-------
The depuration tests for TeCB dosed small rainbow trout indicated that different
metabolic rates resulted in different rates of TeCB depuration which was reflected in the slope of
the depuration curve (Figure 5). The constant k2 was calculated using BIOFAC® computer
program for each of the depuration tests and was plotted against the mean value offish oxygen
uptake rate in a specific depuration period (Figure 6). A drastic increase in k2 was observed
when fish metabolic rate was increased, which tends to back up the statement that fish gills are
the major site of toxicant clearance. It was also found that the depuration data profile was similar
for fish exposed only to TeCB and those exposed to three chemicals simultaneously, indicating
there was no interaction among these chemicals during depuration processes. As depuration tests
are relatively time consuming, this finding will facilitate faster data acquisition in future
depuration studies.
Q
O
ra
oo
90
75'
60
45
30
15
0
• 2.0 BL/S,
X
'*v^
• — ,
11,6 °C
-•
80
60
40
20
0
10 15 20 25
10 15 20 25
80
60
40
20
0
\
2.9 BL/S, 14.3 °C
10 15 20 25
O
O
O
O
BO
60
40
20
0
-20
2.0 BL/S, 6.5 C
70
60
50
40
30
20
10
-10
_
- • 2.9 BL/S, 9.6 °C
- T
- 1
• t
t
VM. .. .
. *^ — •• • — • • •
-50 5 10 15 20 25
-50 5 10 15 20 25
DEPURATION (day)
Figure 5. The 21-day depuration test of small rainbow trout swimming at low and high speed
under different temperatures after 24-hour double-dose exposure to 260|ig/L TeCB.
-------
Oxygen Consumption^ Rate
, (mg/kg/h)
Figure 6. Relationship between oxygen consumption rate (lylO2) and depuration rate constant (k2)
in small rainbow trout.
A very similar relationship between k2 and fish lipid content was found to exist if
compared with that between k, and lipid level (Figure 7). Geyer et al, (1995) demonstrated that
the depuration half lives (tj/2), which is inversely correlated with k2, of TCDD in mussel and fish
at different developmental stages increases with their lipid content (Figure 7). The lipid content
of these small rainbow trout were all close to 5%, although k2 was also calculated with lipid
normalized data in these five depuration experiments, no significant difference was seen from the
data expressed on wet body basis. '•
00
o
C
ffl
_*j
M
C
o
CJ
q
o
a_J
<0
l-l
a,
- 1.0 -
-l.O'r-
-1,5 r-
-2.0
-3,0 I
0.8 0.4 0.6 0,8 1.0 1.2 1,4
Lipid Content (%, log)
-101 2 3 4
Body Weight (g, log)
Figure 7. Correlation between l^^S-tetrachlorodibenzo-^-dioxin (TCDD) depuration rate
constant k2 and lipid content/body weight in different fish species at various developmental
stages. Adapted from Geyer et al. 1995.
7
-------
MODEL DEVELOPMENT AND PRELIMINARY TEST
The overall objective of this research is to develop a simple model in which some easily
measured fish physiological parameters and certain chemical properties are incorporated to
predict the amount of xenobiotics in a fish body. The presumption of this study is that the fish
gill dominates as the xenobiotic exchange site between the fish body and the environment during
both uptake and depuration periods. It has been shown that both uptake and depuration kinetics
can be affected by fish oxygen uptake rate. The most commonly used one-compartment, first
order kinetic (1CFOK) model (Spacie and Hamelink 1982) was chosen as the basis of this work.
As this is a rate-constant (kj and k2) based model, it becomes a matter of predicting the uptake
and depuration rate constants according to fish oxygen consumption rate by using the
correlations found in this study (Figures 4 and 6). A test was conducted to look at the feasibility
of predicting chemical concentration hi fish by incorporating into the 1CFOK model, the
relationships established between fish MO2 and k, and k2.
Bishop and Maki (1980) did some bioconcentration studies on juvenile bluegill (Lepomis
macrochirns) with dichlorodiphenyltrichloroethane (DDT), sodium dodecylbenzenesulfonate
(LAS) and tetradecylheptaethoxylate (AE> among other chemicals. Based on the kinetic data and
all the information necessary for the physiological model, predictions were made about DDT, AE
uptake and LAS depuration in bluegills, and the results of the model test on these data together
with those of UBC Tests 243 and 244 depuration experiments are shown in Figures 8 and 9,
respectively. The predicted uptake data of DDT at both test concentrations agreed with the
measured ones quite well up to exposure time 120 hours. The experimental data of AE,
however, seemed to level off after 24 hours while the predicted values kept increasing. As for
the depuration data, model values of LAS concentrations fitted extremely well with the test ones
at two water concentrations, as did the data for UBC Test 244 (high speed depuration) but the
calculated data for UBC Test 243 (low speed depuration) didn't fit nearly as well. The
discrepancy between the predicted and measured values could come from several sources. For
example, the relationships between MO2 and kinetic rate constants may vary with fish body
weight, as was indicated hi the study by Brauner et al. (1992) where they found that the
correlation was slightly different between fish below and above 1 Ig. The bluegill used by
Bishop and Maki (1980) were less than Ig, so the relationship derived in our experiments might
need further adjustment before being applicable for fish in this weight range. Also, the
relationship between k2 and fish MO2 needs to be further strengthened with more depuration data
and the effect offish size on this correlation remains to be investigated.
The main advantage of the modified 1CFOK model over the other compartmental models
is that the prediction is based on actual physiological processes and fish oxygen consumption
rate is far easier and accurate to measure than some of the parameters required by other
physiological models (Hayton and Barren 1990), such as gill surface area and blood flow,
epithelial thickness and diffusion coefficients. Moreover, oxygen consumption rate is available
through the oxygen data bank (OXYREF) (Thurston and Gehrke 1993) provided the fish body
weight is known. This modified model, therefore, possesses some functional reality which might
enable more realistic predictions, and is relatively convenient to be applied in practical
biomonitoring and toxicant risk assessment.
8
-------
MODEL TEST: AE UPTAKE
Bishop and Maki, 1980
Predict (0.151ppm)
Measure {0.151ppm)
Predict (0.014ppm)
leasure (0.014ppm)
MODEL TEST: DDT UPTAKE
Bishop and Maki, 1980
1 3 6 24 48 120
TIME (hour)
Predict (0.23ppb)
Measure (0.23ppb)
•edict (0.026ppb)
Measure (0.026ppb)
Figure 8. Comparison of predicted and measured uptake data for dichlorodiphenyltrichloroethane
(DDT) and tetradecylheptaethoxylate (AE) (measured data from Bishop and Maki 1980).
MODEL TEST: TCB DEPURATION
UBC 243, 244 Data (1993)
Measure (low speed)
Predict (low speed)
0.5
1 248
TIME (day)
16 21
MODEL TEST: LAS DEPURATION
Bishop and Maki, 1980
Measure (0.64ppm)
Predict (0.64ppm)
Measure(0.063ppm)
Predict(0.063ppm)
24 48
TIME (hour)
Figure 9. Comparison of predicted and measured depuration data for tetrachlorobenzene (TeCB)
and sodium dodecylbenzenesulfonate (LAS) (measured data for LAS from Bishop and Maki
1980). :
9
-------
Some authors have reported that the uptake and clearance rate constants of some
lipophilic compounds by various fish species have a fixed relationship to the Kow (Hawker and
Connell 1985, Saarikoski et al. 1986). Studies of the relationship between k, and Kow in fish
have shown that (1) kj increases with K^, if logKow is low (<-4), (2) kj is constant if logKow is
large (between 4 to 6), and (3) kt drops with increasing Kow for chemicals with extremely high
K™ (logK^ above 6) (McKim et al. 1985, Gobas et al. 1986, Gobas and Mackay 1987). Hawker
and Connell (1985) reported a very good linear correlation between k2 and logKow, within the
range from 2.5 to 6.0, as depuration rate constant decreases with increased Kow. The Kow of TCB
and TCG are 4.99 and 4.41, respectively. Thus, the relationship between these uptake/depuration
rate constants and fish MO2, derived from this experiment, can only cover a certain range of
hydrophobic chemicals. As discussed above, it is possible to derive the correlation for chemicals
with a lower K^ based on the equations concerning kl5 k2 and Kow in the literature. However, the
relationship predicted according to the data with TeCB and TeCG needs to be validated by actual
tests using a chemical of lower Kow.
CONCLUSIONS
(1) Toxicant uptake rate constant (kj and depuration rate constant (k2) increase with fish
oxygen consumption rate (MCy and decrease with fish lipid content; (2) A modified one-
compartment first-order kinetic (1CFOK) model incorporating the relationships between kj/kj,
MO2 and lipid content offish can be used to predict body toxicant burden with reasonable
accuracy.
ACKNOWLEDGEMENTS
This work was supported by Cooperative Agreement CR816369 from the U.S.
Environmental Protection Agency, the British Columbia Science Council, and the National
Science and Engineering Research Council of Canada.
REFERENCES
Barron, M. G., G. R. Stehly, and W. L. Hayton. 1990. Pharmacokinetic modeling in aquatic
animals. I. Models and concepts. Aquatic Toxicology 18:61-86.
Bishop, W. E., and A. W. Maki. 1980. A critical comparison of two bioconcentration test
methods. Aquatic Toxicology, ASTM STP 707. J.G. Eaton et al. (Eds.), pp. 61-77.
Black, M. C., and J. F. McCarthy., 1988. Dissolved organic macromolecules reduce the uptake
of hydrophobic organic contaminants by the gills of rainbow trout (Salmo gairdneri).
Environmental Toxicology and Chemistry 7:593-600.
Black, M. C., D. S. Millsap, and J. F. McCarthy. 1991. Effects of acute temperature change on
respiration and toxicant uptake by rainbow trout, Salmo gairdneri (Richardson).
Physiological Zoology (in review).
10
-------
Brauner, C. J. 1991. The relationship between oxygen and 1,2,4,5-tetrachlorobenzene uptake in
five species of fishes. M.Sc. Thesis. The University of British Columbia.
Brauner, C. J., J. F. Newman, R. V. Thurston, and D. J. Randall. 1992. The relationship
between the uptake of organic chemicals and oxygen consumption in fish. In: Proceedings of
the 3rd International Symposium on Fish Physiology, Fish Toxicology, and Water Quality
Management. Nanjing, People's Republic of China. November 3-5, 1992. Office of
Research and Development, U.S. Environmental Protection Agency. EPA/600/R-94/138.
207pp. :
Brauner, C. J., D. J. Randall, J. F. Newman, and R. V. Thurston. 1994. The effects of exposure
to 1,2,4,5-tetrachlorobenzene and the relationship between toxicant and oxygen uptake in
rainbow trout (Oncorhynchus mykiss). Environmental Toxicology and Chemistry 13:1813-
1820.
Connell, D.W. 1990. Bioaccumulation of Xenobiotic Compounds. CRC Press, Inc. Boca
Raton, Florida. 213pp.
Dobbs, A. J., and N. Williams. 1983. Fat solubility - a property of environmental relevance?
Chemosphere 12:97-104.
Gehrke, P. C., L. E. Fidler, D. C. Mense, and D. J. Randall. 1990. A respirometer with
controlled water quality and computerized data acquisition for experiments with swimming
fish. Fish Physiology and Biochemistry 8:61-67.
Geyer, H., I. Scheunert, R. Bruggemann, V. Zitko, C. E. W! Steinberg, R. Pruell, A. Kettrup,
J. Olson, P. Schmieder, and D. C. G. Muir. 1995. Relationship between elimination half-
lives (T1/2) of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in different mussel and fish
species and their lipid content or body size. Organohalogen Compounds 25:273-278.
Gobas, F. A. P. C., and D. Mackay. 1987. Dynamics of hydrophobic organic chemical
bioconcentration in fish. Environmental Toxicology and Chemistry 6:495-504.
Gobas, F. A. P. C., A. Opperhuizen, and O. Hutzinger. 1986. Bioconcentration of hydrophobic
chemicals in fish: Relationship with membrane permeation. Environmental Toxicology and
Chemistry 5:637-646.
Hawker, D. W., and D. W. Connell. 1985. Relationship between partition coefficient, uptake
rate constant, clearance rate constant and time to equilibrium for bioaccumulation.
Chemosphere 14:1205-1219.
i
Hayton, W. L., and M. G. Barren. 1990. Rate-limiting barriers to xenobiotic uptake by the gill.
Environmental Toxicology and Chemistry 9:151-157. ,
Heath, A.G. 1987. Water pollution and fish physiology. CRC Press Inc., Boca Raton.
11 ;
-------
Hamelink, J.L., and A. Spacie. 1971. A proposal: Exchange equilibria control the degree
chlorinated hydrocarbons are biologically magnified in lentic environments. Transactions of
the American Fisheries Society 100:207-218.
McKim, J., P. Schmieder, and G. Veith. 1985. Absorption dynamics of organic chemical
transport across trout gills as related to octanol-water partition coefficient. Toxicology and
Applied Pharmacology 77:1-10.
Murphy, P. G., and J. V. Murphy. 1971. Correlation between respiration and direct uptake of
DDT in the mosquito fish Gambusia affinis. Bulletin of Environmental Contamination and
Toxicology 6:581-588.
Randall, D. J., and C. Daxboeck. 1984. Oxygen and carbon dioxide transfer across fish gills. In:
Fish Physiology Volume 10A. W. S. Hoar and D. J. Randall (Eds). Academic Press Inc.,
New York. pp. 263-307.
Randall, D. J., and C. J. Brauner. 1990. Toxicant uptake across fish gills. In: Proceedings of the
Second International Symposium on Fish Physiology, Fish Toxicology, and Water Quality
Management. Sacramento, California, September 18-20, 1990. United States Environmental
Protection Agency, Environmental Research Laboratory, Athens, Georgia, USA.
EPA/600/R-93/157. 245pp.
Randall, D. J., C. J. Brauner, and R. Yang. 1993. Uptake of chemicals by fish. In: The Aquatic
Resource Research Project Technical Report. A.P. Farrell (Ed.). Simon Fraser University.
pp. 181-194.
Saarikoski, J., R. Lindstrom, M. Tyynela, and M. Viluksela. 1986. Factors affecting the
absorption of phenolics and carboxylic acids in the guppy (Poecilia reticulatd).
Ecotoxicology and Environment Safety 11:158-173.
Spacie, A., and J. L. Hamelink. 1982. Alternative models for describing the bioconcentration of
organics in fish. Environmental Toxicology and Chemistry 1:309-320.
Thomas, R. E., and S. D. Rice. 1981. Metabolism and clearance of phenolic and mono-, di-, and
polynuclear aromatic hydrocarbons by Dolly Varden char. In: Physiological Mechanisms of
Marine Pollutant Toxicity. W.B. Vernberg et al. (Eds.). Academic Press, New York. pp.
425-448.
Thurston, R. V., and P. C. Gehrke. 1993. Respiratory oxygen requirements of fishes:
description of OXYREF, a datafile based on test results reported in the published literature.
In: Proceedings of the Second International Symposium on Fish Physiology, Fish
Toxicology, and Water Quality Management. Sacramento, California, September 18-20,
1990. United States Environmental Protection Agency, Environmental Research Laboratory,
Athens, Georgia, USA. EPA/600/R-93/157. 245pp.
12
-------
FUNCTIONS OF NEUROPEPTIDES AND NEUROTRANSMITTERS
IN THE REPRODUCTIVE PROCESS OF AMPHIOXUS
Zhang Chong-Li1, Shen Wei-Bin1, Yin Hong1, and Jiao Li-Hong1
ABSTRACT
For a long period of time, we knew very little about the reproductive neuroendocrinology
of amphioxus which is a unique and important living species of the chordates. The animals used
in this experiment were Branchiostoma belcheri Gray and B. belcheri var. Tsingtaonesis,
collected from Xiamen in the southern part of China and Qingtao on the northeast coast of China,
respectively. Using immunohistochemistry, radioimmunoassay, high performance liquid
chromatography and high performance liquid chromatography with an electro-capture detector,
we measured the neuropeptides and neurotransmitters in both sexes of mature and young
amphioxus, and found the relationship between their contents and gonadal development as well
as spawning/spermiation. The results showed that amphioxus contained not only a number of
regulatory peptides (gonadotropin releasing hormone and opioid peptides), but also
neurotransmitters, e.g., norepinephrine and dopamine. These endogenous bioactive substances
existing in the bodies and gonads may be responsible for their synthesis and release as well as
participation in the regulation of the development of gonads and spawning/spermiation.
This finding is of significance in understanding the evolutionary process of reproductive
neuroendocrinology of vertebrates.
INTRODUCTION
The small fish-like lancelet, represented by amphioxus, is phylogenetically a unique
and important living species of chordate, and has long been considered as being in the line of
evolution between vertebrate and invertebrate hypothetical phyla. Many primative features
characterizing early vertebrates are found hi amphioxus. Chang et al (1985) have reported that
the luteinizing hormone-releasing hormone (LHRH), the follicle stimulating hormone (FSH),
and luteinizing hormone (LH) immuno-positive reactions were visualized in the epithelia tissue
of Hatschek's pit, and further demonstrated mammalian gonadotropin and sex steroid hormones
in amphioxus. Recently, we have identified two forms of LHRH, opioid peptides and
neurotransmitters, in the amphioxus (Shen et al. 1991, Zhang et al. 1991, Fang and Zhang 1991,
Zhang et al. 1993). The study of reproductive neuroendocrinology of amphioxus is of interest in
the reproductive process. It may provide new insights into the evolutionary aspects of
reproductive patterns, reproductive hormones, and neurotransmitters. Due to the lack of fossils
for evolution of the phylogenetical development of reproduction, the amphioxus (a living fossil)
has become a desirable animal model for investigation. It is; the species which may represent the
form that has been suggested to be the likely kinsman of the ancestral type of vertebrates.
'State Key Lab of Reproductive Biology, Institute of Zoology, Academia Sinica, Beijing, PRC.
13 '• .
-------
The data obtained from this study would be expected to provide evidence filling
a neuroendocrine gap in a continuous evolutionary process from the lower to the higher forms
of living organisms. A more extensive physiological and biochemical consideration of the
lower chordates would probably contribute much toward the solution of this problems.
The developmental physiology and morphological structure of amphioxus have been well
documented. However, for long periods we knew little about their reproductive
neuroendocrinology. Using immunohistochemistry, radioimmunoassay (RIA), high performance
liquid chromatography (HPLC), and HPLC with an electro-capture detector (HPLC/EC), we
have measured the neuropeptides and neurotransmitters in both sexes of mature and young
amphioxus, especially to elucidate the relationship between their contents and gonad
development as well as spawning/spermiation.
MATERIALS AND METHODS
Animals
Branchiostoma belcheri Gray was collected from Xiamen (Amog) in the southern part of
China and B. belcheri var. Tsingtaonesis, a variety of small body size, was obtained in Qingdao
(Tsingtao) on the north-east coast of China. Most of the experiments were carried out with the
Qingdao variation.
Immunohistochemistry of p-endorphin
Both sexes of amphioxus were washed with fresh sea water, then fixed with Bouin's
solution. The immunoreactivity of p-endorphin (p-END) on the routine paraffin section was
tested with antiserum against P-END with an immunogold-silver staining method (Wang 1985).
RIA of Gonadotropin Releasing Hormone (GnRH) and P-END
Mature individuals of both sexes of amphioxus were usually collected in the evening in
late June to early July. When the animals swam to the surface of sea water in the laboratory tank
to spawn/spermiate they were collected immediately or at intervals of 2-12, 24,48 and 72 hours
following spawning/spermiation. The samples were cut into several pieces, boiled in 2 mol/L
acetic acid, homogenized with an ultrasonic homogenizer, and centrifuged. The clear
supernatant was lyophilized for measuring the contents of GnRH and P-END. The young
amphioxus were collected every month and GnRH levels were detected as methods for mature
samples. The method of RIA for GnRH was carried out according to the procedure described by
Zhang at al. (1989), and the method for P-END was carried out according to the modified
method of Zhu et al. (1986).
14
-------
Extraction and Purification of Peptides
Both sexes of mature amphioxus were cut into three;parts: "head", "middle" (including the
gonad), and "tail" regions; these, as well as whole body, were analysed. Each was sliced into
small pieces, put into 2 mol/L acetic acid (w:v=l :10), and the samples were then homogenized
With a high-speed homogenizer. After centrifugation, the supernatant was passed through the
preparative column of HPLC. The elute fractions were detected by RIA of GnRH and 3-END.
The immunoreactive fractions were pooled and lyophilized, then further analysed by reverse
phase HPLC, and eluted with linear gradient (Wu et al. 1986). Aliquots of fractions collected
each minute were measured for RIA of GnRH and (3-END.
Detection of neurotransmitters by HPLC/EC ;
The "head" regions of both sexes of amphioxus were added to 0.2 mol/L perchloric acid
(PCA), sliced into small pieces, homogenized with an ultra-sonic homogenizer, and centrifuged
(13,000 rpm, 30 minutes). The clear supernatant was injected into a CIS column of HPLC and
detected with an electro-chemical (EC) detector (LC-150, Bioanalytical System Co.). The
different kinds of neurotransmitters were identified according to the method of Mefford (1981).
RESULTS
GnRH in the Reproductive Process of Amphioxus
Annual Changes of GnRH in Amphioxus. The GnRH level in female amphioxus
increased significantly with the beginning of the reproductive season and reached its maximum
in June (159± 3.2 mg/10 animals), then declined obviously after breeding (39.1 + 1.6), and
reached the lowest value in February (30.8±0.61) (Figure 1). So the GnRH content in female
amphioxus during the reproductive season was approximately two to five times greater than that
of the non-breeding season. The annual pattern of GnRH content in male amphioxus (Figure 2)
was different from that of the female. As compared with the females, the GnRH level in the
males increased earlier than that of females, and remained at a relatively steady level through the
whole breeding season; its content was just about two to three times greater than that of the non-
breeding season. ;
Relationship Between GnRH and Gonadal Development. The relationship between
gonadal development and GnRH was examined in the head; middle, and tail regions as well as
the whole body in a different developmental stage. It was found that the GnRH existed in every
region of the adult amphioxus, and a little in the larval stage. The GnRH level in the head
region, in which the Hatschek's pit localizes, increased in parallel with gonadal development and
maturity. During the mature stage, the GnRH contents of the head region were 0.0325 ng/10
heads of females (Figure 3) and 0.0248 ng/10 heads of males (Figure 4), which was
approximately four to five times greater than those detected in Stage 0 (0.006ng/10 heads) of the
immature amphioxus. The gonadal development of the amphioxus could also be correlated to its
total GnRH contents; total content of GnRH increased with gonadal development, and reached
its maximum at the time of gonadal maturity (Figures 5 and 6).
15 !
-------
'.SO
160
2 100
£ so
I
ff
O 60
40
20
I 2 3 4 5 <> 7 8 9 . 10 II 12-
Month
Figure 1. Annual change of GnRHin the female amphioxus.
60
V)
•a
£
2 40
X
20
1 2 3 4 5 6 7 & 9 10 11 12
Month
Figure 2. Annual change of GnRH in the male amphioxus.
16
-------
0.035
0.030
.«
•s
.8 0.025
1 0.020
o
5 °-015
•£-
i| 0.010
0.005
'0 i fi .in IV; v
Stage of ovaiy development
Figure 3. Relationship between GnRH of head region and gonadal development
in the female amphioxus.
"3
0.030
0.025
•a
1 0.020
CO
O
§ 0.015
£,
= 0.010
C3
0.005
0 I II ffi F V
Stage of testicaUr development
Figure 4. Relationship between GnRH of head region and gonadal development
in the male amphioxus.
17
-------
i
30
2. 'MI-
'S
DC
10
5
o i n m iv v
Stage of ovuy development'
Figure 5. Relationship between total GnRH and gonadal development
in the female amphioxus.
35
30
I"
o 20
*«
& IS
10
5
i
o i n m iv v.
Stage of letticulttr development
Figure 6. Relationship between total GnRH and gonadal development
in the male amphioxus.
18
-------
Profile of GnRH Level in Amphioxus Before and After Spawning/ Spermiation. Using
the RIA of GnRH, we measured the GnRH level in the different times of pre- and post-
spawning/spermiation. The results showed that the level of GnRH in females declined at
spawning, came to the lowest level at 12 hours of post-spawning (PO.05), and then gradually
increased to normal level after 48 and 72 hours (Figure 7). The changes in the profile of GnRH
in the male were not statistically significant.
-a- male
-o- fwnal*
•: P<0.06
o
DC
O
SO
40
30
20
10
Pre O 2 12 24 48 72
(Post)
Time (hours)
Figure 7. Changes of GnRH level in amphioxus before and after spawning (female) or
spermiation (male). Pre: before spawning or spermiation Post: after spawning or
spermiation.• *: Pre vs post 12 hours, PO.05. ;
3-END in the Reproductive Process of Amphioxus
Distribution offt-END in the Gonad. It was shown that 3-END was mainly localized in
the gonad by means of immunohistochemistry. In the testis? the P-END positive immunoreactive
substances were observed in the whole testis region (Figure 8-1). In the ovary, the 3-END
positive immunoreactive substances were observed in the oocyte plasma and around the nucleus
(Figure 8-2). Besides the gonads, the 3-END positive immunoreactive substances were also
visualized in the neurons and neurofibril of the neural tube (Figures 8-3 and 8-4). The control
tests, using normal rabbit serum as first antibody are shown for testes in Figure 8-5, and for
ovary in Figure 8-6.
Profile of$-END in Amphioxus Before and After Spawning/Spermiation. Using the
RIA of 3-END, we measured the 3-END level in the different times of pre- and post-
spawning/spermiation. The results showed that the level of ;3-END declined markedly at a
period of 2-12 hours after spawning/spermiation (PO.05), and restored gradually to normal level
by 72 hours (Figure 9). Similar profile changes occurred mboth female and male amphioxus.
