods of Evaluating Removal of Pesticides
and Biogenous Substances in Unorganized Surface Run-Off From Land
Farmed by the ISionirrigated Method). Kharkov, USSR. b4 p.
Aytsam, A. M., Kh. A. Vel'ner, and L. L. Paal'. iyb8. Calculation of
changes in pollutant concentrations in rivers, Gigiyena 1 Sanitariya.
11:3-10.
194
-------
Kovalenko , M S. and G. A. Karabash. 1953. Evaluation ol eilect ot removal
ot pesticides in surtace run-olt on water quality in streams. In:
Okhrana vody ot zagryazneniya poverkhnostnym stokom (.Protection oi
Water Against Pollution by Surtace Run-Oti). Kharkov, USSR. pp bO-y2.
Kovai'cyuk, P- 1., B. A. Akisin, V- L . Pavelko, and A. A. Matveyev. lS»7y.
Synthesis oi models ol transformation ot nonconservation substances in
streams during diversions in a system ot tield observations. Gidrokh.
Mater. 72 . il 1-ilb .
Kuchment, L. S., V. N . Demiaov , and Yu. G. Motovilov iyb3. Formirovaniye
rechnogo stoka (Formation ot River Run-Oli). Moscow, USSR. 2ib.
Nash, J. h. iyi>7 . The rorm ol the instantaneous unit hydrograph.
3(45;; 14-121.
Yeremenko, Ye. V., Yu. M. Plis, and N. 1. Selyuk. iy«l. Model ot water
quality in Connecticut and North Donets Rivers. In: Metodologiya i
praktika planirovaniya okhrany vod rechnykh basseynov (.Methodology and
Practice ot Planning Protection for River Basin Waters). Kharkov,
USSR. pp 120-170.
APPENDIX
Values for elements of transition matrices in solutions of (3)
195
-------
= 7T
U
31
3
196
-------
I2 ir-r
U S 5
Q K3-
'V
Q
Q ^-
197
-------
-£,
* -
X) \
c
198
-------
-a,)
199
-------
^ -O*
where
Translator's note: Subscript "cp" is Russian abbreviationfor average
or mean.
200
-------
CONTROL OF URBAN NONPOINT SOURCES IN GREAT LAKES BASIN
by
D. Athayde, P. Bubar, and J. Meek1
ABSTRACT
The Nationwide Urban Runoff Program (NURP) was conducted by the U.S.
Environmental Protection Agency (EPA) and many cooperating federal, state,
regional, and local agencies which were distributed across the United States.
The program consisted of individual project studies that were overseen by a
technical team at EPA. The program was managed centrally because it covered
a broad spectrum of technical planning issues at many geographic locations,
several of which were in the Great Lakes Basin. In 1982, the International
Joint Commission (IJC) established a Nonpoint Source Control Task Force to
review and evaluate the effectiveness of pollution reduction activities recom-
mended by the commission. An August 1983 report provided an overview of the
extent of implementation and the effectiveness of various nonpoint source
programs in the Great Lakes Basin. This paper summarizes the results of the
NURP projects in the Great Lakes Basin and compares them to the findings of
the IJC's Nonpoint Source Task Force.
This paper has been reviewed in accordance with the U.S. Environmental
Protection Agency's peer and administrative review policies and approved for
presentation and publication.
INTRODUCTION
The Nationwide Urban Runoff Program (NURP) was initiated as part of the
water quality planning process required under Section 208 of the U.S. Clean
Water Act of 1977- The program began in 1978 and concluded in 1983. NURP
had six projects in four states within the Great Lakes Basin out of a total
of 28 demonstration projects participating. Results from five of those six
projects are now available and will be discussed in this paper.
-'-Office of Water, U.S. Environmental Protection Agency, Washington DC 20460
USA
201
-------
The NURP was conducted by the U.S. Environmental Protection Agency (EPA)
and many cooperating federal, state, regional, and local agencies distributed
widely across the United States. The program consisted of individual project
studies that were overseen by a technical team at EPA. The program was mana-
ged centrally because it covered a broad spectrum of technical and planning
issues at many geographic locations, several of which were in the Great Lakes
Basin.
The program was developed, implemented and managed by the Office of
Water's Water Planning Division at EPA with additional contributions from
EPA's Office of Research and Development and the EPA regional offices. The
data summary and analysis was done mainly by a team of consultants under con-
tract with EPA.
In 1972, the Pollution from Land Use Reference Group (PLUARG) of the
International Joint Commission (IJC) was established for the purpose of de-
termining the levels and causes of pollution from land use activities in the
Great Lakes Basin. PLUARG reported its findings and recommendations to the
IJC in 1978. Subsequently, the IJC forwarded a set of recommendations to the
U.S. and Canadian governments in 1980.
In January 1982, the Water Quality Programs Committee of the IJC' s
Water Quality Board recommended that a Nonpoint Source Control Task Force be
established. Consequently, the Water Quality Board set up a 14-member Task
Force (seven members each from the United States and Canada) . The Task Force
was to review and evaluate the effectiveness of the activities recommended by
the IJC to reduce nonpoint source pollution. The Task Force did this and
prepared a report in August 1983. The report provided an overview of the
extent of implementation and the effectiveness of various nonpoint source
programs in the Great Lakes Basin.
This paper summarizes the results of the NURP projects in the Great
Lakes Basin and compares them to the findings of the IJC's Nonpoint Source
Task Force.
BACKGROUND OF NATIONWIDE URBAN RUNOFF PROGRAM
The NURP was begun in 1978 in an effort to provide the U.S. government
with answers to questions concerning urban runoff including: (1) Is It a
problem? (2) What are its effects on receiving waters? and (3) What control
mechanisms are available?
The program consisted of 28 projects for collecting and analyzing data
in an effort to answer the local questions that existed as well as to provide
information to U.S. Environmental Protection Agency (EPA). Because there
were local objectives that had to be met, EPA played a strong management role
to ensure that each project was not off on its own tangent. This role called
for providing technical assistance in work plan development, in data collection
and instrumentation programs, in data analysis, and in report writing. The
result of this approach was 28 projects working separately but also within
the direction of the NURP team.
202
-------
A major objective of the NURP was the acquisition of data. The data
would be used to characterize the urban runoff problem, evaluate receiving
water impacts, and evaluate management practices. Because of this, quality
assurance and quality control were important elements in each project. There
was considerable effort to ensure that the NURP data were valid. Questionable
data would be of no use.
NURP collected data on an event basis with anywhere from 12 to 30 events
at one site. Both discrete samples and composite samples were collected,
usually by an automatic sampler. Rainfall volume, runoff volume, and runoff
quality data were collected from a number of urban catchments in each city.