19
-------
Figure 8. Distribution of p-endorphin in the gonad of amphioxus. 1. The P-END positive
immunoreactive substance localized in the testis of the male amphioxus. 2. The P-END
positive immunoreactive substance localized in the oocyte plasma in the ovary of the
female amphioxus. 3. The P-END positive immunoreactive substance distributed in the
neural tube (arrow shows the neurofibril protruded from the neural tube). 4. The P-END
positive immunoreactive neuron and its dendron. 5. The control test of testes by using
normal rabbit serum as first antibody. 6. The control test of ovary by using normal rabbit
serum as first antibody.
20
-------
140
120
a 100
f
Ul
^ 80
60
i J
(17)
(7)
Pie
24
48
72
0 Post
2—12
Time(h)
Figure 9. Changes of (3-endorphin level in amphioxus before and after spawning (female)
or spermiation (male). Pre: before spawning or spermiation Post: after spawning or
spermiation *: Pre vs post 2-12 hours, P<0.05.
Identification of GnRH and Opioid Peptides in Amphioxus.
The acid extract of amphioxus was isolated and purified with a preparative HPLC and
the activities of (3-END and GnRH were checked by RIA (Figure 10). The immunoreactive part
was pooled and lyophilized, and then separated by HPLC, with a Cl 8 column. The result
showed that GnRH (mGnRH and sGnRH) and opioid peptides (3-END), Dynorphin A (Dyn A),
and Leu-enkephalin (Leu-ENK) are present in amphioxus (Figure 11) according to their retention
time compared with the chromatographic behavior of the standards and RIA (Figurel2).
The two forms of GnRH also distributed in the different regions of the body, such as head,
middle, tail and gonads. The content of GnRH is higher than that of sGnRH in ovaries and testes
(Figures ISA and B). In both sexes of amphioxus, the content of mGnRH is higher than that of
sGnRH in the female. Contrarily the sGnRH in the male is higher than that of mGnRH
(Figures 13C and D). ;
Neurotransmitters in the Reproductive Process of Amphioxus
Kinds ofNeurotransmitter in Amphioxus. The extract of the head region of the
amphioxus was injected into a CIS column detected by HPLC/EC. Compared with the
chromatogram of standard neurotransmitters, the norepinephrine (NE) and dopamine (DA) were
detected in the head region; the metabolites of NE, 4-hydroxy-3-methoxy-phenylglycol (MHPG)
and DA, 3,4-dihydroxyphenyl-acetic acid (DOPAC) was present in the partial samples
(Figure 14).
21
-------
r
1
8I 1
1
8
S
is u it u u is u a
Tub* Ho.
Figure 10. Chromatogram of amphioxus' extract by the preparative-reverse HPLC. The extract
was injected into a CIS column. Aliquots of the collected fractions were assayed with
RIA of p-END. Solvent system: linear gradient from (A) to (B), 40 minutes. (A):
0.1%TFA.(B): 80%MeOH/0.1%TFA. Flow rate: 5ml/minute.
Figure 11. Chromatogram of amphioxus' extract by HPLC. The chromatographic parameters are
as described in Figure 12. After treatment with preparative HPLC, the concentrated
extract was injected into CIS column. Aliquots of the collected fractions were assayed
with RIA of P-END and GnRH. The elution peaks of opioid peptides (p-END, Dyn A,
Leu-ENK) and GnRH (mRnRH and sGnRH) are indicated by the arrows.
22
-------
Tin (-In)
Figure 12. Chromatogram of reverse-phase HPLC of standard opioid peptides (P-END, Dyn A,
Leu-ENK) and GnRH (mGnRH and sGnRH) for the column calibration. Column size:
ALTEX Ultrasphere ODS column, 0.46x25 cm. Solvent system: linear gradient from (A)
to (B), 70 minutes. (A): 50 mmol/L KH2PO4-H3PO4 (pH 2.0):CH3CN=90:10 (v/v). (B):
50 mmol/L KH2PO4-H3PO4 (pH 2.0):CH3CN=50:50 (v/v). Flow rate: O.6 ml/min.
(Ill
50
40
^ 30
3= 20
at
I 10
i£ ^
O 40
2= 30
3.)
10
n
— S.
tAl jjj
c:
£
•
J1-
:<0 Ij
X'
*
hn.
n
. mL -
/Hi =
(**» ! 3g
o
=
. :. .J
— 1
X.
j=l
-
•
.
Lrv.
111} :
.
J •
. _J.II
"
Si
40
30 •;•
31 if
u.
in §
0 "*"~
40 5
30 SS
20
10
0
Retention Time(min)
Figure 13. The contents of mGnRH and sGnRH in the gonads or whole body of both sexes of
amphioxus measured by HPLC with RIA. A: ovaries; B: testes; C: female bodies;
D: male bodies.
23
-------
NE
Figure 14. Chromatogram of neurotransmitters in amphioxus' head, region.
Profile ofNE and DA in Amphioxus Before and After Spawning/Spermiation.
The NE and DA in the head region at the different times of pre- and post-spawning/spermiation
was detected by means of HPLC/EC. The results showed that in the female, the contents of NE
and DA increased markedly at beginning of spawning/sperniiation (PO.05), and then decreased
gradually until post-24 hours, and finally at 48 hours increased again even over the pre-spawning
levels. In the male, the contents of NE and DA increased slightly, but not statistically significant
during the spawning/spermiation period. Then the levels decreased to their lowest level at 2
hours after spermiation, and then increased gradually until they were restored to their normal
levels at 48-hours post-spermiation (Figures 15 and 16).
24
-------
i
a.
a
i
ui
150
100
i
(9)
(5)
-*- Ma!e
-a- FamsJ*
PRE O 2h 12h 24h 48h
Thn« (hours) ,
Figure 15. Changes of NE levels in the "head" of amphioxus in different times
before and after spawning/spermiation.
200
S
100
*:P<0.05: **:P<0.01
PRE O 2h 12h 24h 48h
Tim* (hours)
Figure 16. Changes of DA levels in the "head" of amphioxus in different times
before and after spawning/spermiation.
25
-------
DISCUSSION
In the recent years, studies on reproductive endocrine in lower vertebrates have
demonstrated that in fishes such as Caribe Colorado and Pygocentrus notatus, the GnRH
(LH-RH) contents of the hypothalamus and telecephalon of the female in contrast to those of the
male, consisted with annual gonadosomatic index (GSI), and exhibited some changing pattern
corresponded to an annual cycle (Gentile et al. 1986). Based on immunocytochemical studies
using the anti-GnRH antibody, Crim et al. (1979) investigated the immunoreactive cells in the
preoptic nucleus and neurohypophysis of larval and adult pacific lampreys (Lampetra tridentatd).
They reported that the GnRH specific immunoreaction was very weak hi larvae, but stronger in
adults during the non-breeding season and became the strongest in adults during the breeding
season. Therefore, a close relationship existed between the GnRH in lower vertebrates and the
reproductive cycle. The amphioxus is a transitional form of organism in the evolution cycle that
links invertebrates to vertebrates. The existence of GnRH in amphioxus has been reported by
Chang et al. (1985), with the GnRH immunoreactive granules localized in the basal region of
epithelial cells in Hatschek's pit. Zhang et al. (1993) have further demonstrated there are two
forms of GnRH (mammalian and salmon) in the amphioxus. The present data not only verified
the previous findings, but also reveal that the GnRH of both sexes of amphioxus, especially in its
head region, appeared an obvious annual change parallel to the gonadal development and GSI,
which is comparable to what has happened in lower vertebrates. This observation is the same
with the total content of GnRH. The data indicated that GnRH in the head region may play a
pivotal role in reproduction, because Hatschek's pit is localized in the head. Interestingly, this
changed pattern of GnRH hi the head region is also in consistency with the results of other
reports on teleost fishes and Agnatha. However, it should be noted that both fishes and Agnatha
have a hypothalamus, which includes a higher center that regulates reproductive activity; while
in the amphioxus, GnRH neural cells are scatteringly distributed and no neural nuclear mass is
formed. This is a clue suggesting that the scattered GnRH secretory cells are gradually
concentrated into a nuclear mass. Therefore, the primitive status of the amphioxus may be
regards as new evidence for the evolution of reproductive neuroendocrine function.
In addition, the middle and tail regions of amphioxus, as well as gonads, also contain
GnRH. This reflects the extensive distribution of GnRH hi the primitive character. It has been
well known that the neural tube of amphioxus extends almost completely from the head to tail,
and there are many neufosecretory cells surrounding the tube. It seems that some of them might
produce GnRH. It is ineresting to note that only in the head region is GnRH responsible for
gonadal development and reproduction, while GnRH in the middle and tail regions is not.
A possible implication might be that the effective GnRH-producing cells in the amphioxus are
concentrated towards the head region, and act as assertion of the hypothalamic primordium
regulating reproductive activity, which might represent another aspect of evolution.
Up until now, six forms of GnRH distinct from mammalian GnRH. have been identified
in the last 10 years. All the GnRH structures hi vertebrates identified to date are decapeptides
with at least 50% of the sequence identified. The seven forms of GnRH, with their distinct
primary structures, are named for the animal from which they were first isolated by peptide
26
-------
chemistry. That is mammalian, chicken I and II, salmon, catfish, dogfish, and lamprey-GnRH
(Sherwood et al. 1993). Additionally, the presence of multiple forms of GnRH in one species,
some of which may be common to different classes of vertebrates, may help to establish the
evolutionary thread for GnRH in vertebrates. We found that two forms, mGnRH and sGnRH,
exist in the amphioxus and they are distributed in the head, middle, and tail as well as gonad. In
both sexes of arnphioxus, the content of mGnRH is higher than that of sGnRH in the ovaries and
testes (Figures 13 A and B). In the bodies, the content of mGnRH is higher than that of sGnRH
in the female, but not so in the male body (Figures 13C and D). These results show that the
changes of GnRH content during breeding season are similar to the changes during the
mammalian reproductive process in both sexes and further indicate that mGnRH may play a
more important role in the development of gonads and the reproductive process, especially in
female arnphioxus.
The opioid peptides are other peptides related to reproduction. It is known that the opioid
peptide may play a role in the regulation of reproduction in mammals, including primates.
P-END has demonstrated its inhibition to the secretion of hypothalamic GnRH which regulates
pituitary LH and FSH secretion (Fritz and Speroff 1982, Kalra 1985). The exogenous opiate-like
material such as morphine, could inhibit the ovulation in rats (Packman and Rothehild 1976).
The physiological function of opioid peptides, especially 0-END in arnphioxus, would be a
matter of conjecture, possibly an inhibitory factor for spawning/spermiation, since the
concentration in both sexes was low for a few hours after spawning or spermiation, and was
restored to a normal level by 74 hours. It is, therefore, suggested that the decline in the level of
3-END during breeding may remove the inhibitory factor for gonadotropin production or its
function, and results in spawning and spermiation.
The nervous system and its neurotransmitters regulating gonadotropin hormone and
ovulation has been reported in mammals (Lipner 1973, Fuxe et al. 1979, Kawakami 1979,
Barraclough and Wise 1982, Kalra 1985). It is considered that NE and DA play an important
role on the secretion of GnRH, FSH, and LH as well as on;ovulation, but we know little about
the presence and function of neurotransmitters in the arnphioxus. The present study
demonstrated that the neurotransmitter-NE and DA, not only verified the presence, but also
revealed the relationship with spawning and spermiation iri the amphioxus. On the basis of
increasing the content of NE and DA during spawning and spermiation, it is suggested that these
neurotransmitters may also play an active effect on the secretion of gonadotropin hormone and
ova release, similar to that in higher vertebrates (Barraclough and Wise 1982). The changes in
pattern of NE and DA content before and after spawning in female amphioxus are similar with
NE levels in the rat hypothalamus during the estrous cycle (Stefano and Donoso 1967). In the
males, the contents of NE and DA are higher than those of females and there are no obvious
changes of NE and DA contents during spermiation. The reason is not known, but it may be a
general character of males. Interestly, the changes in patterns of NE and DA during pre- and
post-spawning in female amphioxus are parallel, but it is not similar to changes in human
ovulation in which DA activity decreases and there is no change of NE activity in the human
brain (Paradishi et al. 1987). This functional differentiation of neurotransmitters may represent
one of the evolutionary characteristics.
27
-------
The results presented here demonstrate clearly that the amphioxus not only produced a
pair of regulatory peptides (GnRH and opioid peptides) of reproduction, but also caused
neurotransmitters and steroid hormones (Change et al 1985). These endogenous bioactive
substances existing in the bodies and gonads may be responsible for their synthesis and release,,
as well as participation of regulation in the development of gonads and spawning/spermiation.
Therefore, this regulatory system consisting of steroids, peptides homones, and neurotransmitters
in the amphioxus might be regarded as new evidence for the evolutionary process of reproductive
neuroendocrinology in the vertebrate.
REFERENCES
Barraclough, C. A., and P. M. Wise. 1982. The role of catecholamines in the regulation of
pituitary luteinizing hormone and follicle-stimulating hormone secretion. Endocrinology
Reviews 3:91-110.
Chang, C. Y., Y. X. Liu, and H. H. Zhu. 1985. The reproductive endocrinology of amphioxus.
/;;; Frontiers in Physiological Research. D. Carlick, P. I. Korner P. L, editors. Canberra,
Australia. Academy of Science, pp 79-86.
Crim, L. W., A. Urano, and A. Gorbman. 1979. Immunocytochemical studies of luteinizing
hormone-releasing hormone in brains of agnathan fishes. II. patterns of immunoreactivity in
larval and nurturing western brook lamprey (Lampetra vichardsori). General Comparative
Endocrinology 38:290-299.
Fang, Y. Q., H. Wang, and C. L. Zhang. 1991. Correlation of annual change of luteinizing
hormone-releasing hormone (LH-RH) with gonadal development of Amphioxus. Science in
China, Series B, 34:814-822.
Fritz, M. A., and L. Speroff. 1982. The endocrinology of the menstrual cycle: the interaction of
folliculogenesis and neuroendocrine mechanisms. Fertility and Sterility 38:509-529.
Fuxe, K., K. Andersson, A. Lofstrom, T. Hokfelt, L. Ferland, L.F. Agnati, M. Perez de la Mora,
and R. Schwarcz. 1979. Neurotransmitter mechanisms in control of the secretion of
hormones from anterior pituitary. In: Central regulation of the endocrine system. K. Fuxe, T.
Hokfelt, and R. Luft (Eds). New York, Plenum Press, p. 349.
Gentile, F., O. Lira, and D. M. Cottle. 1986. Relationship between brain gonadotropin-releasing
hormone (GnRH) and seasonal reproductive cycle of "Caribe Colorado" Pygocentrus notatus.
General Comparative Endocrinology 64:239-245.
Kalra, S. P. 1985. Neural circuits involved hi the control of LHRH secretion: a model for
estrous cycle regulation. Journal of Steroid Biochemistry 23:733-742.
28
-------
Kawakami, M., J. Arita, F. Kimura, and R. Hayashi. 1979. The stimulatory role of
catecholamines and acetylcholine in the regulation of gqnadotropin release in
ovariectiomized estrogen-primed rats. Endocrinology of Japan 26:275-284.
Lipner, H. 1973. Mechanism of mammalian ovulation. In: Handbook of Physiology, Section 7,
Endocrinology. R. O. Greep, Editor. Washington D.C., American Physiological Society.
Vol.2, p. 409. :
Mefford, I. N. 1981. Application of high performance liquid chromatography with
electrochemical detection to neurochemical analysis: measurement of catecholamines,
serotonin and metabolites in rat brain. Journal of Neuroscience Methods 3:207-224.
Packman, P. M., and J. A. Rothehild. 1976. Morphine inhibition of ovulation: reversed by
naloxone. Endocrinology 99:7-10. !
Paradishi, R., G. Gross, S. Ventuvoli, M. Capelli, O. Magrin, E. Porcu, R. Pasquali, C. Flamigni.
1987. Evidence for a physiological reduction in brain dopamine but not norepinephrine
metabolism during the preovulatory phase in nomal women. Acta Endocrinology
116:293-298. ;
Shen, W. B., H. Yin, H. Wang, and C. L. Zhang. 1991. Opioid peptides in amphioxus during
breeding season. Chinese Science Bulletin 36:1481-1484. (In Chinese)
Sherwood, N. M., D. A. Lovejoy, I. R. Coe. 1993. Origin of mammalian gonadotropin-releasing
hormone. Endocrinology Reviews 14:241-254.
Stefeno, F. J. E., A. O. Donoso. 1967. Norepinephrine levels in the rat hypothalamus during the
estrous cycle. Endocrinology 81:1405-1406.
Wang, B. Y. 1985. Sensitive immunocytochemiscal methbd-immunogold-silver staining
method. Chinese Journal of Clinical Experimental Pathology 1:41-43. (In Chinese)
Wu, P., J. F. Ackland, N. Ling, and I. M. D. Jackson. 1986; Purification and characterization of
luteinizing hormone-releasing hormone from codfish brain. Regulatory Peptides 15:311-321.
Zhang, C. L., H. Wang, H. Yin, and B. Zhang. 1989. Radioimmunoassay of LH-RHandits
application. Chinese Journal of Applied Pysiology 5:87-89. (In Chinese)
Zhang, C. L., H. Yin, W. B. Shen, and H. Wang. 1993. Two differnt forms of
gonadotropin-releasing hormones in amphioxus. Developmental and Reproductive Biology
2:33-37.
29
-------
Zhang, C L., H. Yin, H. Wang, L. H. Jiao, and W. B. Shen. 1991. The changes of
neurotransmiters in amphioxus during spawning (spermiation) process. Chinese Science
Bulletin 36:694-696. (In Chinese)
Zhu, Y. X., X, B. Guan, R. Y. Cui, H. N. Zhu, and C. L. Zhang. 1986. Preparation of (3-
endorphin antiserum and development of its radioimmunoassay. Academic Journal of
Second Military Medical College 7:332-336. (In Chinese)
30
-------
NEUROENDOCRINE REGULATION OF GONADOTROPIN SECRETION
IN TELEOSTS
Lin Hao-Ran1
ABSTRACT \
The neuroendocrine regulation of gonadotropin (GtH) secretion in teleosts is a complex
and interactive mechanism. The major regulatory mechanism is a dual neurohormonal system,
with GtH release stimulated by a gonadotropin-releasing hormone (GnRH) and inhibited by
dopamine, which functions as a gonadotropin release-inhibitory factor (GRIP) by acting directly
at the pituitary to modulate the action of GnRH as well as the spontaneous release of GtH, and by
inhibiting release of GnRH. sGnRH-A (D-Arg6, Pro9-Net-sGnRH) has the highest affinity to
GnRH receptors in the pituitary and a very high resistance to metabolic degradation, and is the
most potent agonist analog of GnRH in terms of stimulating GtH release in fishes. The specific
D-2 type dopamine receptor antagonist domperidone (DOM) which does not cross the blood-
brain barrier, is the most potent drug in terms of blocking the inhibitory actions of dopamine on
GtH release and potentiate the stimulatory action of GnRH on GtH release. Some other
neurohormones are also involved in regulation of GtH secretion. Norepinephrine, serotonin, and
neuropeptide Y stimulate GnRH and GtH release. GABA (r-amino-bytryic-acid) stimulates
GnRH release and then, causes GtH secretion. The endogenous opioid peptides also influence
GtH secretion.
In common carp, the basal serum GtH level, the GtH-releasing action of DOM and the
GtH-inhibitory action of dopamine and its agonist apomorphine are in correlation with the
ageing. Temperature directly influences basal GtH release, the responsiveness to sGnRH-A, and
the dynamics of GtH secretion from the perifused pituitary fragments of common carp (sexually
recrudescing fish).
Effects of pollution on fish reproduction are mainly due to the disruptive effects of
pollutants on the hormonal production of the endocrine system. For example, in immersion
treatment with high concentration of CdCl2 (9 mg/L) for 12 days, the serum GtH level of
common carp was significantly lower than the controls; hi immersion treatment with high (9
ml/L) and low (0.3 mg/L) concentration of CdCl2 for 14 days, the serum GtH level of common
carp in response to LHRH-A stimulation was also significantly lower than the controls. These
results indicated that long term exposure of common carp to certain concentrations of cadmium
caused a decrease in serum GtH levels and decrease the serum GtH response to LHRH-A
stimulation.
'Department of Biology, Zhongshan University, Ghangzhou, PRC.
31
-------
INTRODUCTION
The neuroendocrine regulation of gonadotropin (GtH) secretion in teleosts is a complex
and interactive mechanisms. Environmental cues (e.g. temperature, photoperiod, rainfall) and
pheromones are perceived by the sensory organs and interpreted by the brain which is involved
in both stimulatory and inhibitory regulation of GtH release from the pituitary. GtH secretion in
teleosts is mainly regulated by a dual neurohormonal system, with GtH release stimulated by a
gonadotropin-releasing hormone (GnRH) and inhibited by dopamine, which functions as a
gonadotropin release-inhibitory factor (GRIP) by acting directly at the level of the pituitary to
modulate the actions of GnRH as well as the spontaneous release of GtH and by inhibiting
release of GnRH (Peter etal 1986,1991).
Recent studies indicated that some other neurohormones are also involved in regulation
of GtH secretion. Norepinephrine (NE) or serotonin (5-HT) treatment induces an increase in
serum GtH levels in goldfish (Carassius auratus) (Chang and Peter 1984, Somoza et al. 1988);
this may, in part, be due to the stimulatory effects of NE or 5-HT on GnRH release (Yu and Peter
1992, Yu et al 1991). NE also has a direct effect on GtH release (Chang et al. 1991).
Neuropeptide Y (NPY) directly stimulates GtH release from dispersed and cultured goldfish
pituitary cells, although NPY also has stimulatory effects on GnRH release (Peng et al. 1990).
Injection of r-amino-butyric acid (GABA) causes an increase hi serum GtH levels in goldfish in
early stages of gonadal recrudescence, but not in fishes that are prespawning or sexually
regressing (Ibid). GABA does not stimulate GtH release from dispersed goldfish pituitary cells
in static or perfusion culture, although it does have a stimulatory effect on GnRH release from
goldfish pituitary fragments, which is explained by its stimulatory effects on GtH release (Ibid).
The endogenous opioid peptides also influence GtH secretion. Treatment of male or female
goldfish, hi late stages of gonadal recrudescence, with the opioid receptor antagonist naloxone,
caused a decrease in serum GtH levels and decreased the serum GtH responses to LHRH-A
(Rosenblum and Peter 1989). However, hi goldfish in the early stages of gonadal recrudescence,
naloxone was stimulatory on serum GtH levels when given alone as well as on the responses to
LHRH-A; naloxone also potentiated the responses to the dopamine antagonist domperiodone in
fish at this stage. The mechanisms underlying the interactions of endogenous opioids with
gonadotrophs remain to be investigated.
GtH released by the pituitary stimulates gonadal development and maturation, in turn,
gonadal steroids feed back to both the brain and pituitary to modulate the release of GtH.
Gonadectomy causes increased blood GtH levels hi rainbow trout (Oncorhynchus mykiss)
(Bommelaer et al. 1981), African catfish (Clarius gariepinus) (Habibi et al. 1989a), and goldfish
(Kobayashi and Stacey 1990) that are suppressible by replacement therapy with estradiol and
testosterone, providing evidence of negative feedback of sex steroids hi teleosts. Recent studies
suggest that the neurotransrnitter GABA (Kah et al. 1990) and dopamine (Peter et al. 1991) may
be involved the mechanisms underlying sex steroid negative feedback. In sexually immature
teleosts, sex steroids appear to have a predominantly positive feedback. For example, hi
Japanese silver eel, injections or implantations of estradiol and testosterone both stimulated an
32
-------
increase in pituitary GtH content and serum GtH levels. Combining the sex steroids treatment
with LHRH-A, the dopamine antagonist domperidone, or LHRH-A plus domperidone further
stimulated pituitary and plasma GtH levels (Lin et al. 1990,1991). A model of known
neuroendocrine regulatory factor which involved in the regulation of GtH secretion in teleosts
is presented in Figure 1.
Other environmental factors
Figure 1. Model for the neuroendocrine regulation of gonadotropin-II release in teleosts, adapted
from Peter et al. (1991). A line with an arrow indicates a stimulatory effect; a line with a bar
indicates an inhibitory effect. Abbreviations: al-noradrenergic receptor, ct-1; estradiol, E2;
y-aminobutryic acid, GABA; gonadotropin-II, GTH-II; norepinephrine, NE; neuorpeptide Y,
NPY; serotonin, 5-HT; testosterone, T; type-1 dopamine receptor, D-l; type-2 dopamine
receptor, D-2.
33
-------
GnRH Agonist Analogs
The native form of GnRH, such as mGnRH and sGnRH, are rapidly degraded by
enzymes in the pituitary, kidney, and liver of gilthead seabream (Spams auratd) (Goren et al
1990, Zohar et al. 1990), which limits its effectiveness in vivo. The substitution of selected D-
amino acids in position-6 of GnRH provides resistance to cleavage between the 5 and 6 positions
(Karten and Rivier 1986). There is also a cleavage site between Pro9-Gly10-NH2 in GnRH, and
substitution of ethylamide for amino acid 10 stabilizes the C-terminal of mGnRH in mammals
(Ibid) and of sGnRH in the gilthead seabream (Zohar et al. 1987,1990).
In a combination of in vivo and in vitro studies on structure-activity relations of agonist
analogs of mGnRH and sGnRH (Trp7, Leu8-mGnRH), the analog of sGnRH-A (D-Arg6,
Pro'NEt-sGnRH) was found to be the most potent in terms of GtH release in goldfish both in
vivo (Peter et al. 1985,1987) and in vitro (Peter et al. 1987, Habibi et al. 1989b), as well as in
vivo in common carp (Cyprinus carpio) and Chinese loach (Paramis gurous dabryanus) (Lin et
al. 1988,1991). sGnRH-A was also found to have the highest affinity to GnRH receptors in the
pituitary of goldfish (Habibi et al. 1989b) and African catfish (De Leeuw et al. 1988). sGnRH-A
is also reported to have a very high resistance to metabolic degradation in the gilthead seabream
(Zohar et al. 1987).
It is apparent that the superactive analog of mGnRH, the LHRH-A (D-Ala6, Pro9-NEt-
LHRH) is also highly active in teleosts; in some fish such as the Atlantic salmon (Salmo solar)
and gilthead seabream (Zohar et al. 1987), LHRH-A is more potent than sGnRH-A in inducing
GtH release and ovulation, although additional dose response studies need to be done to confirm
tliis. However, in receptor binding assays on teleost pituitaries, mGnRH analogs frequently have
lower receptor affinity than analogs of sGnRH(De Leeuw et al. 1988, Habibie/ al 1989b). This
suggests that hi addition to receptor affinity, other factors such as resistance to metabolic
degradation and degree of hydrophobicity may also play a role in determining activity of the
analog, particularly in vivo. sGnRH is hydrophobic relative to mGnRH because of substitution
of Trp7 for Leu7 (Sherwood et al. 1983). Substitution of a hydrophobic D-amino acid in position
6 of sGnRH does not confer superactivity (Peter et al. 1985, Habibi et al. 1989b), presumably
because the analog is excessively lipophilic. Superactive analogs of sGnRH are optimally based
on substitution of a positively charged D-amino acid in position 6, such as D-Arg (Peter et al
1987, Habibie/ al. 1989b).
Dopamine Antagonists
A variety of in vivo and in vitro evidence indicates that dopamine acts directly at the level
of the gonadotroph to inhibit GnRH-stimulated GtH release in teleosts (Peter et al. 1986,
Omeljaniuk et al. 1987,1989, van. Asselt et al. 1988). A wide variety of drugs effective in
blocking dopamine synthesis or depleting catecholamines, or in blocking D-2 type dopamine
receptors, block the inhibitory actions of dopamine on GtH release, and potentiate the actions of
GnRH in vzvo (Peter et al. 1986, Goosetal. 1987, Omeljaniuk et al. 1987, Lin etal. 1988, van
Asselt et al. 1988). Among them, the most potent and highly specific D-2 type dopamine
34
-------
receptor antagonist domperidone (DOM) does not cross the brain-blood barrier and, therefore,
has no central action on the catecholaminergic system in the goldfish (Omeljaniuk et al. 1987),
and presumably in other teleosts.
The inhibitory actions of dopamine on GtH secretion can vary in potency between
species. For example, in goldfish and common carp, the dopamine inhibition on GtH secretion is
very strong, and mGnRH-A and sGnRH-A stimulate only a modest increase in serum GtH levels
and are ineffective in inducing ovulation (Peter et al. 1985,1986,1987, Sokolowska et al. 1985,
Lin et al. 1987). Administration of the dopamine receptor antagonist pimozide (PIM) or
domperidone (DOM) greatly potentiates the action of mGnRH-A and sGnRH-A on GtH release,
and combined injections of PIM or DOM and mGnRH-A or sGnRH-A are highly effective in
inducing ovulation and spawning in these species (Peter et al 1988, Lin et al. 1988). In bream
(Parabramis pekinensis) and Chinese loach, the dopamine inhibitory tone on GtH secretion is
not so strong, and injection of a high dose of mGnRH-A or sGnRH-A alone can overcome the
dopaminergic inhibition and is effective in stimulating GtH release and ovulation. The
combination of PIM with mGnRH-A, however, resulted in potentiation of the GtH response and
shortening of the response time from injection to ovulation (Lin et al. 1985,1986a,b).
I
In some marine fishes, such as Atlantic croaker (Micropagonias undulatus) (Copeland
and Thomas 1989), gilthead seabream (Zohar et al. 1987) and striped bass (Morone saxatilis)
(Sullivan et al. 1992), in a series of experiments, the effects of injection of a number of dopamine
agonists and antagonists on the GtH secretion or ovulation response to LHRH-A were tested, and
no evidence of dopamine inhibitory regulation was found. Injection of a high dose of mGnRH-A
or sGnRH-A alone is effective in stimulating GtH release and ovulation in these species. The
apparent differences between species in the degree of GRIP activity exerted by dopamine
suggests that the relative importance of both dopamine and GnRH in regulating GtH secretion in
teleosts may have altered through evolution.
Effects of Aging
Basal serum GtH levels in common carp at different age. The basal serum GtH levels in
common carp are altered with different ages. The 6-monthrold fish (GSI=9-12%) showed
significantly higher basal serum GtH level (7.57 ± 3.34 ng/ml) than those 1-year-old fish
(GSI=8-13%; GtH=2.36 ± 0.60 ng/ml) and 3-year-old fish (GSI=6.35-11.7%; GtH=2.30 ±
1.38 ng/ml); whereas there is not significant difference between the 1-year- and 3-year-old fish.
The GtH-releasing action of domperidone. In 6-month-old fish, DOM (10 ug/g b.w.)
stimulated GtH release significantly in a period of 24 hours after injection; GtH levels were
3.05- and 3.75-fold higher than the controls at 6 and 24 hours postinjection, respectively.
In 1-year-old fish, DOM also stimulated GtH release significantly within 24 hours after injection,
but the increasing magnitudes of GtH levels were less than those hi 6-month-old fish, only 2- to
2.5-fold higher than the controls. In 3-year-old fish, DOM stimulated GtH release significantly
only at 6 and 24 hours after injection, and the increasing magnitudes of GtH level were 1.6- to
2.0-fold higher than the controls.
35
-------
The GtH-inhibitory action ofdopamine andAPO. In 6-month-old fish, high dose of
dopamine (100 ug/g b.w.) and APO (20 ug/g b.w.) treatments caused significant decrease in
serum GtH levels at 2 and 6 hours after injection; APO even suppressed the serum GtH level at
24 hours postinjection. In 1-year-old fish, dopamine and APO inhibited GtH release at 2 and 6
hours after injection. In 3-year-old fish, the GtH-inhibitory effects ofdopamine and APO were
very briefly, only suppressed GtH release at 2 hours after injection. These results indicated that
the GtH-releasing action of DOM and the GtH-inhibitory actions ofdopamine and APO are in
correlation with the aging.
Effects of Temperature
The effects of temperature on GtH release from perfused pituitary fragments of common
carp (sexually recrudescing fish) in response to sGnRH-A were investigated recently (Lin et al
1996). The basal GtH release of the pituitary fragments was increased with the elevation of
perfusion temperature from 10 to 25 C, with a significant increase in GtH release at 20 and 25 C
compared with that at 10 C. The GtH release from the pituitary fragments in response to
pulsatile administration of three doses of sGnRH-A (100,10 and 1 nM) were also increased at
20 and 25 C compared with those at 10 C. The GtH release responses to the low dose of sGnRH-
A were less sensitive to the temperature changes, and sGnRH-A was shown to have higher
potency at wanner temperature. Continuous exposure of pituitary fragments at 20 or 10 C to
10 and 1 nM sGnRH-A resulted in an initial peak of GtH release followed by desensitization to
further sGnRH-A stimulation; a smaller magnitude of desensitization was observed after
continuous administration of 1 nM sGnRH-A. Continuous exposure of the fragments perfused at
20 C caused a higher initial peak of GtH release than at 10 C, and a significantly higher GtH
release rate within each subsequent hour. These results provide evidence for direct influence of
temperature on the basal GtH release, the pituitary responsiveness to sGnRH-A, and the
dynamics of GtH secretion.
Effects of Pollution
Effects of pollution on fish reproduction are mainly due to the disruptive effects of
pollutants on the endocrine system, and the inhibition of hormone production leading to
developmental disruption of gonads and gametes (Kime 1995). Hormonal assay studies have
shown the inhibitory effects of pollutants on the hypothalamic-pituitary system. Paramar M50,
Cythion, hexadrin, and aldrin decreased hypothalamic GnRH-like substance in freshwater catfish
(Heteropneustesfossilis) (Singh and Singh 1982). Plasma GtH was decreased by Matacid-50
and carbaryl hi the murrel (Channa punctatus) (Ghosh et al. 1990), and r-HCH in goldfish
(Singh et al. 1994). Exposure of Atlantic croaker to Arochlor 1254 inhibited the in vitro
production of GtH (Thomas 1989). Pituitary and plasma GtH was inhibited by Cythion,
hexadrin, aldrin and parathion (Paramer) in freshwater catfish (Singh and Singh 1980a,b,c, 1981,
1982). GnRH stimulated plasma GtH less in fish which had been exposed, to bleached kraft pulp
mill effluent than controls (Van der Kraak et al. 1992). Hypothalamic GnRH may also be
affected by pesticides (Singh and Singh 1982; Ghosh et al. 1990) and this in turn could be
responsible for some effects on GtH secretion.
36
-------
Our recent studies on the effects of cadmium on serum GtH levels in common carp
indicated that in immersion treatment with high concentration of CdCl2 (9 mg/L) for 12 days, the
serum GtH level was significantly lower than the controls (1.82 ± 0.58 ng/ml versus 3.50 ±
0.78 ng/ml). In immersion treatment with high (9 mg/L) and low (0.3 mg/L) concentration of
CdC12 for 14 days, the serum GtH level in response to LHRH-A stimulation was also
significantly lower than the controls (5.43 ± 1.35 ng/ml and 6.81 + 1.72 ng/ml versus 9.79 ± 2.88
ng/ml, respectively) (Ma et al. 1995). These results demonstrated that long-term exposure of
common carp to certain concentrations of cadmium caused;a decrease in serum GtH levels and
decrease the serum GtH response to LHRH-A stimulation.
ACKNOWLEDGEMENTS
This work was supported by the National Natural Sciences Foundation of China and the
International Development Research Centre of Canada. ;
REFERENCES
Bommelaer, M. C., R. Billard, and B. Breton. 1981. Changes in plasma gonadotropin after
ovariectomy and estradiol supplementation at different stages at the end of the reproductive
cycle in the rainbow trout (Salmo gairdneri). Reprod. Nutr. Devel. 21:989-997.
Chang, J. P., and R, E. Peter. 1984. Influences of norepinephrine and a-adrenergic mechanisms
on gonadotropin secretion in female goldfish, Carassius auratus. Gen. Comp. Endocrinol
55:89-95.
Chang, J. P., F. Van Goor, and S. Acharya. 1991. Influences of norepinephrine and adrenergic
agonists and antagonists on gonadotropin secretion from dispersed pituitary cells of goldfish,
Carassius auratus. Neuroendocrinology 54:202-210.
Copeland, P. A., and P. Thomas. 1989. Control of gonadotropin release in the Atlantic croaker
(Micropogonias undulatus): Evidence for lack of dopaminergic inhibition. Gen. Comp.
Endocrinol. 74:474-483.
De Leeuw, R., P. M. conn, C. Van't veer, H. J. Th. Goos and P. G. W. J. Oordt. 1988.
Characterization of the receptor for gonadotropin-releasing hormone hi the pituitary of the
African catfish, Clarias gariepinus. Fish Physiology and Biochemistry 5:99-107.
Ghosh, P., S. Bhattacharya, and S. Bhattacharya. 1990. Impairment of the regulation of gonadal
function hi Channa punctatus by Metacid-50 and Carbaryl under laboratory and field
conditions. Biomed. Env. Sci. 3:106-112.
37
-------
Goos, H. J. Th., K. P. Joy, R. De Leeuw, P. G. W. J. Van Oordt, A. M. L. Van Delft, and J. Th.
Gielen. 1987. The effects of luteinizing hormone-releasing hormone analog (LHRH-A) in
combination with different drugs with anti-dopamine and anti-serotonin properties on
gonadotropin release and ovulation hi the African catfish, Glorias gariepinus. Aquaculture
63:143-156.
Goren, A., Y. Zohar, M. Fridkin, E. Elhanati, and Y. Koch.' 1990. Degradation of gonadotropin-
releasing hormones hi the gilthead seabream, Spams aurata. 1. Cleavage of native salmon
GnRH and mammalian LHRH in the pituitary. Gen. Comp. Endocrinol. 79:291-305.
Habibi, H. R., R. De Leeuw, C. S. Nahorniak, H. J. Th. Goos, and R. E. Peter. 1989a. Pituitary
gonadotropin-releasing hormone (GnRH) receptor activity in goldfish and catfish: seasonal
and gonadal effects. Fish Physiology and Biochemistry 7:109-118.
Habibi, H. R., T. A. Marchant, C. S. Nahorniak, H. Van Der Loo, R. E. Peter, J. E. Rivier, and
W. W. Vale. 1989b. Functional relationship between receptor binding and biological activity
for analogs of mammalian and salmon gonadotropin-releasing hormone in the pituitary of
goldfish (Carassius auratus). Biol. Reprod. 40:1152-1161.
Kah, O., B. D. Sloley, V. L. Trudeau, K. L. Yu, J. P. Chang, and R. E. Peter. 1990. GAB A and
the neuroendocrine regulation of gonadotropin release in the goldfish. Neuroendocrinology
52 (Suppl) 93 (Abstract p2.88).
Karten, M. J., and J. E. Rivier. 1986. Gonadotropin-releasing hormone analog design. Structure-
function studies toward the development of agonists and antagonists: rationale and
perspective. Endocrinol. Rev. 7:44-66.
Kobayashi, M., and N. E. Stacey. 1990. Effects of ovariectomy and steroid hormone
implantation on serum gonadotropin levels hi female goldfish. Zool. Sci. 7:715-721.
Kime, D. E. 1995. The effects of pollution on reproduction in fish. Fish Biology and Fisheries
5:52-96.
Lin, H. R., C. Peng, L. Z. Ziu, S. J. Zhou, G. Van Der Kraak, and R. E. Peter. 1985. Induction
and ovulation in the loach (Paramisgurnus dabyanus) using pimozide and (D-Ala6,Pro9-N-
Gthylamide)-LHRH. Aquaculture 46:333-340.
Lin, H. R., G. Van Der Kraak, J. Y. Liang, C. Peng, G. Y. Li, L. Z. Liu, X. J. Zhou, M. L. Chang,
and R. E. Peter. 1986a. The effects of LHRH analogue and drugs which block the effects of
dopamine on gonadotropin secretion and ovulation hi fish cultured in China. In: Aquaculture
of Cyprinids. R. Billard and J. Marcel (Eds.), INRA, Paris, France, pp. 139-150.
38
-------
Lin, H. R., G. Van Der Kraak, R. E. Peter, and B. Breton. 1986b. Effects of (D-Ala6,Pro9-NEt)-
LHRH and catecholaminergic drugs on gonadotropin secretion and ovulation in the Chinese
loach (Paramisgurnus dabryanus). Gen. Comp. Endocrinol. 64:389-395.
Lin, H. R., X. J. Zhou, G. Van Der Kraak, and R. E. Peter. 1987. Comparison of LHRH-A and
sGnRH-A, in combination with pimozide or domperidone, in stimulating gonadotropin
release and ovulation in the Chinese loach (Paramisgurnus dabryanus). In: Proceedings of
the Third International Symposium on Reproductive Physiology of Fish. L. W. Crim, D. R.
Idler, and J. Walsh (eds.). p. 33. '.
Lin, H. R:, G. Van Der Kraak, X. J. Zhou, J. Y. Liang, R. E. Peter, J .E. Rivier, and W. W. Vale.
1988. Effects of sGnRH-A and LHRH-A, in combination with pimozide or domperidone, on
gonadotropin release and ovulation in the Chinese loach and common carp. Gen. Comp.
Endocrinol. 69:31-40.
Lin, H. R., M. L. Zhang, S. M. Zhang, G. Van Der Kraak, and R. E. Peter. 1990.. Effects of sex
steroids, (D-Ala6,Pro9-N-ethylamide)-LHRH (LHRH-A) and domperidone (DOM) on
gonadotropin secretion in female silver eel, Anguillajaponica. In: The Second Asian
Fisheries Forum. Temminck and Schlegel (eds.). Asian Fisheries Society, Manila,
Philippines, pp. 591-594.
Lin, H. R., M. L. Zhang, S. M. Zhang, G. Van Der Kraak, and R. E. Peter. 1991. Stimulation of
pituitary gonadotropin secretion and ovarian development by chronic administration of
testosterone in female Japanese silver eel, Anguillajaponica. Aquaculture, 96:87-95.
Lin, X. W., H. R. Lin, and R. E. Peter. 1996. Direct influences of temperature of gonadotropin-II
release from perfused pituitary fragments of common carp (Cyprinus carpio L.) in vitro.
Gen. Comp. Endocrinol. Comp. Biochem. Physiol., 114A:341-347.
Ma G., H. R. Lin, and W. Zhang. 1995. Effects of cadmium on serum gonadotropin and growth
hormone in common carp (Cyprinus carpio). Journal of Fisheries of China 19:120-126.
Omeljaniuk, R. J., S. H. Shih, and R. E. Peter. 1987. In vivo evaluation of dopamine receptor-
mediated inhibition of gonadotropin secretion from the pituitary gland of the goldfish,
Carassius auratus. Journal of Endocrinology 114:449-458.
Omeljaniuk, R. J., H. R. Habibi, and R. E. Peter. 1989. Alterations in pituitary GnRH and
opamine receptors associated with the seasonal variation and regulation of gonadotropin
release in the goldfish {Carassius auratus). Gen. Comp. Endocrinol. 74:392-399.
Peng, C., Y. P. Huang, and R. E. Peter. 1990. Neuropeptide Y stimulates growth hormone and
gonadotropin release from the goldfish pituitary in vitro. Neuroendocrinology 52:28-34.
39
-------
Peter, R. E., C. S. Nahorniak, M. Sokolowska, J. P. Chang, J. E. Rivier, W. W. Vale, J. A. King,
and R. P. Millar. 1985. Structure-activity relationships of mammalian, chicken and salmon
gonadotropin-releasing hormone in vivo in goldfish. Gen. Comp. Endocrinol. 58:231-242.
Peter, R. E., J. P. Chang, C. S. Nahorniaki R. J. Omeljaniuk, M. Sokolowska, S. H. Shih, and R.
Billard. 1986. Interactions of catecholamines and GnRH hi regulation of gonadotropin
secretion in teleost fish. Recent Prog. Horm. Res. 42:513-548.
Peter, R. E., H. R. Lin, and G. Van Der Kraak. 1987. Drug/hormone induced breeding of
Chinese teleosts. In: Proceedings of the Third International Symposium on Reproductive
Physiology of Fish. D. R. Idler, L. W. Crim and L. M. Walsh (eds.). St. John's,
Newfoundland, Canada, pp. 120-123.
Peter, R. E., H. R. Lin, and G. Van Der Kraak. 1988. Induced ovulation and spawning of
cultured freshwater fish in China: Advances in application of GnRH analogues and dopamine
antagonists. Aquaculture 74:1-10.
Peter, R. E., V. L. Trudeau, and B. D. Sloley. 1991. Brain regulation of reproduction in teleosts.
Bull. Inst. Zool., Academia Sinica, Monograph 16:89-118.
Rosenblum, P. M., and R. E. Peter. 1989. Evidence for the involvement of endogenous opioids
in the regulation of gonadotropin secretion hi male goldfish, Carassius auratus. Gen. Comp.
Endocrinol. 73:21-27.
Sherwood, N. M., L. Eiden, M. Brownstein, J. Spiess, J. Rivier, and W. Vale. 1983.
Characterization of a teleost gonadotropin-releasing hormone. Proceedings of the National
Academy of Science, USA 80:2794-2798.
Singh, H., and T. P. Singh. 1980a. Effects of two pesticides on testicular 32P uptake,
gonadotrophic potency, lipid and cholesterol content of testis, liver and blood serum during
spawning phase in Heteropneustes fossilis (Bloch). Endokrinologie 76:288-296.
Singh, H., and T.P. Singh. 1980b. Short-term effect of two pesticides on the survival, ovarian
32P uptake and gonadotrophic potency hi a freshwater catfish (Heteropneustes fossilis)
(Block). Journal of Endocrinology 85:193-199.
Singh, H., and T. P. Singh. 1980c. Effects of two pesticides on ovarian 32P uptake and
gonadotropin concentration during different phases of annual reproductive cycle in the
freshwater catfish, Heteropneustes fossilis (Bloch). Env. Res. 22:190-200.
Singh, H., and T. P. Singh. 1981. Effect of parathion and aldrin on survival, ovarian 32P uptake
and gonadotrophic potency hi a freshwater catfish, Heteropneustes fossilis (Bloch).
Endokrinologie 77:173-178.
40
-------
Singh, H., and T. P. Singh. 1982. Effects of pesticides on fish reproduction. Ichthyologia
15:7181. ;
Singh, P. B., D. E. Kime, P. Epler, and J. Chyb. 1994. Impact of r-hexachlorocyclohexane
exposure on plasma gonadotropin levels and in vitro stimulation of gonadal steroid
production by carp hypophyseal homogenate in Carassius auratus. Journal of Fish Biology
44:195-204.
Sokolowska, M., R. E. Peter, and C. S. Nahorniak. 1985. The effects of different doses of
pimozide and (D-Ala6,Pro9-N-ethylamide)-LHRH (LHRH-A) on gonadotropin release and
ovulation in female goldfish. Canadian Journal of Zoology 63:1252-1256.
!
Somoza, G. M., K. L. Yu, and R. E. Peter. 1988. Serotonin stimulates gonadotropin release in
female and male goldfish, Carassius auratus. L. Gen. Comp. Endocrinol. 72:373-382.
Sullivan, C. V., R. G. Hodson, Y. Tao, W. King, L. F. Jackson, P. Swanson, P. Thomas, L.
Helfrich, W. Blythe, R. M. Harrell, and L. C. Woods III. 1992. Reproduction of striped bass,
Morone saxatilis: Basic and applied research on hormones. In: Abstract L25 of 2nd
International Symposium on Fish Endocrinology, June 1-4,1992, St. Malo, France.
Thomas, P. 1989. Effects of Arochlor 1254 and cadmium on reproductive endocrine function
and ovarian growth in Atlantic croaker. Mar. Env. Res. 28:499-503.
van Asselt, L. A. C., H. J. Th. Goos, W. Smit-Van Dijk, P. A. M. Speetjens, and P. G. W. J. Van
Oordt. 1988. Evidence for the involvement of D-2 receptors in the dopaminergic inhibition
of gonadotropin release in the African catfish, Glorias gariepinus. Aquaculture 72:369-378.
Van der Kraak, G. J., K. R. Munkittrick, M. E. McMaster, C. B. Portt, and J. P. Chang. 1992.
Exposure to bleached craft pulp mill effluent disrupts the pituitary-gonadal axis of white
sucker at multiple sites. Toxicol. Appl. Pharmacol. 115:224-233.
Yu, K. L., P. M. Rosenblum, and R. E. Peter. 1991. In vitro release of gonadotropin-releasing
hormone from the brain preoptic-anterior hypothalamic region and pituitary of female
goldfish. Gen. Comp. Endocrinol. 81:256-267.
Yu, K. L., and R. E. Peter. 1992. Adrenergic and dopaminergic regulation of gonadotropin-
releasing hormone release from goldfish preoptic-anterior hypothalamus and pituitary in
vitro. Gen. Comp. Endocrinol. 85:138-146.
i
Zohar, Y., G. Pagelson, M. Tosky, and Y. Finkelman. 1987. GnRH control of gonadotropin
secretion, ovulation and spawning in the gilthead seabream, Sparus aurata. In: Proceedings
of the Third International Symposium on Reproductive Physiology of Fish. D. R. Idler, L.
W. Crim and L. M. Walsh (eds.). St. John's, Newfoundland, Canada, pp. 106.
41
-------
Zohar, Y. 1989. Fish reproduction: Its physiology and artificial manipulation. In: Fish Culture
in Warm Water Systems: Problems and Trends. M. Shilo and S. Sarig (eds.). CRC Press,
Inc., Boca Raton, Florida, pp. 65-119.
Zohar, Y., A. Goren, M. Fridkin, E. Elhanati, and Y. Koch. 1990. Degradation of gonadotropin-
releasing hormones in the gilthead seabream, Spams aurata. II. Cleavage of native salmon
GnRH, mammalian LHRH, and their analogs hi the pituitary, kidney, and liver Gen Comp
Endocrinol. 79:306-319.
42
-------
XENOBIOTIC-METABOLIZING ENZYMES IN RAINBOW TROUT
AS WATER POLLUTION BIOMONITORS
Arvo Tuvikene,1-2 Sirpa Huuskonen,3 Sashwati Roy3 and Pirjo Lindstrom-Seppa3
ABSTRACT
Although residues of chemicals in fish and in the environment are routinely measured,
such measurements do not provide information of what kind of non-lethal stress these chemicals
can cause for fish. Such can be detected, however, by studying metabolic effects caused by
certain xenobiotics. The aim of this study was to evaluate the xenobiotic metabolism of rainbow
trout (Oncorhynchus mykiss) caged for 3 weeks in selected inland waterbodies of Estonia. Such
monitoring helps to keep an eye on the quality of water to prevent hazards to human and aquatic
health. Those waterbodies are important fishing and recreation areas and also sources of
drinking water. Hepatic monooxygenase as well as conjugation enzyme activities were measured
in the livers of the caged rainbow trout. In order to evaluate the degree of pollution the content
of PAHs, the most expected contaminants, were analyzed in selected fish muscle samples.