The event mean concentration (EMC), defined as the total constituent mass
discharge divided by the total runoff volume, was chosen as the primary mea-
sure of the pollutant load. EMCs were calculated for each event at each site
in the accessible data base. If a flow-weighted composite sample was taken,
its concentration was used to represent the EMC. Where sequential discrete
samples were taken over the hydrograph, the EMC was determined by calculating
the area under the loadograph (the curve of concentration times discharge rate
over time) and dividing it by the area under the hydrograph (the curve of
runoff volume over time). For the purpose of determining EMCs, rainfall events
were defined to be separate precipitation events when there was an interven-
ing time period of at least 6 hours without rain.
EMCs were chosen as the primary water quality characteristic subjected
to detailed analysis, even though it is recognized that mass loading char-
acteristics of urban runoff (e.g., pounds/acre for a specified time interval)
is ultimately the relevant factor in many situations. The reason is that,
unlike EMCs, mass loadings are strongly influenced by the amount of precipi-
tation and runoff, and estimates of typical annual mass loads will be biased
by the size of monitored storm events. The most reliable basis for charac-
terizing annual or seasonal mass loads is using EMCs and site-specific rain-
fall/runoff characteristics. In addition, EMCs were chosen because we were
not interested in looking at one storm event but rather many storm events
over long periods of time. What occurs during one storm event held little
interest to us.
A statistical approach was adopted for characterizing the properties of
EMCs for standard pollutants. Standard statistical procedures were used to
define the probability distribution, central tendency (a mean or median) and
spread (standard deviation or coefficient of variation) of EMC data. EMC
data for each pollutant from all storms and monitoring sites were compiled
in a central data base management system at EPA.
The underlying probability distribution of the EMC data was examined
and tested by the NURP team and consultants, by both visual and statistical
methods. With relatively few isolated exceptions, it was found that the
probability distribution of EMCs at individual sites can be characterized by
log-normal distributions. Because of this, the log (base e) transforms of
all urban runoff data were used in developing the statistical characteriza-
tions.
203
-------
In addition to summarizing and analyzing the concentration data, sever-
al management control practices were evaluated and analyzed. The management
practices analyzed were the use of detention basins and street sweeping. De-
tention basins were also analyzed by the Nonpoint Source Control Task Force.
SUMMARY OF NURP PROJECTS IN GREAT LAKES BASIN
ASSESSMENT OF CONCENTRATIONS AND LOADS IN NURP PROJECTS
As mentioned previously, six projects participating in the NURP were
located in the Great Lakes Basin. Five of the six project's data have been
analyzed by the NURP. These five areas were Ann Arbor, Michigan; Lansing,
Michigan; Glen Ellyn, Illinois; Milwaukee, Wisconsin; and Rochester, New
York (see Figure 1).
The data analysis focused on ten pollutants, which were selected after
initial screening of the data. The pollutants were total suspended solids
(TSS), biochemical oxygen demand (BOD), chemical oxygen demand(COD), total
phosphorus (as P)(TP), soluble phosphorus (as P) (SP) , total kjeldahl nitro-
gen (as N) (TKN), N - nitrite + nitrate (as N) (N02+3), total copper (Cu) ,
total lead (Pb), and total zinc (Zn).
Figure 1. Nationwide Urban Runoff Program locations in United States
204
-------
The NURP analysis summarized the statistical parameters—mean, median,
and coefficient of variation. The data from the 22 sites in the 5 representa-
tive NURP projects are shown in Table 1.
TABLE 1. STATISTICAL PARAMETERS FOR 22 SITES IN GREAT LAKES NURP
-
Pollutant No.
and Project observ.
Total Susp .
Solids (yg/1)
NY3
NY3
WI1
WI1
WI1
IL2
NY3
NY3
MI 3
Mil
Mil
MI 3
MI 3
Mil
MI 3
WI1
WI1
WI1
NY3
WI1
Mil
Mil
Total Phos-
phorus (yg/1)
NY3
NY3
WI1
WI1
WI1
IL2
NY3
NY3
MI 3
Mil
Mil
MI 3
MIS
Mil
10
7
45
33
23
27
11
9
5
35
23
6
6
23
5
47
58
42
12
29
18
20
13
8
45
35
23
26
13
12
5
35
22
6
6
23
Coeffic.
Mean variation
134
294
266
170
251
250
154
63
33
85
158
46
68
172
80
383
212
202
141
412
•92
188
301
448
229
258
453
506
193
195
91
198
458
103
268
394
1.15
1.12
0.44
0.68
0.69
0.75
0.92
0.74
0.77
1.28
1.26
0.37
0.47
0.85
0.91
0.78
0.86
0.68
0.76
0.97
0.82
0.94
0.54
0.47
0.45
0.51
0.69
0.79
0.46
0.47
0.38
0.64
0.65
0.50
0.47
0.54
90 percent
Median confidence limits
88
196
243
141
206
200
113
51
26
52
98
43
61
131
59
302,
161
167
112
296
71
137
265
405
209
230
373
397
175
177
85
167
384
93
243
347
52-150
85-263
219-270
117-169
165-258
161-249
74-173
34-77
14-50
39-69
69-139
32-58
42-88
101-171
28-124
255-357
131-197
142-196
79-159
229-382
53-95
101-186
206-340
300-456
188-233
201-264
298-466
314-501
141-217
140-223
60-121
141-197
309-477
63-137
168-351
285-410
205
-------
TABLE 1 (cont'd). STATISTICAL PARAMETETERS FOR 22 SITES IN GREAT LAKES NURP
Pollutant
and Project
MI 3
WI1
WI1
WI1
NY3
WI1
Mil
Mil
Soluble Phos-
phorus (yg/1)
NY3
NY3
WI1
WI1
WII
IL2
NY3
NY3
MI 3
Mil
Mil
MI 3
MI 3
Mil
MI3
WII
WII
WII
NY3
WII
Mil
Mil
TKN (yg/1)
NY3
NY 3
WII
WII
WII
IL2
NY3
NY3
MI 3
Mil
No.
observ.