INTRODUCTION
The residues of pollutants in organisms or in the environment can be measured by
chemical analyses. However, such analyses do not provide information regarding what kind of
biologic stress these chemicals may have caused the organism. Some of the biologic stress
indicators within organisms are not specific. These are affected by many factors and do not
reveal the exact effects of a pollutant. However, xenobiotic metabolism in fish has been
introduced as a sensitive, as well as specific, sign of aquatic pollution caused by xenobiotic
chemicals (Stegeman et al. 1992). Certain chemicals can induce detoxification systems within
fish, and the induction can be detected by measuring the enzyme activities. For example,
polynuclear aromatic hydrocarbons (PAHs) and halogenated aromatic hydrocarbons (HAHs)
have been shown to increase the amount of cytochrome P4501A protein and the activities of
monooxygenases (MOs) representing this enzyme protein (Stegeman and Hahn 1994). The
activation of conjugation reactions can be studied by measuring transferase activities (Clark et al.
1991, George 1994). Xenobiotic metabolism is regarded as an adaptive physiological response
which usually helps organisms to cope with pollution. In some cases this biotransformation
produces reactive metabolites which can bind to cell macromolecules like DNA, RNA and
proteins. This can raise genotoxic, mutagenic, or carcinogenic responses.
'University of Tartu, Institute of Zoology and Hydrobiology, Tartu, Estonia.
2Limnological Station, Institute of Zoology and Botany, Estonian Academy of Sciences, Rannu, Estonia.
3University of Kuopio, Department of Physiology, Kuopio, Finland. '
43
-------
In the field, xenobiotic metabolism offish has previously been used in biomonitoring
bleached and unbleached pulp mill effluents (Hodson et al. 1991, Huuskonen and Lindstrom-
Seppa 1995), or PAH contaminated sites (Van Veld et al. 1990, Di Giulio etal. 1993). In the
field studies it has been possible to measure the rate of induction hi enzyme activities. Based on
this knowledge, the size of the contaminated area and the degree of the effluent dilution can be
estimated.
The aim of the present study was to evaluate the xenobiotic metabolism of rainbow trout
(Oncorhynchus mykiss) caged for 3 weeks hi selected inland waterbodies of Estonia. Such
monitoring helps to keep an eye on the quality of water to prevent health hazards. These
waterbodies are important fishing and recreation areas, and also sources of drinking water for
approximately 80,000 people. Hepatic monooxygenase as well as conjugation enzyme activities
were measured. In order to evaluate the degree of pollution the content of PAHs, the most
expected contaminants, were analyzed hi selected fish muscle samples.
MATERIAL AND METHODS
Introduction of Study Areas
Lake Vortsjarv (270 km2) is the second largest lake hi Estonia. It is a strongly eutrophic
lake, although it is an important commercial fishing and recreation area. In Lake Vortsjarv, the
xenobiotics originate mostly from the surrounding arable land, and from urban sewage.
Professional and amateur fisheries are relatively well developed at this lake. Fishing vessels may
cause oil spillage, especially hi the harbors most intensively used.
The middle course of the River Suur Emajogi is one of the most polluted waters hi
southern Estonia. The main pollution comes from the town of Tartu (population 114,000), which
in 1994 released roughly 14 million m3 of mostly untreated, or insufficiently purified, industrial
and municipal sewage into the river.
Serious pollution of natural waterbodies hi northeast Estonia is mostly connected with
oil-shale mining (River Rannapungerja), and its combustion and thermal processing (River
Narva), thereby releasing PAHs, phenols, and heavy metals into the environment.
Test Fish
Immature (0+) rainbow trout were caged for 3 weeks hi selected sites hi Lake Vortsjarv
and River Suur Emajogi during winter and spring, hi the Rivers Rannapungerja and Uhe during
winter, and in the River Narva during fall (Figure 1).
44
-------
CAGING SITE 0 10 20 "30 40 50km
Figure 1. Caging sites in Tartu County (southern Estonia) and Ida-Virumaa County (northeast
Estonia). (VO, EO, RO and NO express reference areas; VI, E1-E3, Rl and N1-N2 are marks
for exposed areas).
Analytical and Statistical Methods
After exposure in the cages, the fish were removed and fish liver samples were collected
and analyzed for enzyme activity as follows: hepatic monooxygenase activities by deethylation
of 7-ethoxyresorufin (EROD) according to Burke and Mayer (1974), benzo(a)pyrene
hydroxylase (AHH) activity according to Nebert and Gelboin (1968), conjugation activities for
UDP-glucuronosyltransferase (UDP-GT) activity according to Hanninen (1968), and glutathione
45
-------
S-transferase (GST) activity according to Habig et al. (1974). Protein content was measured
according to Bradford (1976). The PAH contents in fish, sediments, and water samples were
measured using gas chromatography-mass spectrometry (GC-MS). The assumption of equal
variances was tested with Cochran's C test and thereafter the data were tested with a
nonparametric Kruskal-Wallis (K-W) one-way analysis of variance with SPSS/PC+ software
(Chicago, Illinois, U.S.).
RESULTS AND DISCUSSION
In winter, there were no statistically significant changes in biotransformation enzyme
activities in fish in Lake Vortsjarv and River Suur Emajogi (K-W: EROD P=0.087, AHH
P=0.060). However, in River Suur Emajogi the trend was toward increased MO activities in fish
at the exposed areas (El, E2), compared to fish from the upstream reference area (EO) (Figure 2).
vo vi
L. Vortsjarv
EO
E1 E2
R. Suur Emajogi
E3
VO V1
L. Vortsjarv
EO
E1 E2 E3
R. Suur Emajogi
Figure 2. Responses of ethoxyresorufin O-deethylase (EROD) and arylhydrocarbon hydroxylase
(AHH) activity in rainbow trout caged in Lake Vortsjarv (VO as a reference area; VI as an
exposed area) and River Suur Emajogi (EO as a reference area; El, E2, E3 as exposed areas)
in winter and in spring.
46
-------
At its highest, there was a five-fold increase in EROD activities and a three-fold increase
in AHH activities at 3 km downstream from Tartu, compared to the reference area upstream from
Tartu. When the MO activities in rainbow trout cages in winter at the harbor area in Lake
Vortsjarv (VI) were compared to the reference values upstream from Tartu, there was an eight-
fold increase in EROD and a four-fold increase in AHH activities. The lack of statistical
evidence in K-W was due to inadequate number offish at EO and high variation between
individuals.
In spring significant changes were observed in EROD and AHH activities (K-W, PO.001
with both MOs). In Lake Vortsjarv, EROD activities were three times higher and AHH activities
two times higher in fish at the harbor area (VI) than fish at the lake area (VO). Caging studies in
River Suur Emajogi did not show any significant differences of MO activities in fish between
these study sites. The rate of MO induction in this region was higher in winter than in spring
when the contaminants in River Suur Emajogi were much more diluted due to the much higher
run-offs at that time. The effluent dilution was nearly 77:1 and 333:1 for winter and spring,
respectively. This may be the reason for the lack of strong MO induction at the downstream
study location in spring.
While the MO activities of rainbow trout showed elevated levels at exposed areas of
southern Estonia (Figure 2), there were no significant changes in conjugation enzyme activities.
Based on these findings it can be suggested that cytochrome P450 was mainly affected by
pollutants which included at least PAHs.
In the River Narva (northeast Estonia) there were no significant changes in MO activities.
However, GST activities were significantly higher (1.2 times) at the study site near the piles of
waste ash from the thermal power plant (Nl), compared to the reference area (NO). In River
Rannapungerja, no significant changes could be observed hi xenobiotic metabolism offish. In
the River Rannapungerja and River Narva, the lack of indu6tion in MO activities was surprising
because the content of PAHs in these waters was higher than in the waterbeds of southern
Estonia, e.g., the total content of PAHs in the white muscle of rainbow trout at site Rl (River
Rannpungerja) was five-fold higher than the highest concentrations in southern Estonia (VI,
Lake Vortsjarv), and the content of benzo(a)pyrene at site Rl (River Rannapungerja) which is
shown to be a strong inducer of MO was especially high (Table 1).
Table 1. Total PAHs in muscle samples of rainbow trout.
Study Area
PAHs in fish ng/g DW
VI
EO
El
Rl
Nl
738
20
165
3760
2013
47
-------
Probably there were other contusing components in water which could cause the lack of
MO activities. Such an event could be due to heavy metals which are known to inhibit MO
activities. Heavy metals have been detected in reasonable amounts in these waters. The role of
sulphates and phenols, which are also prominent hi these waters, is unclear. Further
investigations of this problem are needed.
CONCLUSIONS
The highest EROD and AHH activities in caged rainbow trout were seen at the harbor
area in Lake Vortsjarv. The effect of urban sewage was seen in elevated EROD and AHH
activities in caged fish downstream from Tartu. The rate of MO induction in the River Suur
Emajogi was higher in whiter than in spring, which is attributed to the more diluted contaminants
in spring during much higher run-offs. The only significant difference among the conjugation
enzymes (GST) was seen at River Narva. The results obtained in the cagings conducted in the
River Suur Emajogi and Lake Vortsjarv confirm that EROD and AHH assays can be used hi
environmental biomonitoring to detect the induction caused by PAHs. However, when the
pollution source is more complex, like it was hi the Rivers Rannapungerja and Narva, the role of
other enzyme activities (e.g., GST) hi xenobiotic metabolism offish may increase.
ACKNOWLEDGMENTS
This study was supported by the Environmental Fund of Tartu County, and by the Fish
Capital of Estonia. Marina Trapido, Lea Tuvikene, Arno Tull, and Evald Jegorov from Estonia,
as well as Eeva-Liisa Palkispaa and Riita Venalainen from Finland, are gratefully acknowledged
for their technical help.
REFERENCES
Clark, D. J., S. G. George, and B. Burchell. 1991. Glucuronidation in fish. Aquatic Toxicology
20:35-56.
Bradford, M.M. 1976. A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry
72:248-254.
Burke, M.D., and R.T. Mayer. 1974. Ethoxyresorufin: direct fluorimetric assay of a
microsomal O-dealkylation which is preferentially inducible by 3-methylcholanthrene. Drug
Metabolism and Disposition 2:583-588.
48
-------
Di Giulio, R. T., C. Habig, and E. P. Gallagher. 1993: Effects of Black Rock Harbor sediments
on indices of biotransformation, oxidative stress, and DNA integrity in channel catfish.
Aquatic Toxicology 26: 1-22.
George, S. G. 1994. Enzymology and molecular biology of phase II xenobiotic-conjugating
enzymes in fish. In: Aquatic Toxicology: Molecular, Biochemical, and Cellular Perspectives,
D. C. Malins and G. K. Ostrander (Eds.),CRC Press, USA: 37-85.
Habig, W.H., MJ. Pabst, and W.B. Jakoby. 1974. Glutathidne S-transferase. The first
enzymatic step in mercapturic acid formation. Journal of Biological Chemistry 249-7130-
7139.
Hanninen, O. 1968: On the metabolic regulation in the glucuronic acid pathway in the rat
tissues. Annales Academiae Scientiarum Fennicae. A2 Chemica 1-96.
Huuskonen, S., and P. Lindstrom-Seppa. 1995. Hepatic cytochrome P4501A and other
biotransformation activities in perch (Percafluviatilis): the effects of unbleached pulp mill
effluents. Aquatic Toxicology 31: 27-41. ;
Hodson, P, V., D. Bussieres, M. M. Gagnon, J. J, Dodson, C. M. Couillard, and J. C. Carey.
1991. Review of biochemical, physiological, pathological and population responses of white
sucker (Catostomus commersoni) to BKME in the St. Maurice river, Quebec. Environmental
fate and effects of bleached pulp mill effluents. Proceeding of a SEP A Conference,
Stockholm, pp. 261-269. ;
Nebert, D.W., and H.V. Gelboin. 1968. Substrate-inducible microsomal aryl hydroxylase in
mammalian cell culture I. Assay and properties of induced enzymes. Journal of Biological
Chemistry 243:6242-6249.
Stegeman, J. J., M. Brouwer, R. T. Di Giulio, L. Forlin, B. A. Fowler, B. M. Sanders, and P. A.
Van Veld. 1992. Enzyme and protein synthesis as indicator of contaminant exposure and
effect. In: Aquatic Toxicology: Molecular, Biochemical, and Cellular Perspectives, D. C.
Malins and G. K. Ostrander (eds.), CRC Press, USA, pp.: 87-206.
Stegeman, J. J., and M. E. Hahn. 1994. Biochemistry and molecular biology of
monooxygenases: current perspectives on forms, functions, and regulation of cytochrome
P450 in aquatic species. In: Malins, D. C. and Ostrander, G. K. (eds.) Aquatic Toxicology:
Molecular, Biochemical, and Cellular Perspectives, CRC Press, USA: 87-206.
Van Veld, P. A., D. J. Westbrook, B. R. Woodin, R. C. Hale, C. L. Smith, R. J. Huggett, and J. J.
Stegeman. 1990. Induced cytochrome P-450 hi intestine and liver of spot (Leiostomus
xanthurus) from a polycyclic aromatic hydrocarbon contaminated environment. Aquatic
Toxicology 17: 119-132.
49
-------
-------
PARASITES AS BIOINDICATORS FOR ESTIMATION OF AQUATIC POLLUTION
Boris I. Kuperman1 ;
ABSTRACT
We propose three new approaches for the evaluation of anthropogenic impact on water
bodies by using fish parasites as bioindicators of water contamination. The first approach is
based on the fact that pathogenic factors in water bodies cause increases in numbers of fishes
with asymmetric bilateral structure. It has been shown that disturbances offish morphogenesis
stability are correlated with asymmetry in the invasion offish gills by monogenean of the genus
Dactylogyrus, and that the ratio of asymmetrically infected :fish to symmetrically infected ones
increases in polluted areas of the Volga River system. The second approach relates to
acidification of water bodies. Acidification in the Darwin National Reserve (Vologda region)
has been accompanied by a decrease of parasite species diversity and an increase in the rate of
dominance of a lesser number of species. There is a definite correlation between the level of
perch infestation with parasitic Apiosoma and the pH of water in some Karelian lakes; the most
intensive infestation of of Apiosoma in perch gills is in the most acidified lakes. The third
approach relates to some parasite species being resistant to complex environmental pollution.
Industrial wastes of the Cherepovetz metallurgical works in the Sheksna part of the Rybinsk
Reservoir contained PCBs, PAHs, oil-products, and heavy metals, and the number of
ectoparasites in the gills of bream collected in this region were significantly lower than elsewhere
in the reservoir. At the same time, however, the monogenean Diplozoon paradoxum, and the
endoparasite cestode Caryophyllaeus laticeps showed a high resistance to the water pollution.
The number of these parasites in bream was increased in the area of direct action of pollutants.
Thus, some species of parasites can be used as the biological indicators of anthropogenic
pollution, and the ecological and sanitary state of the reservoir.
INTRODUCTION
Abnormalities and fish diseases associated with pollution have been reported from a
number of regions and countries (Khan and Thulin 1991, Qverstreet 1993, MacKenzie et al.
1995). During the last decade it has become clear that several fish diseases in wild fish
population occur more frequently in highly polluted areas. Data on the influence of industrial
and domestic sewage waters on fish parasites in the water bodies of Russia are scarce (Laimann
1957, Kostarev 1980, Anikieva 1982, Bogdanova 1988). New approaches to the estimation of
the anthropogenic effect on water bodies of northwest Russia using parasites have been proposed
in the Laboratory of Ecological Parasitology, Institute of Biology of Inland Waters (Kuperrnan
1992, Kuperman et al. 1994, Zharikova 1993, Zhokhov and Tyutin 1994). These approaches are
based on the concept that the parasites offish and invertebrates constitute a significant integral
part of the ecosystem and frequently function as regulating factors in the biota of aquatic
'Institute of Biology of Inland Waters, Russian Academy of Sciences, Borok, Yaroslavl, Russia.
51
-------
systems. Pollution of aquatic systems leads to sufficient changes in the structure of the
hydrobiont communities, and among others B parasites communities. As a result, biocenotic
relationships in a waterbody get disturbed. This, in turn, causes reduction in number or total
disappearance of highly sensitive invertebrates and vertebrates--the intermediate hosts of
parasites. In contrast, some groups of invertebrates are resistant to certain pollution and their
numbers in polluted areas increase. Free-living stages of helminths - oncomiracidia, coracidia,
and miracidia - are rather susceptible to the effect of pollution.
The response of several groups of parasites to the impact of contamination of diverse
chemical nature varies (Overstreet 1993). The invasion rate of some species of parasites
increases, while of others it decreases, under pollution. The influence of toxicants on the course
of infective processes in fishes, and fish susceptibility to parasites has been experimentally
shown (Flerov et al. 1982). Our goals are to study the influence of various pollutants on the
species diversity offish parasites, on their prevalence and morphology in the northwestern
waterbodies of Russia - the Volga River basin, the Vologda region, and Karelia - and to use
parasites for the evaluation of the ecological state of these waterbodies.
MATERIALS AND METHODS
Thermal Effects
Studies on the infestation of yearlings of cyprinid fish by the protozoan Ambiphrya
ameiuri (Ciliophora, Peritricha) have been carried out at the spawning places in the shallow
well-heated shore zone in the Volga part of the Rybinsk Reservoir (the Volga basin) in
June-July 1989. The infestation of 76 specimens of larvae and fry of white bream (Blicca
bjoerknd) and 24 larvae and fry specimens of common bream (Abramis bramd) have been
examined. The count and location of A. ameiuri on the body of live fish were recorded. In
addition, these parasites were fixed with Schaudin fixative and subsequently stained by
Heidenhein hematoxilin. The A. ameiuri distribution on the fish body surface and its
ultrastructure were studied using a scanning electron microscope (SEM).
Acidification Effects
The influence of water acidification on parasites was investigated in 13 lakes in South
Karelia (Russia) in 1991. Two of the lakes were circumneutral (pH 7.0 - 7.9), while others were
acidic (pH 4.5 - 5.8); the lakes were low in mineral content, and were colored or had clear water.
The gills of 200 perch (Percafluviatilis) from all these lakes were examined by SEM.
Industrial Chemical Effects
The impact of the complex pollution in the industrial sewage from Cherepovetz
metallurgical work was studied hi the Sheksna part of the Rybinsk Reservoir in 1988-91.
The wastes included PCBs, PAHs, oil-products, and heavy metals (Pb, Cd, Cu, Cr). For
comparison, fish parasites from other, relatively clean, parts of the Rybinsk Reservoir (Volga and
Mologa parts) were also examined. The fish examined were common bream; 120 specimens
52
-------
were collected from the Sheksna part, 106 specimens from the Volga part, and 196 from the
Mologa part. Structural anomalies of the attachment apparatus ofDiplozoonparadoxum were
studied using SEM. All tissues for SEM investigation were fixed in 3% glutaraldehyde in 0.1 M
sodium cacodylate buffer (pH -7.4) for 3 hours and postfixed with 1% OsO4 in the same buffer
for 1 hour, dehydrated, critical-point dried, and gold-sputter coated. Specimens were then
examined under the SEM (JSM-25S). '
RESULTS AND DISCUSSION
Thermal Effects
The influence of the thermal waters has been traced on the protozoan Peritricha,
A. ameiuri. Previously A. ameiuri was considered as a specific parasite of North American
catfish (Ictalurus punctatus) transferred to Europe and found in fish farms only. A. ameiuri was
brought together with its host to many thermal water fish farms of Russia and infected salmonid
fry and cyprinid fishes (Solomatova and Lussin 1981, Strelbitzkaya 1986). We have found a
mass infection with peritrichoid A. ameiuri of larvae and fry of the carp fishes, white bream
(Blicca bjoerknd) and common bream in the shallow, well-heated shore zone of the Rybinsk
Reservoir (Kuperman et al 1994). The prevalence of invasion was 100%, with intensity from
50 to 200 protozoan specimen per fish. This was the first recording of this protozoan species in
natural conditions. The outburst of the peritrichoid infestation of fry in the reservoir is probably
connected with favorable conditions for their reproduction during the hot summer period of
1989. The mass invasion of the fry of cyprinid fish by A. ameiuri in spawning-places of the
Rybinsk Reservoir was initiated by the following factors: the high temperature of the shallow
waters (up to 29 C); the low water level; the high organic content; and the great number of
bacterioplankton which is the main food for these infusoriae.
The A. ameiuri were evenly spread on the larvae body surface and more densely on the
fins (Figure 1 ). It is the skin respiration and gas exchange offish larvae that is very significant.
On the other hand the parasite scopula, which enables attachment to the larva, covers a great part
of the epithelial surface (Figure lc,d). Totally, it covers up to 50-60% of the body surface area,
and thereby supresses the gas exchange in fish larvae which sometimes leads to their high
mortality.
Aquatic Acidification
The next approach is connected with evaluation of parasite species diversity under the
acidification of water. In recent years acid atmospheric deposition has had a significant impact
on the biota of the northern lakes of Russia. It is especially true for small lakes from the Vologda
region and Karelia. Acidification is associated with mobilization of some metals. The combined
effect of the metals and low pH is toxic to fishes. That is why fish communities are represented
only by perch. The parasitofauna of perch has been studied in 12 lakes with pH 4.4-7.9 in the
Darwin National Reserve (Vologda region). Acidification in these lakes is accompanied by a
decreasing of parasite species diversity and increasing in the degree of dominance of separate
species (Table 1). The maximal number of parasite species (19) has been observed in the perch
53 .
-------
D
Figure 1. Distribution ofAmbiphrya ameiuri on the larvae body surface ofBticca bjoerkna, SEM.
(a) head; (b) pectoral fin; (c) caudal fin; (d) attachment of A. ameiuri with scopula. Scale bar :
(a) (c) 100 nm; (b,d) 10
54
-------
from neutral lakes (pH 6.6-7.5), while in acidic lakes (pH 6.0-6.5) there were seven parasitic
species, and in the most acidified ones (pH 4.1-5.5) - the number was reduced to four species:
two species of helminths and two protozoa. The most adapted to acid waters are protozoan
Apiosoma minimicronudeatum, cestoda Proteocephalus percae and nematoda Camallanus
lacustris. The level of invasion with C. lacustris and P. percae in acid lakes reaches 100%.
The spreading of these helminths in acid lakes is provided by the presence of intermediate hosts
resistant to acidification, so the life cycle of C. lacustris includes a wide circle of intermediate
hosts: Copepoda, Isopoda, larvae of insects, and yearlings of cyprinid fishes. Besides, these
nematoda are live-bearing and from fully developed larvae with a dense protective cuticular
tegument.
Table 1. Perch parasites acidified lakes in Darwin National ;Reserve (Vologda region)
(Zhokhov and Tyutin 1994).
Parasite species
Lake type
pH 6.6 - 7.5 pH 6.0 - 6.5
pH 4.1-5.5
Protozoa
Glugea luciopercae
Hennegitya creplini
Hemiophrys branchiarum
Capriniana piscium
Apiosoma campanulatum
A. minimicronudeatum
Trichodina urinaria
Trichodinella epizootica
Monogenea
Ancyrocephalus percae
Gyrodactylus cernuae
Cestoda
Triaenophorus nodulosus
Proteocephalus percae
Trematoda
Bunodera luciopercae
Ichthyocotylurus variegatus
Ich. pileatus
Nematoda
Eustrongylides tubifex
Camallanus lacustris
Acanthocephala
Acanthocephalus lucii
Crustacea
Argulusfoliaceus
Total number of species
0
0
+
+
+ ;
+ ,
19
0
7
0
0
0
0
+
0
0
0
0
0
0
0
0
4
Note: "0" - absence of the parasite species, "+" - presence of the parasite species, "-" -data are not available.
-------
Copepodae serve as intermediate hosts for P. percae, a common component of
zooplankton of acid lakes. At the same time the eggs of these cestoda have a thick shell
envelope reliably defending the embryo while they are living in water. Thus, the species of
parasite composition and the infection of the perch with P. percae and C. lacustris can be used
as parasitological indication of the level of lake acidification (Zhokhov and Tyutin 1994).
The following investigations of acid lakes were carried out by us (together with Victoria
Matey) on 13 Karelian lakes in 1991. Only two of them were circumneutral, the others were
acidic with pH from 5.8 to 4.5, but their color was different: all lakes from the Suoyarvi system
were brown or humic, and from Kondopoda they were clear. Using SEM we have investigated
the gills of 200 perch from all these lakes and discovered that they are infected with two species
of Peritricha: Apiosoma campanulatum and A. minimicronucleatum. The presence of A.
minimicronudeatum has been reported for the gills of perch from acid lakes of West Siberia
(Razmashkin 1988) and the Vologda region (Zhokhov and Tyutin 1994), but A. campanulatum
were not reported.
The level of infestation and distribution of parasitic Peritrieha in the acid clear water
lakes and colored water lakes were different (Table 2). In clear water lakes the infestation of fish
gills with parasitic Peritricha was extremely low (Figure 2a). In the neutral lake and weak-
acidified lake the gills were infected with a single species (A. campanulatum). As for perch from
lakes with pH lower 5.9, their gills were absolutely free from Peritricha. In the colored acid
lakes we saw another situation. In the less acidic lake the perch gills were infected with a single
species (A. campanulatum) at a moderate level of infestation. In the lakes with water pH from
5.5 to 4.9 we noted an increasing number of A. campanulatum, and the appearance of
A. minimicronucleatum (Figure 2 b-d). A. campanulatum localized on gill filaments, while
A. minimicronucleatum preferred respiratory lamella. But when water pH decreased below 4.9,
A. campanulatum disappeared and fish gills were infected only withal, minimicronucleatum.
Table 2. Infestation of perch from Karelian lakes with parasitic Peritricha.
Lakes
pH
Apiosoma
minimicronucleatum
Apiosoma
campanulatum
Colored lakes
Suoyarvi
Kabozero
Vegarusyarvi
Lcukunyarvi
IlyukaJhenyarvi
Vuontolenyarvi
Lamba Vegarous
Sargozero
Venderskoye
Urosozero
Chuchyarvi
Blue lamba
Grushina lamba
5.8
5.5
5:1
4.9
4.6
4.6
4.5
7.9
7.0
5.9
5.0
4.6
4.6
Clear water lakes
+
-f-
56
-------
Figure 2. Localization of peritrichs Apiosoma campanulatum and A. minimicronucleatum
on the gills of perch from clear-water (a) and colored (b,c,d) Karelian lakes, SEM.
(a) A. campanulatum (pH 5.9); (b) A. campanulatum on the primary lamellae of gills (pH 5.8);
(c)A. campanulatum and A. minimicronucleatum (pH 5.1); (d)A. minimicronucleatum
(pH4.6). Scale bar-100 |im
57
-------
Figure 3. infrastructure and localization alApiosoma campanulatum and A. minimicronudeatum
on the gills of the perch from colored acid Karelian lakes, SEM. (a) ultrastructure of
A. campanulatum; (b) ultrastructure of A. minimicronudeatum; (c) localization of
A. minimicronudeatum on the basement primary lamellae of gills (pH 4.5);
(d) A. minimicronudeatum on the secondary lamellae of perch gills (pH 4.5).; Scale bar:
(a,b) 10 jim; (c,d) 100 [im
58
-------
The level of infestation was extremely high in perch gills from the most acidic colored lake
(pH 4.5). A. minimicronucleatum covered gill petals very tightly: respiratory lamella and
filaments (Figure 3). There is a definite correlation between the level of infestation offish gills
with parasitic Peritricha and the ecological situation in the water bodies. The most tolerant to
acidification, A. minimicronucleatum, can be used as indicator of water acidification.
Industrial Pollution
The proposed method for biotesting of a fish population is based on the phenomenon of
symmetry of bilateral structures. This feature reflects the morphogenesis stability offish
organisms. At the same time the percent of the asymmetric specimens in a population under
anthropogenous impacts increases noticeably. It has been shown that alteration offish
morphogenesis stability from the polluted areas are correlated with asymmetry in the invasion of
fish gills by monogenean of genus Dactylogyrus (Zharikova 1993). For each fish examined, the
presence of parasites was determined on every gill arch both on the right and the left sides. The
number of gill rackers of bream has been considered as one of the morphological features that
have taxonomic significance.
The ratio of symmetric and asymmetric specimens of common bream infected by
Dactylogyrus varied in distinct parts of the Gorky, Kuibyshev, and Rybinsk Reservoirs
(the Volga River system) and other waterbodies with different degrees of pollution (Table 3).
The high percentage of the asymmetrically infected breams has been revealed in the water bodies
under anthropogenic pollution: the region of warm waters of the Kostroma hydro-power station,
Gorky Reservoir; the Kama River, Kuibyshev Reservoir; and the Sheksna part of the Rybinsk
Reservoir.
Table 3. Asymmetry of bream infection with monogeneans Dactylogyrus spp. in some part of
reservoirs with antropogenous pollution (Zharikova 1993).
Reservoir Total number of fishes Groups of fishes
Gorky Reservoir, region of warm waters
Kuibyshev Reservoir, Kama part
Rybinsk Reservoir j Sheksna part
61
34
45
Symmetrical
26 (42.9%)
i 13 (38.2%)
17 (37.8%)
Asymmetrical
35 (57.1%)
21 (61.8%)
28 (62.2%)
Note in brackets - percent of total number
On the other hand, the low percentage of the asymmetrically infected breams has been
found in the relatively clean water bodies, such as the Danube delta, the river part and the
Kostroma widening of the Gorky Reservoir, and the Volga part of the Rybinsk Reservoir
(Table 4). According to the ratio of the symmetrically and asymmetrically infected specimens
we can predict the state of the environment which the fish inhabits (Zharikova 1993). This
method allows for the determination of changes in the physiological state of a fish population
and consequently changes in the ecological state of the waterbody.
59 !
-------
Table 4, Asymmetry of bream infection with monogeneatis Dactylo^fus spp, in relatively
clean waters (Zharikova 1993).
Reservoir
Delta of Danube River
Gorky Reservoir, river's part
Gorky Reservdlr, Kostroma Widening
Rybinsk Reservoir, Volga part
Total number of fishes
23
64
60
66
Group of Wsjies
Symmetrical Asymmetrical
19 (82.6%) 4 (17.4%)
48(75.0%) 16(23.0%)
42(70.0%) 18(30.0%)
45(68,2%) 21(31.8%)
Note in brackets - percent of total number
The next approach is associated with the search of some parasite species resistant to
complex environmental pollutants. We have studied the action of industrial wastes of
Chcrepovetz metallurgical works on fish parasites from the Sheksna part of the kybinsk
Reservoir in 1988-1991, The wastes contained PCBs, J*AMs? oil-products, and heavy metals
(Pb» Cd, Cu, Cr) (Flerovl990). For the sake of comparison fish parasites from the other,
relatively clean, parts of the Rybinsk Reservoir, the Volga and Mologa parts were examined,
Endo- and ectoparasites of common bream from various parts of the Rybinsk Reservoir Were
examined. There were 120 specimens from the Sheksna part, 106 specimens from the Volga
part, and 196 specimens from the Mologa part.
Influence of the pollution has initially been studied on parasites of common bream from
the Sheksna part of the Rybinsk Reservoir. The number of ectoparasites on the bream gills
Protozoa, crustaceans Ergasilus sieboldl, leeches Caspiobdeliafadejewi, monogenean genus
Gyrodactylw, and Dactylogynts noticeably decreased. At the same time the monogettean
Dlplozoon pctmdoxum has a high resistance to the water pollution, its number increased and
was higher in the Sheksna part when compared with the relatively clean Volga and Mologa parts
of the reservoir. The specimens of bream most infected by D.paradoxum were found in
which is the closest to Cherepovetz and the most polluted area (table 5).
Table 5. Occurence tfDipiozoonpamdomm mAbrantis brattia from kybinsk Reservoir under
antropogettous effect (Kuperman 1992).
Study area
Sheksna part1
ToroVd1
Volga part2
Mologapart2
Number of fish
77
24
64
66
Prevalence (%)
79.2±16.6
4§.4il2,5
13.6±8.4
Mean intensity inf. fish
4.3
5.S
: 3,2 .'
3,4
Range
1-19
1-19
h 12
1-8
Note: ' - polluted area;2 - relatively clean area
-------
D
Figure 4. Structural anomalies of the clamps on the opisthoptor of Diplozoon paradoxum in bream
from the polluted area of the Sheksna part of Rybinsk reservoir, SEM. (a) normal structure,
correlation of clamps is 4:4; (b-d) asymmetric localization and reduction of clamps of
D.paradoxum fronrthe polluted area: (b) correlation:of clamps is 4:3; (c) correlation of
clamps is 3:3; (d) correlation of clamps is 4:1. Scale bar -10 jam.
61
-------
Despite the high resistance of those parasites they, as organisms, undergo a direct effect
of toxicants at the population level. We have observed for the first time structural anomalies of
£>. paradoxum in the polluted zone of the Sheksna part of the Rybinsk Reservoir. Abnormal
forms were noticed with reduced clamps on the opisthaptor, with disturbed symmetry and
number (Figure 4). Normal arrangement for Diplozoon clamps is 4:4. We have seen specimens
with 4:3, 4:1,4:0, and 3:3 clamps. About 12% of D.paradoxum had abnormal opisthaptor
structures. Structural anomalies are probably a result of the mutagenous effect of toxic water in
polluted zones. In relatively clean zones of the reservoir (the Volga and Mologa parts) such
anomalies were not observed.
The intensity of infection of common bream by the endoparasite cestode Caryophyllaeus
laticeps in the polluted zone (Sheksna part) has greatly increased when compared with the Volga
and Mologa parts (Table 6). This is related to toxicoresistant oligochaetae intermediate hosts of
Caryophyllidae, which became the main food for common bream in this area. Biomass of
benthos in the Sheksna part of the Rybinsk Reservoir in this period is mainly represented by
oligochaetae, whereas the number of Chironomideae, more sensitive to the pollution, sufficiently
decreases (Skalskaya 1990). Previously the increasing in intensity of the infestation of common
bream by C. laticeps has been observed in the Kama River under anthropogenous pollution
(Kostarev 1980).
Table 6. Occurence of Caryophyllaeus laticeps mAbramis brama from Rybinsk Reservoir
under antropogenous effect (Kuperman 1992).
Study area
Sheksna part1
Volga part2
Mologa part2
Number of fish
43
42
130
Prevalence (%)
90.7±10.4
78.6±11.1
91.8±8.2
Mean intensity inf. fish
56.5
32.0
25.5
Range
2-246
1-170
1-311
Note: - polluted area; - relatively clean area
On the basis of the data obtained by us and from the literature (Khan and Thulin 1991,
Overstreet 1993, MacKenzie et al. 1995) we assume that a number of parasite species can serve
as reliable indicators of anthropogenous pollution of water bodies. Other species of parasites
more highly resistant to toxic impact can be the most objective bioindicators of authropogenous
pollution and ecological and sanitary state of water bodies.
ACKNOWLEDGEMENTS
I thank Tatyana Zharikova, Alexandr Zhokov, and Andrey Tyutin from the Institute of
Biology of Inland Waters, Russian Academy of Sciences, for the helpful discussion of
presentation of data.
62
-------
REFERENCES
Anikieva, L. A. 1982. Effect of wastes of Segezha paper-pulp plant on the helminth fauna of
fishes from Vygozero. Ecologia Parasiticheskikh Organismov v Biogeocenozach Severa.
Petrozavodsk, 83-84 (In Russian). :
Bogdanova, E.A. 1988. Ecological characteristics offish parasitofauna in different areas of
Ladoga Lake in conditions of reconstruction of its biocenosis. In: Influence of Pollution on
the Ecosystem of Ladoga Lake, 285:33-43 (In Russian).
Flerov, B.A. 1990. Ecological situation in the Rybinsk Reservoir as the result of the accident in
Cherepovetz purification of sewage building. In: Vlijanie Stokov Cherepovetzkogo
Promishlennogo Uzla na Ecologicheskoye Sostojanie Rybinskogo Vodochranilischa.
Rybinsk, pp 3-11 (In Russian).
Flerov, B.A., V.R. Mikryakov, and B.I. Kuperman. 1982. Invasive and infective processes in
fish by toxical exposure. In: Helminths in Freshwater Biocenosis. Moscow, pp 58-67. (In
Russian).
Khan, R.A., and J. Thulin. 1991. Influence of pollution on parasites of aquatic animals.
Advances in Parasitology, Academic Press., 30:201-238.
Kostarev, W.F. 1980. Effect of pollution on the dynamics offish parasite fauna of Kama
reservoirs. In: Biologicheskie Resursy Vodoemov Zapadnogo Urala, Perm, pp 147-153 (In
Russian). '
Kuperman, B.I. 1992. Parasites as bioindicators of the pollution of water bodies. Parasitologiya
26:479-482 (In Russian).
Kuperman, B.I., I. Y. Kolesnikova, and A.V. Tyutin. 1994.. Ambiphrya ameiuri (Ciliophora:
Peritricha): the ultrastructure and localisation on the body of young carp fishes. Parasitologia
28:214-221 (In Russian).
Laimann, E.M. 1957. Effect of the wastes of plants on the infection of Kama fishes. Trudy
Moskovskogo Technicheskogo Instituta Rybnoy Promyshlennosti i Chozyastwa, 8:235-240
(In Russian).
MacKenzie, K., H.H. Williams, B. Williams, A.H. McVicar, and R. Siddal. 1995. Parasites as
indicators of water quality and the potential use of helminth transmission in marine pollution
studies. Advances in Parasitology, Volume 35, pp 85-144.
Overstreet, R.M. 1993. Parasitic diseases of fishes and their relationship with toxicant and other
. environmental factors. In: Pathobiology of Marine and Estuarine Organims, CRC Press, Inc.,
Florida, pp 111-156. :
63
-------
Razmashkin, D.A. 1988. Dependence of parasitofauna offish from West Siberia lake farms on
abiotic factors. In: Ecologo-populationary Analisis of Parasite-Host Relationship.
Petrozavodsk; pp 36-56.
Skalskaya, LA. 1990. Stress of communities of zooperiphiton in Rybinsk reservoir. In: Vlijanie
stokov Cherepovetzkogo Promishlennogo Uzla na Ecologicheskoye Sostojanie Rybinskogo
Vodochranilischa, pp 59-72 (In Russian).
Solomatova, V.P., and A.V. Lussin. 1981. Parasites and diseases oflctalurus in the farm
conditions on the wastes of ORES. Sbornic Nauchnykh Trudov GOSNIORKH, 150: 157-
174 (In Russian).
Strelbitzkaya, I.N. 1986. Ambiphrya from one year old Salmonidae. Sbornic Trudov
GOSNIORKH, 247:111-114 (In Russian).
Zharikova, T.I. 1993. Influence of anthropogenic pollution of water bodies on the ectoparasites
of bream, Abramis brama. Zoologicesky Zhurnal, 72:53-58 (In Russian).
Zhokhov, A.E., and A.V. Tyutin. 1994. Parasitofauna of fish under acidification of lakes. In:
Structure and Functioning of Ecosystems in Acid Lakes. Saint Peterburg, Nauka, pp 186-201
(In Russian).
64
-------
WHIRLING DISEASE OF SALMONIDS IN RUSSIA AND EUROPE:
HISTORY, DISTRIBUTION, AND CONTROL
i
Boris I. Kuperman1 and Solomon S. Schul'man2
ABSTRACT
Whirling disease is a dangerous protozoan disease of salmonids caused by Myxobolus
(Myxosoma cerebralis). The transcontinental distribution of whirling disease has great economic
importance for salmonid fish hatcheries. The first outbreak of whirling disease was reported
from Europe (Germany) in rainbow trout (Oncorhychus mykiss) hi 1903. During the 1950s
reports appeared on the spreading of whirling disease in trout hatcheries of other European
countries including France, Italy, Poland, Czechoslovakia, USSR, Bulgaria and Denmark. The
current geographical distribution of M. cerebralis demonstrates its presence on all continents:
the USA in North America, Ecuador and Mexico in South America, South Africa and Morocco
in Africa, and in Australia and New Zealand. In the USSR whirling disease was detected in
1952-1954 in trout farms in the Leningrad region and in trout farms of North Caucasus and
Stavropol regions. Spores of M cerebralis were also found in both hatchery and wild salmonids
from nearby water-bodies on the Sakhalin (Far East) and Kola Peninsula (Polar region).
At present, spores of M cerebralis have not been identified in wild or hatchery salmonids from
the Far East. M. cerebralis occurrence has been recorded in 17 species of the genera Salmo,
Oncorhynchus, and Salvelinus. It has been experimentally demonstrated that the life cycle of
M. cerebralis passes with the participating of tubificid oligochaetes. In the intestinal epithelium
of these oligochaetes spores of M cerebralis transform into^spores of Actinosporea, genus
Triactinomyxon. The process of this transformation has been investigated using light and
electron microscopy.
HISTORY AND DISTRIBUTION
Whirling disease is caused by Myxosoma cerebralis, and is one of the most dangerous
protozoan diseases of salmonids. The transcontinental distribution of M cerebralis has great
economic importance simultaneously with the increased development of salmonid fish farms. In
spite of the obvious economic importance of M cerebralis, information on some questions of its
biology is scarce.
'Institute of Biology of Inland Waters, Russian Academy of Sciences, Borok, Yaroslavl, Russia.
2Institute of Ecology of the Volga River Basin, Russian Academy of Sciences, Toliatty, Russia.
65
-------
The first outbreak of whirling disease was reported from Germany by Hofer (1903) in
rainbow trout (Oncorhynchus myMss). It was examined in more detail by Plehn (1904, 1924).
Whirling disease epizooty in Germany of the 1920s and 30s acquired such serious forms that it
was necessary to liquidate completely these trout husbandries (Schaperclaus 1931). After the
Second World War the importance of this disease did not diminish (Heuschmann 1949, Luling
1952, Schaperclaus 1954). At the same time, in Germany they succeeded in elaborating a full
system of preventive and control measures against whirling disease (Tack 1951, Schaperclaus
1954). In the 1950s and 60s reports appeared on the wide spreading of whirling disease in trout
husbandries in other countries of western and eastern Europe (including the USSR): in France
(Vanco 1952), Poland (Kocylowcki 1953), Italy (Scolaria 1954), Czechoslovakia (Dyk 1954,
Volf 1957, Havelka and Volf 1970, Lucky 1970), Bulgaria (Margaritov 1960), Denmark
(Christensen 1966, Rasmussen 1967), and Scotland (Elson 1969).
In the former USSR whirling disease was detected for the first time in 1952 in the
collective trout farm "Ropscha" in the Leningrad region (Uspenskaya 1955,1957). The disease
was very severe with significant mortality for young trouts. Soon whirling disease was revealed
in some other trout farms hi different regions of the USSR, namely the north Caucasus and
Stavropol regions. In addition, M. cerebralis was also found in salmonids in nature near fish
hatcheries on the Sakhalin Island (Far East) in 1959, and on the Kola Peninsula (Polar region) in
1955 (Bogdanova 1960,1970). At present, spores of M cerebralis are not found in salmonids
from the Sakhalin Island, and other hatcheries and natural waters of the Far East.
As a result of serious measures, whirling disease was liquidated hi the Ropsha trout farm
and in the North Caucasus region. The further spreading of whirling disease has been prevented
in other regions of the USSR. Since the 1970s whirling disease among salmonids in the USSR
has not been found. In the U.S., M. cerebralis was detected in 1956 (Hoffman et al. 1962,
Hoffman and Putz 1970). Whirling disease was widely spreading in many states in the U.S. as a
result of numerous transfers of rainbow trout. Presently whirling disease has been reported in the
wild hi 11 states and in 9 additional states in hatcheries. M. cerebralis has not been observed in
numberous wild and cultured salmonids examined in Canada (British Columbia) from 1968 to
1980 (Margolis et al. 1981). The occurrence of M cerebralis has been reported hi New Zealand
(Hewitt and Little 1972).
THE HOST RANGE OF M. CEREBRALIS
M. cerebralis has been mainly identified in farm-reared salmonids. Table 1 illustrates
the known host range of M cerebralis in both farms and natural populations of salmonids
(Uspenskaya 1957, Hoffman 1970, Havelka and Volf 1970, Bogdanova 1970, Christensen 1972).
and others (Halliday 1976). M. cerebralis has been reported hi 17 species of the genera Salmo,
Oncorhynchus and Salvelinus. Bogdanova (1970) stated that some salmonids participated hi the
distribution of the parasite La nature, but that non-anadromous forms were the most important
hosts. The infection took place only in freshwater but the intensity of invasion was low and no
disease symptoms were observed. As to the susceptibility of salmonids, rainbow trout are most
affected, brook trout (Salvelinus fontinalis) less severely, and brown trout seem most resistant to
clinical infection (Hoffman etal. 1962, O'Grodnik 1979).
66
-------
Table 1. The occurrence of M. cerebralis in salmonids in hatcheries and natural waters
(from Halliday 1976).
Europe and Asia and Far East
America
Species
Natural waters
Hatcheries
^Natural,waters Hatcheries
Salmo clarki
S. ischchanisp. aestivalis
S. ischchan isp. gegerkuni
S. salar
S. trutta
S. trutta labrax
Salvelinus fontinalis
S. leucomasnis
S. malma
S. namaychush
Oncorhynchus gorbushcha
O.keta
O. kisutch
O. masu
O. mykiss
O. nerka
O. tschawytscha
+-H-
+++
+
+ M. cerebralis present; +++ Whirling disease recorded;- 0 Experimental infection; - Fish not examined.
LIFE CYCLE OF M. CEKEBRALIS
Investigations of M cerebralis began about 90 years ago, but its life cycle has not been
completely clarified and it is still being widely discussed. Up until the 1980s the generally
accepted point of view was that the life cycle of M cerebralis mainly passed within the body of
one host fish. It was thought that fish are infected by ingesting invasive spores of M cerebralis
dispersed in water. However experiments designed to infect rainbow trout fry were unsuccessful
or inconclusive. ;
Let us examine the life cycle of M cerebralis in detail. The developmental process of
M. cerebralis in fish includes two stages: an active vegetative stage amoeboid throphozoite, and
the stage of spores. The spores of M cerebralis are of spherical shape and contain at the anterior
end two polar capsules with polar filaments within each of them. Besides the polar capsules
there is sporoplasma. Under the effect of the host's digestive enzymes the polar filament is
extruded, and the spores are anchored to the intestinal wall. The valves open and liberate the
sporoplasm. After release, the sporoplasm passes between the cells of the intestinal tract and
reaches the cartilage by means of the blood, lymph, or coelomic fluid. In the cartilage it
develops into a multinuclear amoeboid trophozoite. Being established, the trophozoite grows by
nuclear division and cytoplasmic growth. It may reach a maximum of 1 mm in diameter
(Hoffman et al. 1962) with 50 or more nuclei (Schul'man 1966). The division of the nuclei is
67
-------
accompanied by plasmotomy in M. cerebralis like in other species of Myxosporidea (Noble
1944). Within the trophozoite are formed pansporoblasts, which contain 12-14 nuclei (Hoffman
-------
two stages were observed. The first corresponds to the amoeboid developmental stage of this
parasite, during which destruction of the cartilage surrounding the auditory organ and spinal
column determine the main symptoms: (1) Disturbing of movement coordination caused by
erosion of cartilaginous part of the auditory organ; (2) Blackening of the caudal part of the fish.
caused by infection of the cartilage of the spinal column in the region of the 25th vertebra; it
exerts pressure on the caudal nerves, which control the pigment cells in the tail; and (3)
Destruction of the cartilage leads also to deformities of the spinal column, cranial deformities
and deformities of the jaws and gill opercula (Hoffman et al. 1962).
The first stage of disease is more dangerous and often associated with severe exhaustion,
mahiutrition and death. This stage of the disease usually lasts 3-4 months. The duration of the
incubation period is between 12-60 days. The second stage of the disease begins when the
parasite is starting spore formation. The external symptoms of the disease disappear. During
this period surviving fish become the agent to spread whirling disease. The process of spore
formation coincides with ossification of the auditory capsule alleviating the effects of the
parasites. Intensity of the disease depends on water temperature. At higher temperatures
(20-25 C) M. cerebralis develops faster and the disease proceeds less severely than at low
temperatures (7-15 C) (Schaperclaus 1931). The ossification of the spinal column and the
intensive growth of trout proceed faster at higher temperature. Histopathology of M cerebralis
depends on three factors: (1) the age of the host at first infection (Hoffman et al. 1962); (2) the
number of infective units (Putz 1969); and (3) the ambient water temperatures (Halliday 1976).
DIAGNOSIS AND CLASSIFICATION
The most reliable diagnosis is assessed by direct revealing of M cerebralis spores. In
general, the technique is to concentrate the spores by homogenizing the skeleton and centrifuging
the homogenate with or without prior filtration. The preparation can then be stained by Giemsa,
malachite of methylene blue to aid recognition of the spores. Also, the spores can be embedded
into glycerin-gelatin preparation and post examined with th.0 phase-contrast method (Donee and
Schul'man 1973). Another technique used to diagnosis whirling disease is an immunological
method.
The classification of the Myxosporidea relies entirely on features of the spore,, and those
of M cerebralis have been characterized by Schul'man (1966), Lorn and Hoffman (1971), and
Hedrick et al. (1991). Spores are usually oval and bivalved and have a length of 9.7 (J, and a
breadth of 8.5 ja. The valves are connected with a well-developed suture ridge. The spores have
two polar capsules and an amoeboid embryo. Polar capsules have a length of 4.2 |i and a breadth
of 3.1 [i. The capsule filaments have a length of 5-6 jx. The valves are characterized by high
stability defending the parasites from pressure and drying up. According to Schaperclaus (1931),
the spores M. cerebralis can preserve life ability and invasiveness for about 12 years on the
bottom of a dry pond.
69
-------
CONTROL AND TREATMENT
To control and treat whirling disease in hatcheries the following measures have been
used: (1) full quarantine of the fish farming where the disease was detected; (2) drying and
disinfection of the ponds; (3) separate rearing of young and aged fish; (4) limiting of fish
transfer from one fish farm to another; (5) water treatment; (6) selection and destruction of
trout having signs of recovery after whirling disease (Bauer 1959; Rasmussen 1967; Bogdanova
1968); some investigators recommend destroying of all fish which have been exposed to the
infection (Schaperclaus 1954; Hoffman etal. 1962); (7) prevention of wild brook trout invading
ponds offish farming; and (8) growing offish larvae in apparatuses with exclusively uninfected
water (preferably brook water). After removing the infected fish, holding facilities can be
disinfected by any of several methods described by Schaperclaus (1931), Tack (1951), Scolaria
(1954), Rasmussen (1967), Putz (1969) and Hoffman (1970).
PREVENTION
Prevention of whirling disease is based on the suggestion that when a fish reaches a
length of 5-6 cm (4-5 months of age) it is generally resistant to infection because ossification of
the skeleton will prevent massive infection. The following method is used successfully in
Europe (Christensen 1966, Rasmussen 1967, Ghittino 1970). Eggs are hatched in spore-free
water. After hatching, the fish are reared either in tanks or in raceways. These replace the
earthen ponds which were a source of infection. The fish are placed in earthen ponds only when
they are considered to be clinically resistant to the parasite. The success of this technique in
Denmark has been assessed by Halliday (1974).