5
47
60
44
12
29
18
17
0
0
0
0
0
24
0
0
5
32
20
6
6
21
5
0
0
0
0
0
14
16
13
7
45
15
1
0
13
10
5
35
Mean
134
289
108
105
216
511
546
435
-
-
-
-
-
98
-
-
33
43
68
13
59
47
39
-
—
_
_
_
127
59
1492
3246
1260
1102
_
1099
1111
889
1490
Coeffic.
variation
0.56
0.59
0.56
0.79
0.26
1.19
0.58
0.71
-
_
-
-
_
1.21
_
_
0.55
0.76
0.68
0.37
0.88
0.47
0.46
_
_
_
0.72
1.24
0.45
0.90
0.50
0.54
_
0.50
0.36
0.11
0.53
Median
117
249
94
82
209
330
472
355
—
_
_
_
_
63
_
_
29
34
56
13
44
42
35
_
103
37
1358
2411
1125
969
982
1045
883
1316
90 percent
confidence limits
71-193
218-284
84-105
69-98
183-239
245-443
378-589
271-465
-
_
_
—
—
45-88
18-47
28-42
44-71
10-17
24-82
45-50
23-53
76-140
24-56
1098-1679
1369-4245
908-1395
801-1173
778-1240
854-1279
796-981
1142-1516
206
-------
TABLE 1 (cont'd). STATISTICAL PARAMETERS FOR 22 SITES IN GREAT LAKES NURP
Pollutant No.
and Project observ.
Mil
MI 3
MIS
Mil
MI3
WI1
WI1
WI1
NY3
WI1
Mil
Mil
Nitrite plus
nitrate (yg/1)
NY3
NY3
WI1
WI1
WI1
IL2
NY3
NY3
MI 3
Mil
Mil
MIS
MIS
Mil
MI 3
WI1
WI1
WI1
NTS
WI1
Mil
Mil
Total copper
(yg/l)
NY3
NY3
WI1
WI1
WI1
23
6
6
23
5
16
27
25
13
8
18
18
0
0
18
24
3
21
0
0
5
35
23
6
5
23
5
17
28
26
0
12
18
17
0
0
0
0
0
Mean
1631
845
1056
1988
1116
1452
1023
1073
1256
1656
1274
1713
-
-
775
625
-
796
-
-
1108
775
883
284
469
875
1033
751
708
781
-
783
686
742
-
-
-
-
-
Coeffic.
variation
0.42
0.29
0.22
0.47
0.15
0.35
0.44
0.61
0.45
0.65
0.57
0.56
-
-
0.48
0.39
-
0.55
-
-
0.17
0.49
0.44
0.48
0.24
0.43
0.76
0.69
0.68
0.69
-
0.50
0.40
0.52
-
-
-
-
-
Median
1506
811
1031
1802
1104
1369
936
916
1144
1389
1107
1493
-
-
699
582
-
699
-
-
1092
696
807
256
456
803
821
618
584
642
-
702
637
657
—
-
-
-
-
90 percent
confidence limits
1304-1740
642-1025
862-1233
1536-2115
958-1273
1180-1589
815-1075
755-1110
925-1414
933-2068
891-1376
1205-1850
-
-
580-843
510-664
-
576-848
—
-
930-1283
192-558
694-938
176-372
364-571
693-931
431-1563
474-805
479-712
520-791
—
549-897
544-746
534-808
~
—
—
—
-
207
-------
TABLE 1 (cont'd). STATISTICAL PARAMETERS FOR 22 SITES IN GREAT LAKES NURP
Pollutant
and Project
IL2
NY3
NY3
MI3
Mil
Mil
MIS
MIS
Mil
MIS
WI1
WI1
WI1
NY3
WI1
Mil
Mil
Total lead
(Mg/1)
NY3
NY3
WI1
WI1
WI1
IL2
NY3
NY3
MI 3
Mil
Mil
MI 3
MI 3
Mil
MIS
WI1
WI1
WI1
NY3
WI1
Mil
Mil
No.
observ.
26
0
0
0
16
13
0
0
9
0
0
0
0
0
0
6
7
13
87
44
35
22
24
10
12
0
24
18
6
5
18
4
45
59
44
13
27
13
13
Mean
49
-
-
-
15
30
-
-
14
-
-
-
-
-
-
36
25
34
193
95
108
303
322.
12
35
-
Ill
122
21
61
170
582
193
121
47
409
116
115
Coeffic.
variation
0.53
-
-
-
0.54
0.63
-
-
0.31
-
-
-
-
-
-
0.53
0.65
0.77
0.89
0.72
0.67
1.14
1.01
0.42
1.65
_
1.09
0.90
1.63
0.71
1.39
0.94
0.83
0.73
0.50
0.86
0.77
0.76
Median
43
-
-
-
13
26
-
-
13
-
-
-
-
_
-
32
21
27
144
77
90
200
227
11
18
_
75
91
11
50
99
424
148
98
42
310
92
92
90 percent
confidence limits
36-51
-
-
-
10-16
20-35
-
-
11-16
-
-
_
_
_
_
21-48
14-32
19-38
86-240
65-91
75-107
143-280
169-304
9-14
10-33
55-102
66-125
4-28
27-92
65-151
348-517
126-173
83-115
33-53
243-396
66-129
66-128
208
-------
TABLE 1 (cont'd). STATISTICAL PARAMETERS FOR 22 SITES IN GREAT LAKES NURP
Pollutant
and
No.
Coeffic.
Project observ. Mean
vari
ation Median
90 percent
confidence limits
Total zinc
(yg/D
NY3
NY3
WI1
WI1
WI1
IL2
NY3
NY3
MI 3
Mil
Mil
MI3
MI 3
Mil
MI 3
WI1
WI1
WI1
NY3
WI1
Mil
Mil
*Site
Proj.
NY3
NY3
WI1
WI1
WI1
IL2
NY3
NY3
MIS
Mil
Mil
MI 3
MI 3
Mil
MI3
WI1
WI1
9
8
18
21
0
27
9
9
2
17
14
4
4
9
2
27
32
19
9
7
7
27
description:
415
488
106
108
_
230
792
1063
-
121
245
-
-
149
-
476
145
156
1416
280
244
223
Site Land use %
Cranston
E. Roch.
Bur bank
Hastings
Lincoln
Comb . Inl
Thornell
Thomas Cr
Traver
Waverly
Grand R.
Pitt AAS
Pitt AAN
Grace N
Swift Run
Wood Ctr
Post Off
100% Res
100% Res
100% Res
100% Res
97% Res.
85% Res.
.
.
.
.
100% Open
91% Open
90% Open
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
100% Comm
0.
1.
1.
1.
_
0.
2.
3.
-
0.
0.
-
-
0.
-
1.
1.
0.
2.
0.
0.
0.
Area
(acres)
166
346
63
33
36
524
28,416
17,728
2,303
30
453
2,001
2,871
164
1,207
45
12
88
10
34
20
69
39
14
45
71
35
21
16
75
55
66
42
54
Pop
(no
. Dens .