Several types of water filtration systems have been developed to eradicate the problem of
whirling disease. Sand-charcoal filters have met with success in France (Hoffman et al. 1962).
Recently a method of irradiating hatchery water supplies with ultraviolet light has been
developed. This treatment in addition to destroying other fish pathogens kill the spores
Myxosporidea (Anonymous 1969). Units of UV at dosages of 35,000,43,000 and 112,000
MWS/cm prevented infection.
ACKNOWLEDGEMENTS
1 thank Oleg Yunchis for useful discussion and a personal communication on recent data
regarding Myxobolm ccrebralis from salmonids in the Far East (Sakhalin Island).
70
-------
REFERENCES ,
Anonymous. 1969. Whirling disease. U.N. Food and Agricultural Organization, Aquaculture
Bulletin 1:10.
Bauer, O. N. 1959. Parasites of freshwater fish and the biological basis for their control.
Bulletin of the State Scientific Research Institute of Lake and River Fishery 49. Israel
Program for Scientific Translations, Jerusalem 1962. 236 pp.
Bogdanova, E. A. 1960. An endemic disease of salmonids in Sakhalin Hand. Doklady
Academii Nauk. SSSR. 134:1501-1503. (Translated in AIBS).
Bogdanova, E. A. 1968. Modern data on the distribution and biology of Myxosoma cerebralis
(Protozoa, Cnidosporidia) as agent of whirling disease of salmonids. Bulletin of the Office
of International Epizootics 69:1499-1506. ;
Bogdanova, E. A. 1970. On the occurrence of whirling disease of salmonids in nature in the
USSR. Second International Congress of Parasitology, Abstract #719, September 6-12,
Washington, D.C. Journal of Parasitology 56:399.
Christensen, N.O. 1966. Fiskesygdomme. Copenhagen. 69 pp. (In Danish).
Christensen, N. O. 1972. Some diseases of trout in Denmark. Symposia of the Zoological
Society of London, 30:83-88.
Donee, Z. S., and S. S. SchuFman. 1973. On the methods of investigation of Myxosporidia
(Protozoa, Cnidosporidia). Parasitologya 7:191-193. (In Russian).
Dyk, V. 1954. Nemoci u Nasich Ryb. Praha, Czechoslovakia. 392 pp. (In Czechoslovakian)
El-Matbouli, M., and R. W. Hoffman. 1989. Experimental transmission of two Myxobolus spp.
developing bisporogeny via tubificid worms. Parasitology Research 75:461-464.
El-Matbouli, M., R. W. Hoffman, and C. Mandok. 1995. Light and electron microscopic
observations on the route of the triactinomyxon-sporoplasm of Myxobolus cerebralis from
epidermis into rainbow trout cartilage. Journal of Fish Biology 46:919-93 5.
Elson, K.G.R. 1969. Whirling disease in trout. Nature 223,968.
Ghittino, P. 1970. Present status of whirling disease in Italian trout farms. Rivista Italiana
Piscicola Ittiopatology 5:89-92.
Halliday, M.M. 1974. Studies on Myxosoma cerebralis, a parasite of salmonids. III. Some
studies on the epidemiology of Myxosoma cerebralis hi Denmark, Scotland, and Ireland.
Nordisk Veterinaermedecin 26:165-172.
71
-------
Halliday, M. M. 1976. The biology of Myxosoma cerebralis: the causative organism, of whirling
disease of salmonids. Journal of Fish Biology 9:339-357.
Havelka, J., and F. Volf. 1970. Whirling disease in New Zealand trout caused by Mixosoma
cerebralis in Czechoslovakia. Abstract No. 253. Journal of Parasitology 56. Proceedings of
the 2nd International Congress of Parasitology, September 6-12,1970, Washington D.C.
Journal of Parasitology 56:137-138.
Hedrick, R.P., A. Wishkovsky, J.C. Modin, and R.J. Toth. 1991. Three Myxosporeans found in
the cranial and branchial tissues of rainbow trout in California. Journal of Aquatic Animal
Health 3:55-62.
Heusehmann, O. 1949. Die Drehkrankheiten der Salmoniden. Allgemaine Fishereizeitung
74:216-217. (In German).
Hewitt, G. C., and R. W. Little. 1972. Whirling disease hi New Zealand trout caused by
Mixosoma cerebralis .(Hofer 1903) (Protozoa: Myxosporidia). New Zealand Journal of
Marine and Freshwater Research 6:1-10.
Hoffer, B. 1903. Uber die Drehkrankheit der Regenbogenforelle. Allgemaine Fischereizeitung
28:7-8. (In German).
Hoffman, G.L. 1970. International control of parasitic diseases of fishes. Abstract No. 277,
Second International Congress of Parasitology, September 6-12,1970, Washington, D.C.
Journal of Parasitology 56:151-152.
Hoffman, G. L., C. E. Dunbar, and A. Bradford. 1962. Whirling disease of trouts caused by
Myxozoma cerebralis in the United States. U.S. Fish and Wildlife Service, Special Scientific
Report-Fisheries No. 427. 14 pp.
Hoffman, G. L., and R. E. Putz. 1970. Host susceptibility and effect of aging, freezing, heat and
chemicals on spores of Myxosoma cerebralis. Progressive Fish-Culturist 31: 35-37.
Kocylowski, B. 1953. Choroby nyb I ich knalkanie. Gespodarka ryb 5:24-26. (In Polish).
Lom, J., and G.L. Hoffman. 1971. Morphology of the spores of Myxosoma cerebralis (Hofer,
1903) andM cartilaginous (Hoffman, Putz, and'Dunbar, 1965). Journal of Parasitology
57:1302-1308.
Lucky, Z. 1970. Parasitological changes and diagnostics of Myxosomosis of the rainbow trout
(Salmo gairdneri irideits). Acta Veterinarra Brno 39:19-29.
Luling, K.H. 1952. Die Emeger der Knotchen-Baulen and Drehkrankheit unserer Fische.
Osterreichs Fischerei 5:171-175. (In German).
72
-------
Margaritov, N.M. 1960. Whirling disease of trout in DRS-Samokov. Rubno Stopanstvo 2:15-
18, Sofia. ;
Margolis, L., T.E. McDonald, and G.E. Hoskins. 1981. Absence of the protozoan Myxosoma
cerebralis (Myxosoma: Myxosporea), the cause of whirling disease, in survey of salmonids
from British Columbia. Canadian Journal of Fisheries and Aquatic Sciences 38:996-998.
Markiw, M.E., and K. Wolf. 1983. Myxosoma cerebralis (Myxozoa:Myxosporea) Etiological
agent of salmonid whirling disease requires tubificid worm (Annelida: Oligochaeta) in its life
cycle. Journal of Protozoology 30: 561-564.
Noble, E.A. 1944. Life cycles in the Myxosporidia. Quarterly Review of Biology 19:213-235.
O'Grodnik, J. 1979. Susceptibility of various salmonids to whirling disease (Myxosoma
cerebralis). Transactions of the American Fisheries Society 108:187-190.
Plehn, M. 1904. Uber die Drehkrankheit der Salmoniden Lentospora cerebralis. Archive fur
Prolistenkunde5:145-166. (In German).
Plehn, M. 1924. Praktikum der Fischkrankheiten. Stuttgard, Schweizerbartsche, 479 pp.
Putz, R.E. 1969. Experimental transmission of Myxosoma cerebralis (whirling disease) and
effect of freezing on the spores. Progressive Sport Fish Research 1969:55-57.
I
Rasmussen, C.J. 1967. Handbog I Orredopodrate. Copenhagen: Rhodos. 242 pp.
Schaperclaus, W. 1931. Die Drehkrankheit in der Forellenzucht und ihre. Bekampfung.
Zeitschrift fur Fischenkunde 29: 521-567. (In German).
Schaperclaus, W. 1954. Fischkrankheiten. Academic-Verlag, Berlin, Germany, 708 pp. (In
German).
Schul'man, S.S. 1966. The Myxosporidian Fauna of the U.S.S.R. Nauka, Moskow and
Leningrad, 504 pp. (In Russian).
Scolaria, C. 1954. Sull'impiego dello stovarsolo nelle profilassi del "capostan" o
"lentosporiasi" delle trote d'allevamento. La Clinica Veterinaria 77:50-53. (In Italian).
Tack, E. 1951. Bekampfung der Drehkrankheir mit Kalkstockstoff. Fishwirtschaft 1:123-130.
(In German).
Uspenskaya, A.V. 1955. Biology, distribution and economic importance of Myxosoma
cerebralis, the causative agent of twist disease in trout. Doklady Akademii Nauk SSSR
105:1132-1135. (In Russian).
73
-------
Uspenskaya, A.V. 1957. The ecology and spreading of the pathogen of trout whirling disease-
Myxosoma cerebralis (Hofer 1903, Plehn 1905) in the fish ponds of the Soviet Union. All-
Union Research Institute of Lake and River Fishery 42:47-55. (In Russian).
Vanco, R. 1952. Contribution a L'etude de la Pathologic des Alevins de Truitte. Paris, 92 pp.
Volf, G. 1957. The first case of whirling disease among our salmonid fish in our hatcheries.
Zivocisna Vyroba 30:425-428.
Wolf. K., and M.E. Markiw. 1984. Biology contravenes taxonomy in the Myxozoa: new
discoveries show alternation of invertebrate and vertebrate hosts. Science 225:1449-1452.
74
-------
WHIRLING DISEASE: THE NORTH AMERICAN EXPERIENCE
Elizabeth MacConnell1 and Charlie E. Smith1
ABSTRACT
Based on circumstantial evidence, whirling disease was introduced into North America in
frozen, processed fish imported from Denmark to Pennsylvania and Nevada in 1956. Whirling
disease was spread from these two original sources by the transfer of live fish and the movement
of infected fish in streams. The first confirmed case of whirling disease was in a Pennsylvania
hatchery in 1958, and since then a total of 20 states have reported confirmed cases of whirling
disease. Until recently, Myxobolus cerebralis, the causative agent of whirling disease, was
classified as a prohibitive pathogen. Diagnosis of M. cerebralis in a hatchery meant that
depopulation and disinfection of the entire facility would be required. In 1984 Wolf and Markiw
published a revolutionary discovery regarding the life cycle 'of this parasite. Their research
demonstrated that M. cerebralis has a two host life cycle that involves a fish and an aquatic
oligochaete. Limited research has been conducted in this country since. Experiences throughout
the U.S. have shown that M. cerebralis does not cause severe disease in hatchery fish, however,
the impact of whirling disease on wild fish populations is unknown. Recent population declines
of wild trout associated with whirling disease in Colorado and Montana have generated a great
deal of controversy and renewed interest in research on this disease.
HISTORY
Whirling disease of salmonid fish is caused by the myxosporean parasite, M. cerebralis.
The parasite has a broad specificity for salmonids with rainbow trout being highly susceptible to
infection (Markiw 1992). Unlike most myxosporeans, M. cerebralis can produce a severe host
reaction causing serious disease.
The disease was first described in rainbow trout (Oncorhynchus mykiss) in Europe
(Hofer 1903). It is widely believed that M. cerebralis evolved as a nonpathogenic parasite of
brown trout (Salmo trutta) in central Europe and northern Asia (Hoffman 1970).
'U.S. Fish and Wildlife Service, Bozeman Fish Technology Center, Bozeman, Montana, USA.
75
-------
Based on circumstantial evidence, whirling disease was introduced into North America in frozen,
processed fish imported from Denmark to Pennsylvania and Nevada in 1956. Whirling disease
was spread from these two original sources by the transfer of live fish and the natural movements
of infected fish. The first confirmed case of M cerebralis was diagnosed in a Pennsylvania
hatchery in 1958 (Hoffman 1990). A positive diagnosis of whirling disease in Nevada was not
documented until 1965, but subsequent examination of specimens collected in 1957
demonstrated the presence of the parasite. Since its introduction into the United States, the
parasite has been identified hi 18 states (Table 1). Except for Montana, the first case in each of
these states was based on finding whirling disease in a hatchery. There are currently no
documented cases of whirling disease in Canada or Mexico.
Table 1. Chronology ofMyxobolus cerebralis detection in the United States.
Year
State
1958
1961
1965
1966
1968
1973
1980
1984
1986
1987
1988
1993
1994
Pennsylvania
Connecticut
Nevada, California
New Jersey, Virginia
Ohio, West Virginia, Michigan
Massachusetts
New Hampshire
New York
Oregon
Idaho, Colorado
Wyoming
Utah
Montana
Importation of frozen trout into the U.S. ceased hi 1958. Until recently, the parasite
M. cerebralis was classified as a emergency prohibitive pathogen. Diagnosis of M. cerebralis in
a hatchery meant depopulation and disinfection of the entire facility. Regulations that govern the
movement of live fish in the U.S. are complicated. In the 1980's numerous experts proposed
that whirling disease was not as serious as once thought and that the prohibitive classification
was no longer justified.
A conference in 1988 brought together professionals from federal and state agencies and
private aquaculture to discuss this issue. One of the main points of the conference was that
depopulation and disinfection of hatcheries had not controlled the spread of whirling disease in
this country. Fish culture activities could minimize the losses to whirling disease and most
hatcheries reported few mortalities associated with whirling disease. Impacts on wild fish
populations were discussed but states reported varied experiences and it was generally thought
that the impact on wild fish was minimal. However, no detailed studies on susceptibility of wild
fish to whirling disease had been conducted in rivers positive for the parasite. The result of the
76
-------
conference was that whirling disease was down-listed from an emergency prohibitive to a
notifiable pathogen. Since the conference, many states adopted management policies that allow
the stocking offish lots with known positives into waters where the disease is enzootic but not
into whirling disease free waters.
A revolutionary discovery regarding the life cycle of this parasite was published by Wolf
and Markiw in 1984. Their research demonstrated that M, cerebralis has a two host life cycle
that involves a fish and an aquatic olgiochaete. Similar life cycles have since been described for
14 species of myxosporean parasites (El-Matbouli et al. 1995). After their discovery, limited
research was conducted in this country on the parasite. Population declines of wild rainbow
trout due to whirling disease were recently reported in areas of the Colorado River that had been
stocked with fish from M. cerebralis positive hatcheries (Walker and Nehring 1995). In
December 1994, declining populations of wild rainbow trout from the Madison River, Montana
were examined and found to be positive for M. cerebralis. These reports generated a great deal
of controversy about the impacts of whirling disease in wild trout populations.
CURRENT STUDIES
I
Finding M, cerebralis in wild rainbow trout from the Madison River, world famous for its
wild trout fishery, has brought national attention to effects of whirling disease on wild fish
populations. Clinical signs of whirling disease and severe pathological lesions associated with
infection have been observed in young-of-the-year rainbow trout from the Madison River.
Unlike most other states in the U.S., the fisheries management program in Montana relies on
wild reproducing populations and there has been a no-stocking rivers policy for the past 20 years.
To date, hatcheries in Montana, including federal, state and private facilities, have been tested
and found negative for the parasite, M. cerebralis.
Research studies at the University California Davis, pish Disease Laboratory have
resulted in the recent development of a DNA based diagnostic test using primers developed from
the 18S rRNA sequence (Andree et al. 1997). Several states in the U.S. are examining wild
rainbow trout populations for the presence of the parasite arid possible population effects. The
Whirling Disease Foundation, U.S. Fish & Wildlife Service and natural resource agencies within
the state of Montana are working cooperatively to conduct necessary research to develop
appropriate management strategies for addressing whirling disease in wild trout.
REFERENCES
Andree, K.B., E. MacConnell, T. McDowell, S. Gresoviac, R.P. Hedrick. 1997. Polymerase
Chain Reaction (PCR): A new approach to Myxobolus cerebralis diagnostics. Whirling
Disease Symposium-Expanding the Database: 1996 Research Progress Reports 111-116.
. Whirling Disease Foundation, Bozeman, Montana, USA.
77
-------
El-Matbouli, M., R.W. Hoffman and C. Mandok. 1995. Light and electron microscopic
observations on the route of the triactinomyxon-sporoplasm ofMyxobolus cerebralis from
epidermis into rainbow trout cartilage. Journal of Fish Biology 46:919-93 5.
Hofer,B. 1903. Ober die Drehkranheit der Regenbogenforelle. Allgemeine Fischerei-Zeitung
28, 1:7-8. (In German).
Hoffman, G.L. 1970. Intercontinental and trancontinental dissemination and transfaunation of
fish parasites with emphasis on whirling disease (Myxosoma cerebralis). American Fisheries
Society Special Publication 5:69-81.
Hoffman, G.L. 1990. Myxobolus cerebralis, a worldwide cause of salmonid whirling disease.
Journal of Aquatic Animal Health 2:30-37.
Markiw, ME. 1992. Salmonid Whirling Disease. U.S. Fish and Wildlife Service, Fish Disease
Leaflet No.17. Washington D.C., USA.
Walker, P.G. and R.B. Nehring. 1995. An investigation to determine the cause(s) of the
disappearance of young wild rainbow trout in the upper Colorado River, in Middle Park,
Colorado. Colorado Division of Wildlife Report, Denver, Colorado, USA.
Wolf, K., and M.E. Markiw. 1984. Biology contravenes taxonomy in the Myxozoa: new
discoveries show alteration of invertebate and vertebrate hosts. Science 225:1449-1452.
78
-------
ACIDIFICATION OF LAKES IN NORTHWESTERN RUSSIA AND RESULTANT
EFFECTS ON MORPHOFUNCTIONAL CHARACTERISTICS OF FISH GILLS
Victoria E. Matey1
ABSTRACT
The gills in wild perch (Percafluviatilis L.) from 23 acid and circumneutral lakes of
northwestern Russia (Karelia, Vologda, and Kostroma districts) have been investigated using
transmission (TEM) and scanning (SEM) electron microscopy and light microscopy (LM).
The gills offish from the acid lakes (pH 4.5-5.9) exhibited uniform structural alterations,
depending on the level of acidity and colourness of water. The most severe damages of gill
morphology occured in fish from the most acidic, highly colored lakes. Because of hypertrophy
and hyperplasia of gill epithelia, the primary and secondary lamellae in perch from all
investigated acid lakes exhibited swelling, and in fish gills from the most acidic colored lakes
(pH 4.5-4.9) the numerous neighbouring secondary lamellae were fused. The gills of perch from
clear and colored lakes were characterized by active mucus production, reduction of the
superficial microridges on respiratory cell surface, and substitution of a part of the respiratory
cells on the secondary lamellae with chloride cells. These differences in gill structure serve to
decrease gill surface area as well as to cause thickening of the secondary epethelium which
enlarges the water-blood diffusion barrier. These may be the reasons for inhibition of respiratory
gases exchange activity. The extremal elevation of chloride cell numbers both in primary and
secondary lamellae of gill in perch from acid lakes may represent a way of maintainence of ionic
regulation. This is especially important for fish from acid colored lakes, where the ultrastructural
alterations of chloride cells are more prominent than in fish from clear-water lakes. The
possibility of osmo-respiratory compromise of gills in fish from naturally acid lakes is discussed.
INTRODUCTION
Anthropogenous acidification of superficial waterbodies represents a serious ecological
problem for the European part of Russia (Komov and Lazareva 1994). The emission of SO2
from coal and petroleum products and deposition of strong acids as rain and snow are the main
sources of water acidification. Because of poor mineralization and low acid neutralizing capacity
(ANC) the lakes from northwestern Russia are especially susceptible to acid loads. Nowadays
about 10% of the lakes from the northern regions of Russia are acidified. In Karelia all lakes are
endangered in terms of acidification (Hettelling et al. 1991, Komov and Lazareva 1994).
'Institute of Biology of Inland Waters, Russian Academy of Sciences, Borok, Yaroslavl, Russia.
79
-------
Like Scandinavia, northeastern United States, and eastern Canada (Haines 1981, Jeffries
1990, Rosseland and Henriksen 1990) the lake ecosystems of northwestern Russia suffer from
negative impacts of acid deposition. This effect is intensified by mobilization of metals from
soils and atmospheric depositions (Haines et al 1992, Komov and Stepanova 1994). As a result,
fish populations in acidified waterbodies of the European part of Russia have declined or are lost,
and acid lakes from the northwestern region of the country are inhabited predominately by perch,
the species most tolerant to acidification.
One of the main reasons of the shortage offish fauna hi acid waterbodies is the direct
action of acidic water on the "target organ", the fish gills. Acidification, especially combined
with low mineralization and high concentrations of metal ions, effect both structural and
functional properties offish gills (McDonald 1983, Evans 1987, Laurent and Perry 1991).
The damage to the main functions of gills, ionic regulation and respiratory exchange, and their
morphology are well documented in experimental studies (Jagoe and Haines 1983, Evans et al.
1988, Wood et al. 1988, Mueller et al. 1991). Relatively few investigations have been conducted
on wild fish from natural acid waterbodies (Matey 1984, Chevalier et al. 1985, Karlsson-
Norrgren et al. 1986a, 1986b, Lewoetal. 1987, Matey et al. 1994). The purpose of the present
study is the comparative investigation of morphofunctional characteristics of gills in perch from
circumneutral and acid lakes from northwestern Russia (Figure 1).
Mestvur
Figure 1. Map of northwestern part of Russia showing the location of the studied regions:
1 - Karelia; 2 - Vologda district; 3 - Kostroma district.
80
-------
MATERIALS AND METHODS
Field studies were conducted during 1989-1992 in two regions of northwestern Russia:
Karelia and the Northern Region, namely the Vologda and Kostroma districts. Lakes occupy
about 20% of Karelia terrain and 2% of the Northern Region. Karelian lakes are based on
Precambrian granitic bedrock covered with thin soil. Lakes from the northern region, also on
Precambrian bedrock, are covered with thin sediments (hi the Kostroma district) or glacial tills
(in the Vologda district). Of 23 lakes examined, the acid ones predominated (Table 1). All lakes
studied were low in minerals (4.0-200 mg/L) with ANC < 0 (Haines et al. 1992, 1995, Komov
and Lazareva 1994). In clear lakes water color varied from 3 to 25°, and about 50% of the lakes
investigated were colored. These are characterized by the high content of dissolved organic
carbon (DOC). The dominant anion in clear and colored lakes is sulfate (Komov and Lazareva
1994). This confirms the anthropogenous nature of the acidification of these lakes (Ibid.).
Table 1. Lakes studied in the northwestern region of Russia.
Lake
Area
pH
Color (Hazen)
Number of fishes
studied
Sargozero
Venderskoye
Uros
Suoyarvi
Kabozero
Vegarousyarvi
Chuchyarvi
Leukunyarvi
Ilyakalkenyarvi
Vuontolenyarvi
Blue Lamba
Grushina Lamba
Lamba Vegarous
Hotavets
Uteshkovo
Motykino
Dorojiv
Dubrovskoye
Karelia
200
998
426
6070
210
1880
112
150
104
394
307
3
7
Northern Region-
160
5
2
200
20
Region
7.9
7.0
5.9
5.8
5.5
5.1
5.0
4.9
4.6
4.6
4.6
4.6
4.5
Vologda
7.5
4.8
4.8
4.6
4.6
Northern Region-Kostroma
Polovchinovskoye
Rybolovskoye
Skomorohovskoye
Rusilovskoye
Podlesnoy
170
19
22
37
10
7.6
7.5
6.9
5.8
5.0
25
23
9
104
141
; 105
8
182
170
186
3
3
I 182
District
188
150
19
; 13
182
District
25
101
78
: 22
125
5
5
7
7
7
8
8
6
5
8
6
7
8
10
6
12
7
10
7
5
6
7
7
81
-------
In contrast to acid lakes in Scandinavia, North America, and Canadia with high
concentration of aluminum (Spry and Wiener 1991), the predominant metal in acidified lakes in
northwestern Russia is iron (Fetota, 0.8-3.5 mg/L) (Komov and Stepanova 1994). In addition, high
concentrations of mercury have been discovered in fish (0.06-3.04 mg/g wet weight) from acid
colored lakes from the Vologda and Kostroma districts and from Karelia (Haines et al. 1992,
1995). A total of 164 perch (Percafluviatilis L.\ average weight 18-89 g, were sampled for
morphological investigations from 23 lakes (Table 1). The second and third gill arches were
dissected from each fish immediately after collection. The specimens for LM, TEM, and SEM
investigations were fixed in either 2.5% glutaraldehyde in 0.1M cacodylate buffer (pH 7.3) or in
a mixture of 1% glutaraldehyde and 2% paraformaldehyde in 0.1M HEPES buffer (pH 7.4).
Specimens for LM and TEM studies were post-fixed in buffered OsO4, dehydrated through a
graded ethanol series with a final change in absolute acetone, and embedded in Araldite® or
a mixture of Epon and Araldite.® Semi-thin sections for LM studies were cut from
perpendicularly-oriented gill filaments, stained with toluidine blue, and examined with a light
microscope (MBI-15, Karl Zeiss, Jena). The morphometric analyses of gill epithelia thickness
were executed on 10 randomly selected gill filaments and lamellae of five fish from each lake
investigated. Ultra-thin sections for TEM studies were stained with uranyl acetate and lead
citrate and examined with a transmission electron microscope (JEM 100C). Whole gill arches for
SEM investigations after the absolute acetone stage were critical-point-dried with liquid CO2,
sputter-coated with gold, and examined using a scanning electron microscope (JSM-25S).
RESULTS
The structure of gills in perch from the circumneutral clear and colored lakes studied
(pH 7.0 - 7.9) is typical for teleost fish. Primary lamellae (filaments) support two rows of
secondary lamellae, separated with wide interlamellar spaces (Figures 2a, 3a). The primary
epithelium covering gill filaments is thick (Figure 4) and shows few chloride (ion transporting)
and mucous cells, surrounded with squamous respiratory cells (Figures 3a, 5a). The chloride
cells of the primary epithelium have shallow apical pits with numerous microvillae; the apical
plasmic membrane of the respiratory cells forms a well-developed system of microridges
(Figure 5a). The epithelium of secondary lamellae is thin (Figure 4), and its surface consists of
respiratory cells only (Figure 3a).