. /acre)
5
18
15
17
18
8
-
1
-
11
5
2
7
5
2
12
0
312
327
63
69
189
306
322
-
110
200
-
-
140
-
303
94
125
517
234
225
196
%
Im
22
38
50
51
57
17
4
11
6
68
38
21
26
28
4
81
195-499
180-594,
42-95
49-99
_
154-232
130-720
124-839
_
92-132
148-271
-
-
113-173
-
222-414
71-124
96-163
214-1247
150-363
167-303
135-284
100
209
-------
TABLE 1 (cont'd). STATISTICAL PARAMETERS FOR 22 SITES IN GREAT LAKES NURP
Site description (cont'M) :
Pro j .
WI1.
NY3
WI1
Mil
Mil
Site Land use %
Rustler 100% Comm
Southgate 100% Comm
State Fr 74% Comm
Indus Dr 100% Ind
Grace S. 52% Ind
Area
(acres)
12
179
29
63
75
Pop. Dens.
(no. /acre)
0
2
10
0
5
% Imp.
100
21
77
64
39
The runoff coefficient statistics also were calculated for the repre-
sented sites (Table 2).
TABLE 2. RUNOFF COEFFICIENT STATISTICS FOR GREAT LAKES NURP SITES
Proj .
NY3
NY3
WI1
WI1
WI1
IL2
NY3
NY3
MI3
Mil
Mil
MI3
MI 3
Mil
MI3
WI1
WI1
WI1
NY3
WI1
•Mil
Mil
Site
Cranston
E. Roch.
Burbank
Hastings
Lincoln
Comb. Inl
Thornell
Thomas Cr
Traver
Waverly
Grand R.
Pitt AAS
Pitt AAN
Grace N
Swift Run
Wood Ctr
Post Off
Rustler
Southgate
State Fr
Indus . Dr
Grace S.
Land
use
%
100% Res.
100% Res.
100% Res.
100% Res.
97% Res.
85% Res.
100% Open
91% Open
90% Open
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
100% Comm
100% Comm
100% Comm
74% Comm
100% Ind
52% Ind
Drainage
area
(acres)
166
346
63
33
36
524
28,416
17,728
2,303
30
453
2,001
2,871
164
1,207
45
12
12
179
29
63
75
Pop.
dens.
(no. /acre)
5
18
15
17
18
8
_
1
-
11
5
2
7
5
2
12
0
0
2
10
0
5
% Imp.
22
38
50
51
57
17
4
11
6
68
38
21
26
28
4
81
100
100
21
77
64
39
No.
obs .
13
9
44
33
19
29
13
13
5
35
23
6
6
23
5
44
54
39
13
27
18
20
Runoff
Median
0.16
0.20
0.27
0.27
0.38
0.17
0.06
0.04
0.11
0.36
0.11
0.19
0.10
0.11
0.21
0.76
0.90
0.79
0.20
0.62
0.10
0.11
Coeffic.
Coeff
variation
0.33
0.42
0.92
0.37
0.55
0.48
0.93
0.56
1.07
0.25
0.50
0.46
0.43
0.41
0.38
0.42
0.19
0.19
0.28
0.24
0.71
0.43
The site mean EMC values are useful but the loading values are usually
what get published in the literature. Using the event mean concentrations and
assuming 100 events per year, annual loads from the Great Lakes projects were
calculated by the NURP team. We used 100 events per year basically because it
210
-------
simplified our analysis. The number of storm events that occur annually in
the Great Lakes Basin, however , ranges from 90 to 120. Loadings from the NURP
Great Lakes sites are summarized in Table 3.
TABLE 3. LOADINGS FROM NURP GREAT LAKES SITES
Pollutant Mean event load (kg/ha)
Total suspended solids
Total phosphorus
Soluble phosphorus
TKN
N02 + N03
Total copper
Total lead
Total zinc
4.7658
0.007836
0.000979
0.037096
0.019934
0.000597
0.004245
0.006399
Annual load (kg/ha/yr)
477
0.78
0.10
3.7
2.0
0.06
0.42
0.64
Table 3 includes all land uses taken together. We found, at the 95%
confidence level, that there is no significant difference in pollutant load-
ing values between the various land uses.
The Nonpoint Source Control Task Force of the IJC also published load-
ing values based on their pilot watershed studies. Table 4 shows that the
NURP numbers compare well with the PLUARG data.
Annual urban runoff loads for total phosphorus from the entire Great
Lakes Basin were compared between the two studies. Both showed that urban
runoff accounted for approximately 7% of the nonpoint load and 3% of the
total load.
ASSESSMENT OF CONTROL PRACTICES IN GREAT LAKES BASIN
In addition to assessing the concentrations of urban runoff, the Michi-
gan, Illinois, and Wisconsin projects all evaluated the effectiveness of
either street sweeping or detention basins. In fact, control practice evalu-
ation was the focus of these projects. Milwaukee, Wisconsin, evaluated street
sweeping. The primary purpose of this project was to evaluate the potential
improvement in stormwater quality caused by an accelerated street sweeping
program. The project selected pairs of small, homogeneous watersheds. One
basin served as the test area and one as the control area. The control area
was swept using the same baseline frequency at which it had customarily been
211
-------
TABLE 4. SUMMARY OF RANGES OF UNIT AREA LOADS OF SELECTED MATERIAL BY LAND USE FROM PILOT WATERSHED STUDIES
DONE BY PLUARG
Annual unit area loads (kg/ha/yr)
Land uses
I Rural
General agriculture
Cropland
Improved pasture
Forest/wooded
Idle/ perennial
Sewage sludge
Waste water spray
Irrigation
II Urban
General urban
Residential
Commercial
Industrial
Developing urban
Suspended
solids
3-5600
20-5100
30-80
1-820
7-820
-
-
210-1750
620-2300
50-830
400-1700
27.500
Total
phosphorus
0.1-9.1
0.2-4.6
0.0-0.5
0.02-0.67
0.02-0.67
0.02
0.2-1.4
0.3-2.1
0.4-1.3
0.1-0.9
0.9-4.1
23
Filtered
reactive
phosphorus
0.01-0
0.05-0
0.02-0
0.01-0
0.01-0
0.01
0.1-1.
0.05-0
0.2
0.02-0
0.3
0.1
.6
.4
.2
.10
.07
3
.3
.08
Total
nitrogen
0.6-42
4.3-31
3.2-14
1-6.3
0.5-6.0
11
2-2.37
6.2-10
5-7.3
1.9-11
1.9-14
63.0
0
0
0
0
0
0
-
0
0
0
2
-
Lead
.002-0.08
.005-0.006
.001-0.015
.01-0.03
.01-0.03
.01
.14-0.5
.06
.17-1.10
.2-7.0
Copper
0.002-0.09
0.014-0.064
0.021-0.038
0.02-0.03
0.02-0.03
0.005
-
0.05-0.13
0.03
0.07-0.13
0.29-1.3
-
Zinc Chlorine
0.