The gills of perch from the acid lakes studied in Karelia, Vologda, and Kostroma
(pH 4.5-5.9) exhibited uniform structural alterations, depending on the level of acidity and
colorness of water. One of the most evident damages offish gills is the distortion of the
secondary lamellae. Due to hypertrophy and hyperplasia, the primary and secondary epithelia
were widened (Figure 4). The secondary lamellae in perch from acid lakes became shorter and
thicker than similar ones in circumneutral lakes in the same region (Figures 2a-f, 3a-d). This
effect was the most prominent for gills offish from highly colored acid lakes. The enlargement
of non-tissue spaces was not sufficient except for gill epithelia of perch from two colored lakes in
Karelia (Leukunyarvi, pH 4.9, and Vegarousyarvi, pH 5.1) and one colored lake from the
Kostroma district (Podlesnoye, pH 5.0)
82
-------
e
Figure 2. Micrographs of gills in perch from Vologda (a-d), Kostroma (e), and Karelia (f) lakes.
(LM). Scale bar =10 um for all cases.
a - circumneutral colored lake, pH 7.5; b - acid clear lake, pH 4.8; c - acid clear lake, pH
4.6; d - acid colored lake, pH 4.6; e - acid colored lake, pH 5.0; f-acid colored lake, pH
4.6. c - capillar; cc - chloride cell; pi - primary lamella; si - secondary lamella; re -
respiratory cell; arrow - secondary lamella fusion.
83
-------
Figure 3. Structure of gills in perch from circumneutral (a) and acid (b-f) Karelian lakes (SEM).
' Scale bar = 100 urn
a - clear lake, pH 7.0; b - colored lake, pH 4.6; c-colored lake, pH 4.5; d-clear
lake, pH 5.0; e - colored lake, pH 5.1; f - colored lake, pH 4.5 arrow - evaginated
region on primary lamella; double arrows - chloride cells on the surface of lamellae.
84
-------
Secondary epithelia gffgi Primary epithelia
* Colored lakes
Figure 4. Thickness of primary and secondary gill epithelia in perch
from lakes of northwestern Russia.
85
-------
Figure 5. infrastructure of primary epithelium surface of perch gills from (a) circumneutral and
(b-f) acid lakes (SEM). Scale bar = 10 jam.
a-d - Karelian lakes; e - Kostroma lake; f-Vologda lake.
a - clear lake, pH 7.9; b - clear lake, pH 4.6; c - colored lake, pH 5.8; d - clear lake,
pH4.6; e - colored lake, pH 5.0; f-colored lake, pH 4.6. ap - apical pit of chloride
cell; mg - mucous granules; mr - microridges; mv-microvillae; sp - secretory pore of
mucous cell.
86
-------
Hyperemia was the typical feature offish gills from all investigated acid clear- and
colored-water lakes (Figures 2b-f). The widening of blood vessels was one of the reasons of gill
lamellae enlargement As a result of swelling of the secondary lamellae the distances between
them in gills of perch from lakes with water pH < 5.0 were reduced. In fish from the most acid
colored lakes (pH 4.5-4.9) from Karelia (Leukunyarvi, Vegarousyarvi, Lamba-Vegarous) and the
Vologda district (Lake Dubrovskoe) about 15-25% of the secondary lamellae were fused
(Figures 2d, 3c). These fish also had locally swollen and evaginated regions on the primary
lamellae (Figure 3e). However, only in the fish from the single clear water lake from Karelia
(Chuchyarvi, pH 5.0) have we discovered partial fusion of secondary lamellae (Figure 3d).
The characteristic feature of gill epithelia structure in perch from all acid lakes studied
( pH 4.5-5.9) was the increased number of chloride Cells. Because of hyperplasia, the abundance
of chloride cells or their clusters has been found on the surface of the primary epithelium
(Figure 3f). Besides, the rich population of chloride cells was localized on the secondary lamellae
(Figure 2b-f, 6a-c). The number of chloride cells in primary and secondary lamellae appeared to
increase along with a decrease of water pH. Maximal number of these cells was found in fish
gills from the highly colored lakes (Karelia: Ilyakalkenyarvi, Vuontolenyarvi, Larhba Vegarous;
Vologda region: Uteshkovo, Dubrovskoye). The gills of perch from all acid lakes investigated '
were characterized by a high level of mucous secretion. A large amount of mucous granules on
the gill surface has been found in the fish from lakes with pH < 5.0 (Figures 5e,f). The color of
water had no definite effect on the process of mucous production.
The decrease of water pH and elevation of color.of the lakes studied were the reasons for
epithelial cell structure damages. As for the respiratory cells from primary and secondary
epithelia, the main characteristic for their alterations was hypertrophy resulting in the reduction
of the superficial system of microridges. Compared to perch from circumneutral lakes (Figure
5a), cell surface microridge density was slightly less in the acid lakes (pH < 5.5) (Figures 5c,d).
The most pronounced reduction of the labyrinth pattern of microridges has been discovered for
respiratory cells in fish from highly colored acid lakes.
Chloride cells from gill epithelia of perch from clear- and colored-water acid lakes were
hypertrophied. Due to this effect the apical surfaces of them were evaginated and the microvillae
were shortened, widened, and partially reduced ( Figures 5b-d). Only a few chloride cells had
deep apical pits (Figures 3f, 5e). The ultrastructure of chloride cells hi perch from acid lakes was
different from the same fish from circumneutral lakes. The most severe alterations took place in
the chloride cells offish gills from colored acid lakes where the concentration of metals and the
mercury body burden was extremely high. •
Chloride cells from gill epithelia of perch from clear- and colored-water acid lakes
consisted of numerous swollen mitochondriae and widely developed tubular reticulum
(Figure 7b). However, only in the most acidic and highly colored lakes from Karelia (Lamba
Vegarous, Vuontolenyarvi) and the Vologda district ( Dubrovskoye and Uteshkovo) the
cytoplasm of numerous chloride cells included some channel-like structures (Figure 7c). Besides,
the chloride cells which are disposed on the superficial layer of gill epithelia of perch from these'
lakes are saturated with a lot of large lysosome-like bodies (Figures 7d, e). The average
87
-------
Figure 6. Distribution of chloride cells in secondary gill lamella in perch from acid colored lakes
(SEM). Scale bar = 10 um
a - Kostroma lake, pH 5.0; b - Vologda lake, pH 4.6; c - Karelian lake, pH 4.9.
arrows - chloride cells
88
-------
Figure 7. infrastructure of chloride (a-e) and respiratory (f) cells from gill epithelium in perch
from Vologda circumneutral lake (a) and acid colored lakes (b-f). (TEM). Scale bar =
1 urn.
a-pH7.5; b-pH4.8; c-pH4.8; d,e-pH4.6; f-pH4.8.
cs - channel-like structure; Gc - Goldgi complex;^ Ib - lysosome-like body; m -
mitochondria; tr - tubular reticulurn; arrow - electron dense particles on the cell
surface.
89
-------
diameters of these structures were 0.5-0.8 urn, and they were filled with an electron-dense
content. The same lysosome-like bodies have also been discovered in respiratory cells, but they
were not as numerous as in the chloride cells (Figure If). The small electron-dense particles
appeared on the apical plasmic membranes of chloride and respiratory cells (Figures 7f). The
lysosome-like bodies have not been detected in chloride and respiratory cells from the middle
and basal layers of gill epithelia, nor were any detected in the mucous cells.
DISCUSSION
The data presented here demonstrate that the alterations in morphology of gills in wild
perch are related to water pH and degree of color. The extent to which perch gill morphology
was damaged increased with decrease of lake water pH. The highest degree of damage was
found in the most highly colored acid lakes. Because of hypertrophy and hyperplasia of
secondary epithelium the diffusion distance was enlarged up to three times for fish from colored
lakes and about one-and-a-half to two times for fish from clear-water lakes. The height of
secondary lamellae of perch from all acid lakes studied was markedly shortened. In the fish from
the most acidic and highly colored lakes, several neighboring secondary lamellae were fused.
These lesions reduce the total diffusion area offish gills (Mueller et al. 1991). The abundance of
chloride cells on the secondary lamellae also diminishes the surface for gas diffusion. In addition,
the reduction of microridge pattern and accumulation of mucus as well as precipitation of small
electron-dense particles on the gills decreases the capacity for gas exchange. Data obtained
corresponded to similar reports for other fish species affected by acidic and acidic/metal
contamination both in laboratory experiments and in the wild (Jagoe and Haines 1983, Karlsson-
Norrgrenefa/. 1986a, 1986b, Leinoetal. 1987, Laurent and Perry 1991, Mueller et al. 1991).
The described lesions on gill structure provoked the remarkable disturbances in respiratory
functions offish gills. On the other hand, observed changes hi gill morphology may represent an
adaptive reaction to support ionic regulation in fishes from natural acid lakes. The first evidence
for this is the extreme elevation in number of chloride cells both in primary and secondary
epithelia.
It was suggested (Laurent 1984, Evans etal. 1988, Matey et al. 1994) that chloride cells
hyperplasia may be considered as a mechanism of maintainance of ionic regulation in fish gills.
It is especially important for fish from acid colored lakes where the ultrastructural alterations of
chloride cells were more pronounced than in fish from the clear lakes with similar water pH.
The numerous lysosome-like bodies were found in chloride cells in perch gills from the most
acidic highly colored lakes of Karelia and the Vologda district. Similar alterations have been.
described in experimental and natural acidified waters with high concentrations of aluminum
(Chevalier et al. 1985,Karlsson-Norrgren
-------
precipitates are quickly desquamated from the gill epithelia surface and are substituted by young
chloride cells, originating from the basal layer of undifferentiated cells of the primary epithelium.
The swelling of chloride cells mitochondriae may be a result of the direct effect of contaminated
water. On the other hand, the presence of well developed tubular reticulum may be evidence of
high ionic pumping activity of chloride cells (Laurent 1984). The thickening of epithelia,
reduction of gill surface area, and accumulation of mucus on the gills are assumed to be the
reasons for the decrease in permeability of gill epithelia (McDonald 1983). Due to active ionic
consumption and limitation of ionic efflux from gill epithelia, the water-salt balance in fish from
the acid lakes we studied may be maintained.
There is definite competition between ionic and respiratory gill function in fish from
natural acid lakes. The maintenance of homeostasis occurs due to osmo-respiratory compromise
(Gonzalez and McDonald 1992). While ionic regulation is supported by responses of gill
epithelia and chloride cells, gas exchange may be provided by responses of the whole organism.
The last mechanism may involve the elevation of ventilation activity, heart beat rates, hyperemia,
and high levels of hemoglobin and hematocrit (Evans et al. 1988, Wood et al. 1988).
ACKNOWLEDGEMENTS
I thank Terry A. Haines, Charles H. Jagoe, and Victor T. Komov for their cooperation in
the numerous research expeditions, and in the valuable discussion about the data presented in this
paper.
REFERENCES '•
i
Chevalier, G., L. Gauthier, and G. Moreau. 1985. Histopathological and electron microscopic
studies of gills of brook trout, Salvelinusfontinalis, from acidified lakes. Canadian Journal
ofZoology 63:2062-2070.
Evans, D. 1987. The fish gills: site of action and model for toxic effect of environmental
pollutants. Environmental Health Perspectives 71:47-58.
Evans, R., S. Brown, and T. Kara 1988. The effect of aluminum and acid on gill morphology
in rainbow trout, Salmo gairdneri. Environmental Biology of Fishes 22: 299-311.
Gonzalez, R., and D. McDonald 1992. The relationship between oxygen consumption and
ion loss in a freshwater fish. Journal of Experimental Biology 16:317-332.
Haines, T. 1981. Acidic precipitation and its consequences for aquatic ecosystems. A review.
Transactions of the American Fisheries Society 110: 669-707.
Haines, T., V. Komov, and C. Jagoe. 1992. Lake acidity and mercury content offish in
Darwin National Reserve, Russia. Environmental Pollution 78: 107-112.
91
-------
Haines, T., V. Komov, C. Jagoe, and V. Matey. 1995. Perch mercury content is related to
acid color of 26 Russian lakes. Water, Air, and Soil Pollution 85:823-828.
Hetteling, J.-P., R. Hovning, and P. deSmet. 1991. Mapping critical loads for Europe.
Coordination Center for Effects, Technical Report 1, Bilthoven, the Netherlands
Jagoe, C., and T. Haines. 1983. Alterations in gill epithelium morphology in yearling Sunapee
trout exposed to acute acid stress. Transactions of the American Fisheries Society 112:689-
695.
Jeffries, D. 1990. Overview of areas for Canada and the United States of America. Proceedings
of the Fifth Program Meeting of Task Force, Freiburg, October 1989, Berlin: 54-65.
Karlsson-Norrgren, L., I. Bjorklund, O. Ljungberg, and P. Runn. 1986a. Acid water and
aluminum exposure: experimentally induced gill lesions in brown trout, Salmo trutta L.
Journal of Fish Diseases 9:11-25.
Karlsson-Norrgren, L., W. Dickson, O. Ljungberg, and P. Runn. 1986b. Acid water and
aluminum exposure: gill lesions and aluminum accumulation in farmed brown trout, Salmo
trutta L. Journal of Fish Diseases 9: 1-9.
Komov, V., and V. Lazareva. 1994. The reasons and consequences of athropogenous
acidification of surface water of the northern region. Comparative limnological investigation
of ecosystems of Darwin Reserve lakes. In: Structure and Functioning of Ecosystems of Acid
Lakes. V. Komov, Ed. Nauka, Saint Petersburg: 3-30 (In Russian)
Komov, V., and L Stepanova. 1994. Hydrochemical characteristic of lakes from Darwin
Reserve. In: Structure and Functioning of Ecosystems of Acid Lakes. V. Komov, Ed. Nauka,
Saint Petersburg: 31-42 (In Russian).
Laurent, P. 1984. Gill internal morphology. In: Fish Physiology. Volume 10A. W. Hoar and
D. Randall, Eds. Academic Press, New York: pp 73-183
Laurent P., and S. Perry. 1991. Environmental effects on fish gill morphology. Physiological
Zoology 64:4-25.
Leino R., P. Wilkinson, and J. Anderson. 1987. Histological changes in the gills of pearl dace
(Semotilus margaritd) and fathead minnow (Pimephales promelas) from experimentally
acidified Canadian lakes. Canadian Journal of Fisheries and Aquatic Sciences 44 (Suppl. 1):
126-134.
Matey, V. 1984. Comparative analysis of the gill epithelium ultrastructure hi perch from basins
with different ionic composition. Tsitologiya 26: 778-782 (In Russian).
92
-------
Matey V., C. Jagoe, and T. Haines. 1994. Morphology of gill apparatus offish from acid lakes.
In: Structure and Functioning of Ecosystems of Acid Lakes. V. Komov, Ed. Nauka, Saint
Petersburg. 213-228 (In Russian).
McDonald D. 1983. The effect of FT upon the gills of freshwater fish. Canadian Journal of
Zoology 63: 691-702.
Mueller, M., D. Sanchez, H. Bergman, D. McDonald, R. Rhem, and C. Wood. 1991. Nature and
tune course of accumulation to aluminum in juvenile brook trout (Salvelinusfontinalis).IL.
Gill histology. Canadian Journal of Fisheries and Aquatic Sciences 48: 2016-2027.
Rosseland, B., and A. Henriksen. 1990. Acidification in Norway - Loss offish populations and
the 1000 lake survey 1986. Sciences of the Total Environment 96: 45-56.
Spry, D., and J. Wiener. 1991. Metal bioavailability and toxicity to fish in low-alkalinity lakes: a
critical review. Environmental Pollution 71: 243-304.
Wood, C., R. Playle, B. Simons, G. Goss, and D. McDonald. 1988. Blood gases, acid base
status, ions, and hematology in adult brook trout (Salvelinus fontinalis) under acid/aluminum
exposure. Canadian Journal of Fisheries and Aquatic Sciences 45: 1575-1586.
Youson, J., and C. Neville. 1987. Deposition of aluminum in the gill epithelium of rainbow trout
(Salmo gairdneri Richardson) subjected to sublethal concentrations of the metal. Canadian
Journal of Zoology 65: 647-656.
93
-------
-------
ADVANCES IN THE CULTURE OF FRESHWATER FISHES IN CHINA
Chen Xianglin1, Wang Chun1, Zhao Jun1, and Chen Went2
ABSTRACT
Chinese freshwater fish culture has witnessed sustainable bumper harvests in the recent
decade. Of the total production in freshwater aquaculture, the output of "pond fish culture"
dominates. This resulted from improvement in traditional culture technology and advances in
the following aspects: overall artificial propagation, feeding suitability, nutritionally complete
feed, and improvement of water quality. In addition, the so-called "large water surface culture"
in such natural and/or semi-natural waters as lakes, reservoirs, and rivers has steadily progressed
under artificial biomanipulation by means of net-cage, net-fencing, and net-encircling cultures.
Progress has also been made in "paddy field fish culture". All these advances have benefitted
from research into fish eco-physiology and control of water quality. A long-term strategy of
Chinese freshwater fish culture should be oriented toward protecting fish resources in natural
waters and sustaining the normal ecosystem offish culture water bodies.
INTRODUCTION
China is among the world's big fishery countries, with the output of freshwater fish
products ranking first in the world. From 1958 to 1985, Chinese freshwater aquaculture was
making sluggish progress with its ups and downs, although the breakthough in 1958 in artificial
propagation of silver carp (Hypophthalmichthys molitrix) and bighead carp (Aristichthys nobilis)
reared in ponds made a fbundamental change in freshwater fisheries from a type of capture
fisheries, mainly catching the natural economic fishes, to an aquaculture system in which fish
culture played a dominant role. In the recent decade, along with research into fish eco-
physiology and with improvement in culture technology, the production offish culture has
developed quickly, and its output was 55% of total fishery production in 1994. Of total fish
reared, the output of pond fish culture dominated. Pond area surpassed 1.33 million ha,
accounting for only 35% of the total area but with the yield in ponds representing 75% in total
volume. In addition, large-scale aquaculture in such natural and/or semi-natural waters as lakes,
reservoirs, and rivercourses also provided bumper yields, and paddy field fish culture also
developed to a certain degree.
'South China Normal University, Department of Biology, Gunagzhou, PRC.
Hjuangdong Department of the Extension of Fishery Technology, Guangzhou, PRC.
95
-------
POND FISH CULTURE
Pond fish culture has all always been the mainstay in Chinese aquaculture. Historically,
Chinese pond culture was famous for grass-fish, mulberry-fish, and sugarcane-fish culture
systems based on a simple microecosystem. In the last 15 years, pond fish culture has tended
towards augmented, standardized, and diversified culture systems conforming to national
conditions. The following aspects helped to bring about sustainable, steady development in pond
fish culture.
Diversifmg Cultured Fishes by Introducing New Species
Such Chinese carps as black carp (Mylopharngodon aethiops), grass carp
(Ctenopharyngodon idelld), silver carp, bighead carp, common carp (Cyprinus carpio), crucian
carp (Carassius auratus), and mud carp (Cirrhina molitorelld) have been the main fishes
cultured. In recent decades, several species have also been introduced and acclimated from
abroad. Native species like long-snout catfish (Leiocassis longirostris), largemouth catfish
(Sihtrus meridionalis), common eel (Anguillajaponicd), Mandarin fish (Siniperca chuatsi),
snakehead fish (Channa argus) and bluntnose black bream (Megalobrama amblycephald) and
fine-scaled Chinese nase (Xenocypris macrolepis) were bred and acclimated to ponds and
became the objects of monoculture or polyculture. In addition, rainbow trout (Oncorhynchus
mykiss), mirror carp (Cyparinus carpio), tilapia (Tilapia mossambica), catfish (Glorias juscus),
channel catfish (Ictalurus punctatus), prussian carp (Carassius auratus gibelio), largemouth bass
(Microptenis salmoides), etc. have been introduced with success. On the other hand, China
strengthened research on selection and breeding of fine cultured varieties. More attention is now
paid to decline in strains of traditional Chinese carps and other cultured species. Farms for
original breeds (e.g. grass carp, bighead carp, and bluntnose black bream) were and will be
established in Hubei and Jiangsu provinces for purification and rejuvenation of reared fishes.
Breeding of new, more desirablely unproved varieties has been in progress and some are being
put into practice. So far, over 10 new hybrid varieties have been bred, including the "bumper"
variety and "Jian" variety of common carp, the "all-female" variety of common carp, and
allogynobenetic crucian carp.
Widespread Popularization of the Techniques of Overall Artificial Propagation in
Production of Seed Stock
Production of seed has been regarded as a gaurantee for aquaculture. So far, China has
achieved a breakthrough in overall artificial propagation from induced spawning of brood fish to
nursing of fry in cultured species, based on eco-physiology offish reproduction; an exception to
this is the eel. Especially in batch production of seed of fine varieties, widespread popularization
of the techniques of overall artificial propagation plays an important role in shift from a low-
96
-------
market value fish culture system to a high-market value one. Some endemic species such as
Mandarin fish, bluntnose black bream, long-snout catfish, and largemouth catfish have also been
acclimated and reared in earthen ponds with success.
Change from Extensive Systems with No Intentional Nutritional Inputs to Semi-Intensive
and/or Intensive Systems
Extensive systems are more appropriate for resource-poor low-income fish farmers. The
four Chinese carps are usually cultured with a low yield of less than 100 kg per 1/15 ha in
extensive systems, but with a high yield of over 400 kg in semi-intensive systems with natural
food increased by fertilization and perhaps also augmented by supplementary feeding with agro-
industrial by-products. Now there is a tendency to rear precious and profitable species in
intensive systems which rely on nutritionally complete diets, either traditonal diets based on live
feeding fish, frozen trash fish, or on formulated diets usually in dry pelleted form. For example,
carnivorous large-mouth catfish and long-snout catfish were cultured in line with polyculture of
the four Chinese carps to control trash fish in earthen ponds, but are now acclimated to intensive
systems (e.g. net-cage culture) in which they are reared with monoculture and fed on formulated
pelleted feed.
Along with the practice of semi-intensive and intensive systems, an augmented culture
system, in which scattered ponds are joined together for good management and for reaping stable
and high yields, emerged and played an important role in founding commercial fish bases.
Standardization of culture techniques has been tested as a guarantee for stable and high yields,
especially in high-market value fish culture systems. During the period 1985-1986, eel culture
in earthen ponds succeeded in Guangdong province, and subsequently was rapidly popularized in
a large-scale. In 1992, standardization of culture technology in earthen ponds on the Pearl River
Delta was designed to guarantee eel production, based on the the following ecological principles:
management techniques, transformation of ponds, integrated utilization of feed, and control of
water quality and disease prevention. The output of cultured eels reached over 7,000 tons with
up to 0.3 billion Chinese renminbi yuan in production value in 1,069 ha of earthen ponds, and
intensive eel culture became an important means to generate foreign exchange through export.
The Chinese pond fish culture systems based traditionally on a simple microecosystem,
e.g., grass-fish, mulberry-fish, and sugarcane-fish pond culture as mentioned earlier, are now
considered minimally productive. Integrated fish farming involving consideration of production
technology, social and economic aspects, and environmental aspects, is considered more
productive. In this agri-ecosystem, fish culture is usually combined with livestock or poultry
production and crop cultivation, with high efficiency of energy conversion and good economic
benefit. It is believed that various modes of integrated fish farming, illustrated in Figure 1,
should be the main orientation for pond fish culture.
97
-------
1 slaughter I J restaurants I*
table fish
J
Mifcat
Figure 1. Model of integrated fish farm in China.
Exploitation of Artificial Feed and Application of Formulated Feed to Fish Culture
In reared species, some are herbivorous like grass carp and blunt snout bream, some are
omnivorous like common carp, tilapia, and some are carnivorous like largemouth catfish, long-
snout catfish, and Mandarin fish, and others are planktonphageous. Hence, more attention is
placed on exploitation of feed resource. In the past, most nutritional research concerned pond or
lake fertilization and the production of natural food. A wide range of potential fertilizers was
assessed as pond inputs: human excrement, livestock manure, plant matter, and inorganic
fertilizers. Recently, more attention has been directed to the use of low-cost supplementary and
complete feed. A large number of aquatic and terriestrial plants, such as green fodder, have been
fed to herbivorous fish, and lots of products and by-products of agri-industrial processes have
been fed to omnivorous fish. Research into nutritionally complete pelleted feed has contributed
directly to the success in the species culture. To give an example, Liang Xufang (1994) •
98
-------
formulated a pelleted feed for Mandarin fish that naturally feed on live fish. This feed was based
on research into the Mandarin fish's sense mechanism related with feeding behavoir. The
acclimation rate and survival rate reached over 78% and 93% respectively. This promoted
further development of semi-intensive culture of grass carp. To date, China has succeeded in
manufacturing nutritonally complete diets for grass carp, common carp, bluntnose black bream,
and allogynogenetic crucian carp. i
Improvement in Water Quality and Disease Preventions in Ponds
Following the expansion of the culture of varieties, there is demand to improve water
quality in ponds. Techniques to cope with oxygen depletion include using aerators, draining the
water and adding fresh water have been adopted. Quick lime is generally used to regulate pH
and to improve pond substratum. Photosynthetic bacteria are added to pond water to increase
energy conversion efficiency. Most diseases can be minimized through proper management of
the culture system and the avoidance of stress caused by abrupt temperature changes, low
dissolved oxygen levels, and accumulation of hydrogen sulfide and ammonia-nitrate. Fish-crab,
fish-turtle, and fish-shrimp polyculture systems have been adopted to reduce incidence of
disease.