0.
0.
0.
0,
0,
0
0
0
3
-
005-0.3
,026-0.083
,019-0.172
,01-0.03
,01-0.03
.2
.3-0.6
.02
.25-0.43
.5-12.0
10-120
10-50
-
2-20
20-35
10
40-160
130-380
1.050
10-150
-
75-160
-------
been swept. The test areas were swept at frequencies that were higher than
in the control areas. The accelerated sweeping frequencies were selected to
represent the possible range of sweeping frequencies that might be socially
and economically acceptable.
A summary of street sweeping performance for the two Wisconsin basins
is shown in Figure 2.
State Fair,
Wisconsin
Rustler,
Wisconsin
Pb
TKN
TP
COD
TSS
100 -50 0 +50 +100
% EMC Reduction
Figure 2. A summary of street sweeping performance.
As shown in the figure, the mean concentrations are as likely to be in-
creased as decreased by street sweeping. Also, it is shown that street
sweeping never produced a dramatic (i.e., >50%) reduction in concentrations.
The analysis of the NURP street sweeping data base indicated that there
is generally no significant difference between median EMCs for swept and un-
swept conditions. This is shown in Figure 3.
Ann Arbor, Michigan, Lansing, Michigan, and Glen Ellyn, Illinois^evalua-
ted the effectiveness of detention basins. The basins monitored were "wet"
basins— that is, they maintain a permanent pool of water. The ponds were
designed such that runoff from an individual storm displaces all or part of
the prior volume, and the residual is retained until the next storm event.
The data from the basins were analyzed and it was determined that the
basins were fairly effective in removing pollutants. The performance efficiency
was determined on the basis of the total pollutant mass removed over all
storms Table 5 summarizes detention basin performance in terms of reduction
in pollutant mass loads over all monitored storm events. The analysis method-
ology used suggests that performance should be expected to improve as the
overflow rate (QR/A=mean runoff rate/basin surface area) decreases and as the
volume ratio (VB/VR=basin volume/mean runoff volume) increases. The basins
in Table 5 .are listed in increasing order of expected performance capabilities.
213
-------
TSS (mg/l)
1 ?
O V |—
£ <
[P u
,_,
-1
Rustler. State Fair.
Wisconsin Wisconsin
400
300
200
100
Total Phosphorus
(mg/l)
1 -
o "5
c <
E^
Rustler. Stt
Wisconsin Wl
-
1 — I
]te Fair,
sconsin
0.8
0.6
0.4
0.2
Street Sweeping Performance
Site Median EMC
Street Sweeping Performance
Site Median EMC
TKN (mg/l)
S .
^ >
° 5
c •<
fh
4j
Rustler,
Wisconsin
State Fair,
Wisconsin
Street Sweeping Performance
Site Median EMC
Figure 3. NURP street sweeping data.
Total Lead (mg/l)
s I
c <
Rustler,
Wisconsin
State Fair,
Wisconsin
Street Sweeping Performance
Site Median EMC
0.8
0.6
0.4
0.2
An alternative approach for characterizing the performance of detention
basins was developed by the NURP team. This approach concentrates on the
variable characteristics of individual storm events and how these are modified
by the _ detention device. A comparison of the mean and coefficient of variation
o± basin inflow and discharge concentrations provides a measure of the perfor-
mance of the device.
Great Lak I ^°^& a/ummary of detention basin performance in the
Great Lakes Basin when assessed in this manner. In most cases, more inlet
storm events were monitored than discharge events, and some inlet events do not
214
-------
TABLE 5. DETENTION BASIN PERFORMANCE
Pollutant and No. of
site storms
Lansing
Grace St. N. 18
Lansing
Grace St. S. 18
Ann Arbor
Pitt-AA 6
Ann Arbor
Traver 5
Ann Arbor
Swift Run 5
^ansing
Waverly Hills 29
NIPC
Lake Ellyn 23
Size
QR/A
8.75
2.37
1.86
0.30
0.20
0.04
0.10
ratios
VB/VR
0.05
0.17
0.52
1.16
1.02
7.57
10.70
Average mass removals-all monitored storms (percent)
TSS BOD COD TP Sol.P TKN NO 2+3 T. Cu T.Pb T. Zn
_14-____ - 9
32 3 - 12 23 7 1 26
32 21 23 18 - 14 7 nm 62 13
5 - 15 34 56 20 27 nm nm 5
85 4 2 3 29 19 80 nm 82
91 69 69 79 70 60 66 57 95 71
84 nm nm 34 nm nm nm 71 78 71
k- indicates apparent negative removals, nm indicates pollutant was not monitored.
-------
have a matching discharge event and vice-versa. For the larger basins where
storm inflow displaces only a fraction of the basin volume, it is unlikely
that influent and effluent for a specific event represent the same volume of
water. The assumption used in this analysis is that the inflow events that
were monitored provide a representative sample of the total population of all
influent EMCs. Similarly, it is assumed that the monitored effluent events
are a representative sample of all basin discharge EMCs.
TABLE 6. DETENTION BASIN PERFORMANCE*
Project and
site
Lansing
Grace St. N.
Lansing
Grace St. S.
Ann Arbor
Pitt-AA
Ann Arbor
Traver
Ann Arbor
Swift Run
No. of Percent reduction in mean EMCT
storms TSS BOD COD Tot.P Sol.P TKN N02+3 Tot.Cu Tot.Pb
23/20 (6) (26) 15 (10) (26) 11 (1) (9) 39
18/17 22 4 (3) 6 0 (5) (20) 25 14
6/6 38 17 23 28 (2) 11 8 n 59
5/5 0 (66) 12 37 63 19 28 n n
5/5 83 11 (3) (38) 21 25 77 n 86
Tot.Zn
(9)
7
22
19
n
Lansing
Waverly Hills 35/30 87 52 52 69 56 30 54 53 93 58
NIPC
Lake Ellyn 25/20 92 n 64 61 62 n 82 88 91 87
In/Out: numbers are approximate and vary with pollutant. Removals in paren-
theses indicate negative removal.
t"n" indicates pollutant either not monitored or number of observations is
too small for reliable estimate of percent reduction.
For each basin influent and effluent, the arithmetic mean and variance were
computed based on the relationships for lognormal distributions. The percent
reduction in the mean concentration and the coefficient of variation are tabu-
lated.
Performance characteristics are generally consistent using either approach,
Performance improves with detention basin size relative to catchment size. (It
is important to note that the Ann Arbor-Traver site had an unstabilized bank at
the outlet of the newly constructed basin. This accounts for the poor suspen-
ded solids removal.)
216
-------
TABLE 7. DETENTION BASIN PERFORMANCE
Project and
site
Lans ing
Grace St. N.
Lansing
Grace St. S.
Ann Arbor
Pitt-AA
Ann Arbor
Traver
Ann Arbor
Swift Run
Lansing
Waverly Hills
NIPC
Lake Ellyn
No. of Percent reduction in coeff. of var. of EMCs
storms TSS BOD COD Tot.P Sol.P TKN N02+3 Tot.Cu Tot.Pb
23/20 14 49 35 (7) (13) 30 0 0 45
18/17 (7) (59) 39 13 0 20 21 17 18
6/6 17 (6) 10 28 (84) 37 0 n 53
5/5 14 (109) 58 (3) 42 (150) (82) n n
5/5 (5) 39 50 (150) 0 20 (150) n 26
35/30 38 5 69 34 26 (8) (198) (22) 34
25/20 44 n 41 71 48 n (115) 60 19
Tot .Zn
(31)
15
(5)
0
n
(36)
41
*In/Out: numbers are approximate and vary with pollutant. Removals in paren-
theses indicate negative removal.
i~"n" indicates pollutant either not monitored or number of observations is too
small for reliable estimate of percent reduction.
The urban component of nonpoint sources received relatively little atten-
tion in the PLUARG studies. The PLUARG studies centered mainly on problem
characterization, watershed planning, technology demonstrations and policy
development. Therefore, it is not possible to compare the results of the con-
trol practice evaluation done in NURP with the IJC Nonpoint Source Task Force
evaluation.
CONCLUSIONS
The NURP team, with the assistance of the technical consultants, analyzed
and summarized the data from 22 of the 28 participating projects. Five of
these projects were located in the Great Lakes Basin.
The data from the NURP projects in the Great Lakes Basin consisted of
assessment data to help determine the existence of water quality problems
caused by urban runoff and best management practice data to determine the
effectiveness of control practices.
217
-------
Examination of the assessment data from the Great Lakes projects shows
that the loading values for the studied pollutants are relatively low. The
numbers calculated for all the pollutants were lower by at least 15% than
the nationwide values reported in the other NURP projects.
The NURP team focused on total phosphorus loads from the Great Lakes
Basin and found that urban runoff accounted for 7% of the nonpoint load to
the lakes and 3% of the total load. PLUARG reported similar loading values.
Thus, based on an analysis of the NURP data, urban nonpoint sources do
not constitute a major portion of the pollutant load to the Great Lakes Basin.
Pollutants in urban runoff, however, may contribute to water quality problems
in the local receiving waters and near shore areas of the Great Lakes.
The PLUARG report also concluded that pollutant loadings from urban run-
off generally do not constitute a significant problem on a lakewide basis.
Two control practices were evaluated in the Great Lakes Basin—street
sweeping and detention basins. Analysis of the NURP street sweeping data
showed that no significant reductions in event concentrations were realized
by street sweeping. In no case did reductions in concentration ever exceed
50%. It can be concluded that street sweeping is generally ineffective as a
technique for improving the quality of urban runoff.
Most of the detention basins monitored in the Great Lakes area were wet
basins—that is, basins that maintain a permanent pool of water. The basins
monitored were designed such that runoff from an individual storm displaces
all or part of the prior volume, and the residual is retained until the next
event. Analysis of the detention basin data in the Great Lakes area generally
showed that they were effective in removing most pollutants. It was found
that performance improved with detention basin size relative to catchment
size and hence the magnitude of the runoff processed. Reduction in overall
mass load ranged from negative to greater than 90% removal.
218
-------
TRENDS IN THE DEVELOPMENT OF WATER TOXICOLOGY IN THE USSR AMD THE US
by
L. A. Lesnikov^
ABSTRACT
Studies of the chronic effects of toxicants on fish in the Union of
Soviet Socialist Republics and the United States are reviewed and maximum
permissible concentrations and water quality criteria of the respective
countries are compared.
INTRODUCTION AND DISCUSSION
The problem of contamination has acquired urgency in connection with
the swift growth of industry and urbanization of the population (Klein 1957),
Water toxicology — the science of the influence of contaminants on water
organisms — began to develop. In the USSR, the first toxicological works
concerned estimates of the effect of petroleum contamination on fish
(Arnold 1897, Nikolskiy 1983, Khlopin 1902, Chermak 1986), and in
Great Britain — of the salts of heavy metals (Carpenter 1927). According
to "Quality Criteria for Water" (Anon. 1972), the study of the effect of
contaminants on fish was begun from the beginning of the century (Marsh
1907, Shelford 1917).
In our country, the dependence of the survival rate of test organisms
on the concentration of the substance being studied was analyzed on the
basis of brief experiments (Beklemishev and Lyubishchev 1924). In our
opinion (Lesnikov 1976), this direction is most promising. In the United
States, the basic indicator in estimating the effect is the establishment
of median lethal concentrations — 24, 48, and 96-hour LC5Q (Anon. 1979,
Anon. 1976). Along with this, proposals to establish 1X59 for a nurber of
periods and only then analyze the shape of the curve are encountered
(Alabaster and Lloyd 1980, Anderson 1948). Probably, both curves should
be identical within the limits of statistical accuracy (Figure 1).
Anderson (1948) proposes considering the value of LC^Q. Alabaster and
Lloyd (1980) consider the asymptote toward which this segment of the
curve strives as the threshold of the median concentration.
State Scientific Research Institute of Lake, River and Fishing Man-
agement, Leningrad, USSR.
219
-------
-------
et al. 1977, McKim et al. 1976); toxaphene (Mayer et al. 1975); phthalic
esters (Sanders et al. 19-); Bayer-73 (Sanders 1977); acrolein, heptachlor,
endosulfan, and trifluralin (Macek et al. 1976); chromium (Sauter et al.
1976); arochlor-1254 (Snarski and Puglisi 1976); diazinon (Allison and
Hermanutz 1977); and hydrogen sulfide (Lloyd 1976). Obviously, the list
of substances with, which chronic experiments were conducted in the United
States is far from complete.
In the USSR, up to now about 500 fishing industry MFC's have been
developed that are based on the results of chronic experiments.
We examine fishing industry MFC's in the USSR and "Quality Criteria
for Water" (Anon. 1976) which protect the hydrobionts in the United States
since these are the standards. According to our data, they are close to
those of public health and, in 60 percent of the cases, below them. There-
fore, in protecting the fish-producing properties of basins, we also pro-
tect other types of water consumption. Protection of fishing industry
interests is not a special point of little significance in the overall
strategy of protecting basins from contamination.
To what degree are the procedures for working out the MFC's in the
USSR and the water quality criteria in the United States comparable?
Some of the test organisms used in both countries are identical. They
are the fish—Salmo irrideus, Cyprinus carpio, Esox lucius, and Ferca
fluviatilis; the invertebrates— Daphnia magna, Chironomus plumosus, and
Asellus aquaricus as well as various species of Gammaridae; and algae—
Scenedesmus quadricanda.
A number of species are similar in toxicological resistance (Lesnikov
1980). Therefore, the data obtained in both countries are completely
comparable. A comparison of the MFC values and the water quality criteria
shows that, as a rule, the figures are either identical or close to one
another (Table 1).
A comparison of the values obtained shows that in the United States
greater significance is devoted to the consideration of local conditions
while in the USSR values common for the country are established. The
latter may be excessively rigid for some cases but more convenient for
monitoring purposes and more economical as regards calculations of water
protection systems. A zero MFC is adopted in the USSR if the MFC of the
substance is below 0.0001 mg/1.
In the USSR one of the mandatory indices is the estimate of the
effect of contaminants on the hydrochemical condition of basins (the pH
dynamics, dissolved oxygen, biological oxygen requirement, permanganate
oxidizability, forms of nitrogen). We did not find similar studies in
accessible literature from the United States.
In turn, a procedure has been developed in the United States to esti-
mate the effect of substances on the spawning of fish, which permits con-
sidering the effect of toxicants on the maternal species, on their progeny
from roe to spawning, and on the roe and larvae from them. Unquestionably,
221
-------
TABLE 1. WATER QUALITY CRITERIA (WQC) OF THE U.S. ENVIRONMENTAL PROTECTION
AGENCY (Anon. 1976) IN COMPARISON WITH CORRESPONDING FISH INDUSTRY
MFC's OF THE USSR (mg/1)
Contaminants
WQC
U.S.
Notes
MPC
USSR
Notes
Ammonia
0.02
nonionized, e.g.,
ammonia gas
Arsenic
Cadmium
Chromium
Copper
Cyanides
Mercury
0.05
0.0004
0.004
0.012
0.005
0.1
0.1
96 h LC50
0.005
0.00005
Nickel
0.0001
0.01
96 h LC5(J
for salmon;
in soft water,
for other species
in soft water
in hard water
in sea water
for most sensitive
species
in fresh water
in sea water
for most sensitive
species
0.05
0.5
0.005
0.005
0.01
0.001
0.001
0.005
0.05
0.0001
0.00001
0.001
0.01
ammonia gas
ammonia salt
in fresh water
in sea water
in fresh water
in sea water
mercuric
chloride in
fresh water
granosan in
fresh water
in sea water
(continued on next page)
222
-------
Table 1. (Continued)
Contaminants
WQC
U.S.
USSR
Notes
MFC
Notes
Petroleum
products and
lubricating
oils
Dissolved
oxygen
DDT
Lindane
-isomer of
hexachloran
Malathion
(karbofos)
Toxaphene
0.01
96 h LC50
5.0
and
more
0.000004
0.0001
for several most
sensitive species:
film on water sur-
face and in soils
prohibited; commer-
cial qualities of
hydrobionts should
not spoil (no exact
figures)
in water and soils
of salmon spawning
grounds
0.05
0.01
for fresh
water film on
water surface
prohibited for
sea water
6.0 and
more
6.0
4.0
0.000001
0.00001 for fresh water
for sea water
0.000005
in summer in
water
in winter in
1st category
basins
in winter in
2d category
basins
absence of
any prepara-
tive forms
absence of
all forms
and isomers
of hexachloran
absent
absent
(Polychlorocamphene)
(continued on next page)
223
-------
Table 1. (Continued)
Contaminants
WOC
U.S.
Notes
USSR
MFC
Notes
PH
Phenol
Phosphate
Phosphorous
6.5-9.0 for fresh water 6.5-8.5
6.5-8.5 for sea water
0.001 to protect com- 0.001
mercial quality
of fish
0.1 for river sections 0.04
far from lakes
0.05 for rivers near
lakes
for K2HP03
(0.31 for salt)
0.025
Phthalic
esters
Zinc
0.003
0.01
96 h LC50
for lakes and res-
ervoirs; in addi-
tion, the charge
for basins is
calculated
according to sensi-
tivity of local
species
0.05 phthalic anhy-
dride
0.01 for fresh water
0.05 for sea water
these are important and needed studies. In the USSR, the introduction of
such procedures is hindered in that our country is located further north.
The fish have relatively long life cycles (4 to 7 years or more); therefore,
such experiments are unrealistic. However, we are now selecting species
with a relatively short life cycle and sufficiently sensitive to contami-
nants. Such experiments on invertebrates and Daphnias have long been con-
ducted in the USSR. It is shown on fish that if the producers (male or
female or one of the parents) are in a toxic solution during the last month
of ovogenesis the viability of the progeny is reduced noticeably (Bykova
1978). The experiments were performed on lamprey (Misgurnus fossilis).
A common scheme for the conduct of experiments to estimate the danger
of contaminants in the U.S. is close. Therefore, cooperation of efforts
and the exchange of data are fully realistic. At present there are about
650 substances on which various toxicological data are available in the
224
-------
USSR and the US or other countries. Now tens of thousands of contaminants
are entering the basins. The conduct of toxicological studies according to
the full program requires from one to three and a half years. Therefore,
cooperative efforts of Soviet and American water toxicologists is not only
desirable, but also necessary to accelerate the process of compiling a
"toxicological catalog" of at least the basic contaminants.
REFERENCES
Alabaster J. S. and R. Lloyd. 1980. Water Quality Criteria for Freshwater
Fish. London; Boston. 297 p.
Allison D. T. and 0. Hermanutz. 1977. Toxicity of Diazinon to brook trout
and fathead minnows. U.S. Environmental Protection Agency, Washington
B.C. EPA-600/3-77-060.
Anderson B. G. 1948. The apparent threshold of toxicity to Daphia magna
for chlorides of different metals when added to Lake Erie water.
Trans. Amer. Fish. Soc. 78:96-113.
Anon. 1976. Quality Criteria for Water, U.S. Environmental Protection
Agency, Washington D.C. 256 p.
Anon. 1979. A review of the EPA Red Book: Quality Criteria for Water.
Bethesda, MD. 313 p.
Arnold, I. N. 1897. On the effect of petroleum on fish. Vestnik
Rybopromyshlennosti. 12(4):167-196. '
Beklemishev, V. N. 1923. On the course of the extinction of the ento-
mostraca population in a salt solution. Trudy pervogo Vserossiyskogo
syezda zoologov, anatomov, i gistologov [Proceedings of the First
Ail-Russian Congress of'Zoologists, Anatomists, and Histologists].
pp. 142-143.
Beklemishev, V. N. and A. A. Lyubishchev. 1924. Formulas to find constants
in toxicity equations. Izvestiya Biologicheskikh Nii I Biologicheskikh
Statyey Permskogo Universiteta. 3(4):41-52.
Benoit D.A., E.N. Leonard, G.M. Christensen and J.T. Fiandt. 197b. Effects
of cadmium on three generations of brook trout (Salvelinus fontinalis).
Trans. Amer. Fish. Soc. 105(4):550-556.
Brungs, W.A. 1972. Effects of pesticides and industrial wastes on surface
water use. In: River Ecology and Man. N.Y. p. 353-365.
Bykova, A.V. 1972. Toxic effect of mercuric chloride on embryogenesis
of fish using the example of lamprey. Express information service of
Central Scientific-Research Institute of Information and Technical-
Economic Studies of the Fishing Industry. Issue 9. pp 11-13.
225
-------
Carlson, A.R. 1972. Effects of long term exposure to carbaril (sevin) on
survival, growth and reproduction of the fathead minnow (Pimephales
promelas). J. Fish. Res. Board Canada. 28(5):583-587.
Carpenter, K.E. 1927. The lethal action of soluble metallic salts on
fishes. Brit. J. Exper. Biol. (4):278-390,
Carwell, R.D., D.G. Foreman, T.R. Payne and D.J. Wilbar. 1977. Acute and
chronic toxicity of chlordane to fish and invertebrates. U.S. Environ-
mental Protection Agency. Washington DC. EPA-600/3-77-019.
Chermak, N. 1986. On the effect of petroleum on fish. Vestnik Rybopromy-
shlennosti. 1:1-30.
Christensen G., E. Hunt and J. Fiandt. 1977. The effect of methylmercuric
chloride, cadmium chloride and lead nitrate on six biochemical factors
of brook trout (Salvelinus fontinalis). Toxicology and Applied
Pharmacology. 42:523-530.
Khlopin, G.V. 1902. Contamination of running water by farm and factory
waste products and measures for its elimination. Ucheniye zapisky
Yuryevetskogo Universiteta (Scientific Notes of Yurevets University).
No 2. p 115.
Klein, L. 1957. Aspect of River Pollution. 621 p.
Lesnikov, L.A. 1973. Methodological Instructions on Establishing Permis-
sible Concentrations of Harmful Substances in the Water of Fishing
Industry Basins. Leningrad, USSR. 22 pages.
Lesnikov, L.A. 1976. A comparison of various procedures for the conduct
of water toxicological experiments. Izvestiya Gosniorkh. 109:3-7.
Lesnikov, L.A. 1980. Comparison of principles of development and use of
water quality standards in the USSR and the USA. U.S. Environmental
Protection Agency: Washington DC. EPA-600/9-80-034.
Lloyd, L.S. 1976. Chronic effects of low levels of hydrogen sulfide on
freshwater fish. U.S. Environmental Protection Agency, Washington DC.
EPA-600/3-78-076.
Lloyd, L.S., D.M. Oseid D.M., I.R. Adelman and Broderius. 1976. Effect
of hydrogen sulfide and invertebrates. Part 1. Acute and chronic
Toxicity Studies. U.S. Environmental Protection Agency, Washington
DC. EPA-600/3-76-062.
Macek, K.J., M.A. Lindberg, S. Sauter, K.S. Buxton and P.H. Costa. 1976.
Toxicity of four pesticides to water fleas and fathered minnows.
Acute and chronic toxicity of acrolein, heptachlor, endosulfan, and
trifluralin to the water flea (Daphnia magna) and and the fathead
minnow (Pimephales promelas). U.S. Environmental Protection Agency,
Washington DC. EPA-600/3-76-899.
226
-------
Marsh, M.C. 1907. The effect of some industrial wastes on fishes. Water
supply and irrigation paper. N 192, p. 337-348.
Mayer, F.L., P.M. Mehrle and W.P. Dwyer. 1975. Toxaphene effects on repro-
reproduction, growth, and mortality of brook trout. U.S. Environmental
Protection Agency, Washington DC. EPA-600/3-75-013.
McKim, J.M., G.F. Olson and G.W. Holcombe. 1976. Long-term effects of
methylmercuric chloride on three generations of brook trout (Salvelinus
fontinalis). Toxicity, accumulation, distribution and elimination.
J. Fish. Res. Board Canada. 33(12):2726-2734.
Nikolskiy, A.M. 1893. On the significance of petroleum in the life of
•fish of the Volga River. Rybnoye Delo. 15:260-288.
Pickering G., W. Brungs and M. Cast. 1977. Effect of exposure time and
copper concentration on reproduction of the fathead minnow (Pimephales
promelas). Water Research. 11:1079-1083.
Sanders, H.O. 1977- Toxicity of the molluscocide and residue dynamics of
Bayer 2352 in aquatic invertebrates. Investigation in Fish control.
78:1-7.
Sanders, H.O. , F.L. Mayer and D.F. Walsh. 19 — - Toxicity, residue dynamics
and reproductive effects of phthalate esters on aquatic invertebrates.
Environmental Res.. 6(1):84-90.
Sauter S. , K.S. Buxton, R.J. Macek and S.R. Petrocelli. 1976. Effects of
exposure to heavy metals on selected freshwater fish. Toxicity of cop-
per, cadmium, chromium and lead to eggs and fry of seven fish species.
U.S. Environmental Protection Agency, Washington DC. (EPA-600/3-76-105),
Shelford, V.E. 1917. An experimental study of the effects of gas wastes
upon fishes with especial reference to stream pollution. Bull.
Illinois State Lab. for Nat. History. 11:381-412.
Snarski, V.M. and F.A. Puglisi. 1976. Effects of arochlor or 1284 on brook
trout (Salvelinus fontinalis). U.S. Environmental Protection Agency,
Washington DC.EPA-600/3-76-112.
Stroganov, N.S. 1941. New ways to solve the problem of the effect of indus-
trial runoff water on water organisms. Scientific notes of Moscow
State University. Biologiya. Issue 60. pp 5-24.
Stroganov, N.S. 1971. A procedure for determining toxicity of the water
Medium. In: Metodiki issledovaniy po vodnoy toksikologii (Procedures
for Studies on Water Toxicology). Moscow, USSR, pp 14-60.
Yeltsina, N.V. 1939. The effect of sea salt on the development of daphnia
and adaptation to conditions of increased salinity. Problems in
Ecology and Biocenology. Issue 4. pp 128-142.
227
-.US GOVERNMENT PRINTING OFFICE 19860646-116-40649
------- |