LARGE WATER-SURFACE FISH CULTURE
The so-called "large water-surface fish culture" means culturing fish in natural and semi-
natural water bodies like lakes, reservoirs, and rivers. In past decades, lake fish culture has by
nature been a type of capture fisheries, as the seed stock was released into lakes for growth and
was later harvested by fishermen. Fishery utilizaton being an important function carried out in
lakes, commercial fish farming in many lakes, especially in shallow lakes, was actually started in
China in the 1960s. Owing to the insufficiency of experience, people tended simply to adopt the
traditional pond pisicultural techniques for the lakes, treating them as huge fish ponds, e.g. dense
planting of silver carp and bighead carp, weeding and even artificially fertilizing. The results
were unsuccessful in most cases, neither making expected profit nor keeping lakes in a healthy
state (Liang and Liu 1995). In 1986, exploitation experiments were conducted at Gehe Lake in
Jiangsu Province, Baoanhe Lake in Hubei Province, and Huangyanhe Lake in Anhui Province by
means of "net-encirclement" and "net-cage" culture techniques. In 1987 the value of total
production of these three lakes was 9.43 million Chinese renminbi yuan, an increase of 60%
compared with the fish crop before experimentation (Liu and He 1992). Although the total
volume of cultured fish has gradually increased year in year out through net-cage, net-encircling,
and net-fencing culture techniques, there still is divergence between maintaining the stability of
lake environments and utilization of lake fishery. Lake aquaculture seems to be inclined toward a
moderate culture system, in which it becomes much more important to conserve and rationally
99
-------
utilize macrophytes and to culture high-market value species, e.g., Chinese mitten-handed crab
(Eriocheir sinensis) and Mandarin fish, rather than traditional Chinese carps. The following
mode may be the guideline for fishery development and ever-lasting utilization of macrophytic
lakes (Figure 2, after Chen and Cao 1991).
EWIROMNT
QUESTION
decline in biodiversity and
diminution in fish resource
CAUSE
lake swamping and drop
in water regulation
barrier beteen lake and
overfishing
CONTEHCVE
succession of
submerged vegetation
extra development of
emergent macrophytes
reform of production,
shift of labour
EEFKT
large-size species and fine
varieties for recruitment
leading to
Figure 2. Optimization between fishery and the environment in a macrophytic lake of China.
Reservoir fish culture has been regarded only as a supplementary function of reservoirs.
Planktonophageous species (e.g. silver carp andbighead carp ) are usually cultured in small-sized
reservoirs. In medium-sized and/or large-sized reservoirs with ovef 5 m of water depth,
moderate net-cage culture systems, in which fine species that are high in the food chain (e.g.
Mandarin fish, largemouth catfish, long-snout catfish, rainbow trout and large-mouth bass
dominate production, are generally adopted. Small numbers of carnivorous species like
mandarin fish and snakehead fish are usually stocked into reservoirs to control wild trash fish.
100
-------
In some places, people culture fish in high-density net-cages or net-fences in rivers.
Cultured species are mainly fed on artificial formulated feed. Guarding against fish escape and
flood prevention are major concerns in everyday management.
PADDY FIELD FISH CULTURE
Paddy field fish culture, based on ecological symbiosis, is an important supplementary
mode of aquaculture hi a bid to produce seed of herbirovous fish like grass carp or omnivorous
fish like common carp for ponds, lakes and reservoirs. There are two main modes of paddy field
fish culture. One is the rice-fish culture system in which fingerlings are reared to young fish
during rice growing. The other is a rice cultivation and fish culture rotation system. In 1988, the
area of paddy field fish culture in China reached 0.67 million ha with a total fish harvest of
118,000 tons. For a developing country like China, emphasis for paddy field fish culture will be
placed upon culvitating rice, with fish production being secondary.
SUMMARY
Systematic studies on freshwater fish culture, especially on cultural eco-physiology of
fish, have led to further development in aquaculture in ponds, lakes, and reservoirs. Freshwater
pisciculture has played an important role in minimizing the catch of wild fish resources in natural
waters and in protecting fish gene pools. There are still some problems that have existed for a
long time and defy solution. As long as the human population continues to grow faster than fish
production, pollution will become more serious and result in a decrease in biodiversity. Long-
term strategies for fish culture should be associated with the maintenance of a normal ecosystems
and the optimization of environmental protection. The emphasis should, however, be placed
upon the direct and indirect impacts of aquaculture on pisci-diversity and environment. China is
now taking the following problems into account and increasing studies on these:
1. Ecological impacts of the introduction of exotic species on endemic species, especially
the impacts of the introduction of carnivorous species on native fauna, and the protection of
wild fish resource;
2. Ways to prevent the escape of hybrid varieties into natural waters resulting in gene
pollution;
3. Prevention of overfishing of economic species hi natural waters;
4. Control of eutrophicatiori and moderate aquaculture under artificial biomanipulation in
lakes and reservoirs. I
101
-------
REFERENCES
Chen Yiyu, and Cao Wenxuan. 1991 Exploitation of the aquatic biological productivity and
strategy for the environment improvements in Lake Honghu. In: Studies on
comprehensive exploitation of aquatic biological productivity and improvement of
ecological envrionment Make Honghu. China Ocean Press, Beijing, pp 1-10. (In
Chinese)
Liang Xufang. 1994. Acclimation of feeding habit in Siniperca. Freshwater Fishery 6:36-37.
Liang Yanling, and Liu Huoquan. 1995. Resource, enviornment and fishery ecological
management of macrophytic lakes, Science Press, Beijing. (In Chinese)
Liu Jiankang, and He Biwu. 1992. Cultivation of the Chinese freshwater fishes. Science Press,
Beijing, pp 5-12. (In Chinese)
102
-------
MERCURY UPTAKE BY SACRAMENTO BLACKFISH:
METHODS AND PRELIMINARY RESULTS
M. Heekyoung Choi1, Joseph J. Cech, Jr.1, and Manuel C. Lagunas-Solar2
ABSTRACT
Methyl mercury (MeHg) uptake by the gills of Sacramento blackfish (Orthodon
microlepidotus) under different temperatures was measured using 203Hg (half life 46.9 days)
radioisotope as a tracer. The efficiency with which fish gills extracted mercury from water was
measured with a respirometer that separated exposure (inspired) water from expired water using
a liquid scintillation counter. Sacramento blackfish gill ventilation volumes and oxygen (O2)
consumption rates increased as temperature increased from 20 to 25 C, while MeHg uptake rates
were not changed significantly. Post-experimental tissue analyses showed that more MeHg
accumulates in Sacramento blackfish gills, liver, spleen, intestine, and heart than in their kidney,
brain, gonad, and muscle after 1-hour exposures. Accumulation of MeHg in organs tended to
increase with temperature, although sample sizes were small.
INTRODUCTION
Hazardous levels of mercury (Hg) have been reported in aquatic systems throughout the
world including Japan, Scandinavia, and North America. Concern for human health has
provided an impetus to study Hg accumulation in fishes (Stiefel 1976). A high concentration of
Hg, exceeding the U.S. Food and Drug Administration (FDA) action level of 1 part per million
(ppm), has been found in fishes from the coastal mountain lakes of northern California since the
early 1970's (Stratton et al. 1987). These sources contain both inorganic and organic forms of
Hg, but biological activity in bottom sediments is capable of converting either form to the more
toxic methyl mercury (MeHg) complexes (Jensen and Jernelov 1969). It has been found that Hg
is present in freshwater fish species almost entirely as MeHg (Westoo 1973).
Sacramento blackfish {Orthodon microlepidotus) are large (up to 1.5 kg) plankton- and
algae-eating members of the minnow family (Cyprinidae). They switch from suction feeding on
individual particles as juvenile fish to filtering available prey as adults (Staley 1980). Blackfish
are native to California, and is one of the most abundant fish species in Clear Lake. They have a
lower overall body burden of Hg than largemouth bass (Micropterus salmoides) in Clear Lake
which may be due to their shorter life span (Stratton et al. 1987) or different Hg uptake or
depuration rates (Richman et al. 1988).
'Department of Wildlife, Fish, and Conservation Biology, University of California, Davis, California, USA.
2Crocker Nuclear Laboratory, University of California, Davis, California, USA.
103
-------
While it is documented that Hg accumulates in animal tissues after exposure (Johnels et
al 1968, Weisbart 1973, Boudouef al. 1979, McMurtry etal. 1989), mechanisms underlying the
uptake of Hg by fishes are not fully understood. There have been numerous direct toxicity
studies as well as studies of biochemical and physiological effects (Weis and Weis 1976,
Hawryshyn et al. 1982, Fletcher and White 1986) of Hg on fishes. Many studies have
demonstrated experimentally that fishes accumulate Hg directly from the surrounding water as
well as from food (Westoo 1973, Olson et al. 1973, Phillips and Buhler 1978, Huckabee et al.
1978, Boudouetal. 1979).
In our experiments, we isolated one route of the bioaccumulation process, uptake of
MeHg across the fish gills, to examine the effects of environmental variables, such as
temperature, dissolved O2, and dissolved organic carbon, on the bioavailability of MeHg to
Sacramento blackfish. The efficiency with which fish gills extract MeHg from water is measured
using a metabolic chamber that separates inspired (exposure, pre-gill) water from expired (post-
gill) water by means of latex membranes (McKirii and Goeden 1982). Direct measurements of
Hg uptake rate eliminate potential kinetic analyses errors of bioaccumulation due to rapid
biotransformation of the compound, pharmaco-dynamics within the organism, or statistical
interdependence (co-correlation) of fitted parameters in the iterative program used to estimate
uptake rate coefficients (Black and McCarthy 1988). This method allows us to investigate the
bioavailability of MeHg for uptake by fish gills under different environmental variables. In this
paper we report on the method used and some preliminary results regarding effects of
temperature.
MATERIALS AND METHODS
Animals
Sacramento blackfish, 900 to 1,200 g, were collected by seining from Clear Lake,
California. They were held in 600-L glass fiber tanks under natural photoperiod for
approximately 1 year before experiments. Tanks were aerated and received a continuous flow of
atmospherically equilibrated well water. Total hardness, total alkalinity, and pH were
determined before each experiment (mean total hardness of well water = 117.6 mg/L as CaCO3;
mean total alkalinity = 246.5 mg/L as CaCO3; mean pH = 8.3). Fish were maintained on Silver
Cup trout pellets (Murray Elevators, Idaho) but not fed for 48 hours before surgery.
Chemicals
In order to investigate the effect of environmental levels of MeHg on physiological
changes of aquatic animals, a procedure for the methylation of radioactive mercuric chloride
(HgCl2) with no addition of non-radioactive Hg is necessary. The 203Hg was purchased from
Buffalo Materials Research Center (Buffalo, New York) as HgCl2 (specific activity of 2.5
mCi/mg) in IN HC1. Methylation is accomplished by reacting the HgCl2 with tetramethyl tin
((CH3)4Sn) (Toribara 1985). Four steps of extraction produce a radioactive Me203HgCl
compound which is finally dissolved in a dilute NajCC^ solution (0.05M) suitable for biological
104
-------
experiments (ibid.). Paper chromatography was used to evaluate the end product of methylation.
The radiolabelled MeHgCl was chosen for this experiment due to the simple and sensitive
analysis. ,
i
Surgical Procedures
Each fish was anesthetized with 100 mg/L MS-222 (tricaine methanesulfonate, Sigma
Chemical Co., Missouri) buffered withNaHCO3 andNaCl, and immobilized by spinal
transection (McKim and Goeden 1982, Schmieder and Weber 1992). Then fish were cannulated
through the dorsal aorta (Cech and Rowell 1976) with PE 50 (Intramedic polyethylene tubing,
0.58 mm ID x 0.965 mm OD, Becton Dickinson, New Jersey) for serial blood sampling for Hg.
An oral latex membrane was sutured around the mouth to allow direct measurements of Hg and
O2 uptake (Table 1). After surgery, fish were placed in the respirometer to recover in aerated
well water (University of California, Davis) from a separate reservoir at least 20 hours before
MeHg exposure. |
Table 1. Abbreviations of measured and calculated variables.
Ventilation volume (L/h/kg)
Inspired oxygen concentration (mg O2/L)
Expired oxygen concentration (mg O2/L)
Oxygen consumption (mgO2/h/kg)
Oxygen extraction extraction efficiency (%)
Physiological vaiables
Vg
CIO2
CEO2
MO2
UO2
Measured
(O2 solubility * PIO2)/Air pressure 1.428 mgO2/ml
(O2 solubility * PEO2)/Air pressure 1.428 mgO2/ml
Vg * (CI02—CE02)
(CIO2—CEO2)/CIO2* 100
Mercury concentrations in inspired and
expired water (ng/L)
Mercuy extraction efficiency (%)
Mercury uptake rate (ng/h/kg)
Mercury uptake variables
Cfflg, CEHg Measured
UHg (Cfflg—CEHg)/CEHg* 100
RHg Vg * (Cfflg—CEHg)
Respirometer ;
The respirometer (modified from McKim and Goeden 1982) measures 20.3 x 20.3 x 66
cm (0.95-cm thick Plexiglas®) and incorporates a water jacket for temperature control, and vents
and filters for using the radioisotope. It is divided into three compartments (A, B, and C),
separated by latex membranes (Figure 1). A latex membrane sewn to the fish's mouth separated
inspired (Compartment A) and expired (Compartment B) water, allowing calculation of MeHg
and O2 extraction efficiency (expressed as percentage reduction of MeHg, or O2 concentration in
inspired water). A second membrane, just posterior to the pelvic fins, prevented exposure of the
skin surfaces to expire water posterior to the fins (in Compartment C). After each experiment,
food dye was added to Compartment B to determine if there had been any leakage between
compartments.
105
-------
BOA contaminated water
Figure 1. Apparatus for direct measurements of methyl mercury uptake by fishes.
Water was circulated through the respirometer via a combination of a mechanical pump,
the fish's gill ventilation apparatus, and gravity. The water containing MeHg was pumped from
Reservoir 1 by a metering pump (Alldos, Minnesota) to compartment A (Figure 1). Water in
Compartment A moved to Compartment B by the fish's gill ventilation. The water not used for
ventilation in Compartment A drained via surface drainpipe back to Reservoir 1. Water
ventilated over the gills was drained via surface drainpipe to Reservoir 2 and pumped through a
series of water purification filters (Adsorber/Universal System, Cole-Parmer, Illinois) to clean up
the 203Hg after the experiment. The radioactivity of water was checked using a Wallac Model
1410 liquid scintillation counter (Pharmarcia, Finland) before disposal. Reservoir 3 held the
water drained from the Compartment C, which was maintained at the same water level as
Compartments A and B after placing the fish hi the chamber. This balance the head pressure
between compartments. The whole system was closed, except for the vents connected to the In-
Line Gas Purifier (400 mL, Alltech, Illinois) in the fume hood, to prevent the release of
volatilized Hg to the atmosphere. Water temperature was maintained by a thermoregulator
(CFT-75, Neslab, New Hampshire) circulating water through the water jackets of the chamber
and reservoir. The supply water in Reservoir 1 was aerated to minimize volatilization by using
chemically inert porous PTFE tubing (IMPRA Inc., Arizona).
106
-------
Exposure and Sampling Procedures
i
The fish were exposed to aqueous solutions of the Me203HgCl under the selected
temperatures in the respirometer. Stock solution of 203Hg was added to Reservoir 1 and allowed
to dissolve in the water and mix via aeration for approximately 1 hour. During the experiment,
water samples (ImL) were collected from Reservoir 1, Compartment A, and Compartment B,
and placed in 20-mL liquid scintillation vials for Hg concentration measurements. Water was
simultaneously sampled for O2 measurements from compartments A and B by a peristaltic pump
(Minipuls, Gilson, Wisconsin) to the O2 meters (Instrumentation Laboratories, Massachusetts).
Heavy-walled, teflon (PFA) tubing was used (except vitori tubing for the peristaltic pump) for
the sampling line to minimize losses of Hg and O2. Fish blood was collected for analysis of
Hg concentration using the dorsal aortic cannula. Blood collection via a cannula circumvents
effects on partial pressures of O2 (PO2) or CO2 (PCO2), or on hormone concentrations
(i.e., catecholamines and cortisol) otherwise caused by blood sampling stress. Ventilation
volume was determined by measuring the volume of water that flowed out of the Compartment B
drainpipe as a function of time. The measuring chamber connected to the drainpipe was
equipped with a light sensor and a solenoid which gave signals to an electronic stopwatch.
Ventilation frequency was measured by visually counting the buccal-opercular frequency using a
video monitor. To minimize stress, the chamber was partially covered with black plastic during
each experiment.
Radioactivity Measurement
Water samples were prepared for scintillation counts and subsequent MeHg
concentrations. One mL of water containing radiolabelled MeHg was mixed with 10 mL of
liquid scintillation cocktail (Universol, ICN Biomedicals, California) in a 20-mL borosillicate
liquid scintillation vial, and the 203Hg activity was measured using the liquid scintillation counter.
Quenching effects were corrected. Methyl mercury concentrations were calculated from the
specific activities of the stock solution and corrected for radioactive decay.
Each fish was sacrificed by over-anaesthetization with MS-222, and then dissected for
tissue Hg concentrations. Tissue samples were taken fronrgills, liver, spleen, muscle, kidney,
intestine, heart, brain, gonad, and gall bladder. Gill filaments were washed with deionized water
and blotted with paper towel to remove extra Hg adsorbed on the surface. For tissue digestion, .
100 mg of tissue of each organ was transferred to a 20-mL borosilicate liquid scintillation vial
with a plastic-lined cap. The use of vials with cork or aluminum lined caps are not suitable since
they may react with the solubilizer vapors producing a harmful product. For blood, 0.2 mL was
obtained for each vial. After addition of 1 mL of 1:1 potassium hydroxide/methanol
(KOH/CHjOH) solubilizer, tissues were homogenized and sealed tightly. The tissue:solubilizer
weight ratio was 1:10. The vials were placed in the water bath and heated at 60 C for 4 hours
until clear. After cooling to room temperature, 100 uL of 60% perchloric acid (HC1O4) and 100
uL of 30% hydrogen peroxide (H2O2) were added to the vials with color (gills, liver, spleen,
kidney, heart, gall bladder, and blood). Vials were then placed in the water bath and heated at 70
C for 30 to 60 minutes until clear and colorless (Mahin and Lofberg 1966). Sample vials were
107 ;
-------
cooled to room temperature. Sample preparation for radioactivity measurements was completed
by addition of 10 mL scintillation cocktail to the vials. Vials were allowed to equilibrate 30 to
60 minutes in the dark before counting.
Statistical Analysis
Abstat 5.0 statistical software (Anderson-Bell, Colorado) was used for the Student's t-test
and one-way analyses of variance to compare the different treatments' means.
RESULTS AND DISCUSSION
Respiration and MeHg Uptake
Sacramento blackfish were exposed to 37.5 ± 0.7 ng/L Me203Hg at 20 or 25 C and
9.43 ± 0.58 mg O2/L. Both the mean ventilation volume (Vg) and oxygen consumption (MO2)
offish exposed to Me203Hg at 20 C were significantly lower (P < 0.05) than those exposed at 25 C
(Table 2). The higher temperature provoked a 68% Vg increase and a 83% MO2 increase
compared with measurements at 20 C. The MO2 Q10 value between 20 to 25 C was high (3.35),
whereas the Vg Q10 was 2.82. Changes in environmental temperature can influence the fish
MO2 because temperature directly affects the metabolic activity (e.g. mitochondrial O2 demand)
of the fish's cells. To take up more or less O2, a fish adjusts its ventilation volume and the gill
surface area in contact with the respiratory water (Randall 1982). Similar increases in MO2 of
Sacramento blackfish were observed in a study by Cech et al. (1979). However, the MO2 values
in our experiment at both temperatures were lower than those measured by Cech et al. (1979).
These differences can be explained by the differences in fish weights. With data calculated on a
per kilogram basis, smaller fish (600g) tend to have a higher MO2 than larger fish (lOOOg). Also,
Cech et al. (1979) used a respirometer for which fish surgery was not needed.
Table 2. Average value of respiration and MeHg uptake in Sacramento blackfish as a function of
temperature. (MeHg concentration of inspired water was 37.5 ±0.7 ng/L.)
Temperature
Variable
Ventilation volume (L/h/kg)
Oxygen extraction efficiency (%)
Oxygen consumption (mg O2/h/kg)
Mercury extraction efficiency (%)
Mercury uptake (ng/h/kg)
20 C (n=5)
27.2±4.0*
23.2±8.0
60.0115.3*
33.4110.4
335.3±110.5
25 C (n=2)
45.819.4*
29.6+3.8
109.5±37.3*
22.616.4
395.71229.5
Values are mean ± SD. * Significantly different (P<0.05) between two different temperatures.
108
-------
The increase in Vg and MO2 did not include significant changes in O2 and UHg. This
meant that fish were increasing water flow across their gills while maintaining a constant percent
removal of O2 and Hg at the lamellar surface. The relatively constant uptake for MeHg with its
log octanol/water partition coefficient (log Kow) of 0.07 (Major etal. 1991) could be due to
diffusion via aqueous pores in biological membranes, which would not be affected by
lipophilicity (Erickson and McKim 1990). This might indicate that MeHg was taken up by the
same diffusion mechanism by which O2 and other lipophilic compounds are taken up. However
the transporting mechanism for Hg might be different. A high affinity Hg binding-to sulfhydryl
(SH) groups in red blood cells might explain higher Hg concentrations in specific, well-
vascularized tissues (Figure 2). Urine was initially collected for Hg concentration using urinary
bladder cannulae. However, we did not find Hg in the urine, presumably due to the short
exposure time.
* -r
•o
2
a
I i
43 \S
C
a
Figure 2. Concentrations of Me203Hg in the organs of Sacramento blackfish
(n = 5.at20C,n = 2at25C)
Methyl Mercury Accumulation in Organs
Concentrations of Me203Hg in the organs of fish after 1 -hour exposure are shown in
Figure 2. These results showed that almost no differences (P > 0.05) existed between MeHg
concentrations in the organs at 20 and at 25 C except in liver. Higher mean Hg concentrations
accumulated in spleen and heart at 25 C, although no statistical significance can be attached
because of the small sample sizes. The results showed that in the initial stages of MeHg
exposure, the gills retain most of the Hg burden. The three times higher Hg burden in gills
compared to other organs reflected surface adsorption rather than true uptake such as the
appearance of metals in the blood (Zia and McDonald 1994). Indeed, it appears that metal
uptake by the gills does not involve direct and rapid transfer from the water to the blood,
109
-------
but rather an intermediate step where the metals first accumulate in the gill tissue. Methyl Hg
levels in other organs confirm that there is an uptake of MeHg through the gills and subsequent
transport by the blood to these organs. During this exposure period, gills, liver, spleen, intestine,
and heart tend to accumulate MeHg more than do kidney, brain, gonad, and muscle. Generally,
the well-vasGularized organs tend to have a high MeHg concentration, supporting the concept
that Hg is efficiently taken up and carried by the blood.
CONCLUSION
Based on these preliminary results, even though the ventilation volume was increased, the
constant MeHg uptake rates at increasing temperatures indicate the existence of a saturation level
in the blood. Therefore, we might assume that the biological uptake of MeHg is directly related
to the SH binding sites in the red blood cell, although a larger sample size and more direct
evidence are still needed. These and other data will form a basis for predicting MeHg uptake
under a variety of physiological and environmental conditions in this fish species.
ACKNOWLEDGMENTS
We appreciate the technical assistance of Albert Chan, Kent English, Timothy Essert, and
Mark van de Water. We thank James McKim, John Nichols, and Gary Gill for their help to
initiate this research, Dick Sijm for support and Paul Lutes and Williams Bentley for animal care.
This study is supported in part by grants from the Ecotoxicology Program of the University of
California Toxic Substnaces Research and Teaching Program, and from the U.S. EPA (R819658)
Center for Ecological Health Research at U.C. Davis. Although this research was funded by the
U.S. EPA, it may not necessarily reflect the views of the Agency and no offical endorsement
should be inferred.
REFERENCES
Black, M. C., and J. F. McCarthy. 1988. Dissolved organic macromolecules reduce the uptake
of hydrophobic organic contaminants by the gills of rainbow trout (Salmo gairdnerf).
Bulletin of Environmental Toxicology and Contamination 7:593-600.
Boudou, A., A. Delarche, F. Riberyre, and R. Marty. 1979. Bioaccumulation and
bioampliflcation of mercury compounds in a second level consumer, Gambusia affinis -
Temperature effects. Bulletin of Environmental Contamination and Toxicology 22:813-818.
Cecil, J. J. Jr., and D. M. Rowell. 1976. Vascular cannulation method for flatfishes.
Progressive Fish-Culturist 38:74-75.
Cech, JJ. Jr., S.J. Mitchell, and MJ. Massingill. 1979. Respiratory adaptations of Sacramento
blackfish, Orthodon microlepidotus (Ayres), for hypoxia. Comparative Biochemistry and
Physiology 63 A:411-415.
110
-------
Erickson, R. J., and J.M. McKim. 1990. A Simple flow-limited model for exchange of organic
chemicals at the fish gills. Environmental Toxicology and Chemistry 9:159-165.
Fletcher, T.C., and A. White. 1986. Nephrotoxic and haematological effects of mercuric
chloride in the plaice (Pleuronectesplatessa L.). Aquatic Toxicology 8:77-84.
Hawryshyn, C.W., W.C. Mackay, and T.H. Nisson. 1982, Methyl mercury induced visual
deficits in rainbow trout. Canadian Journal of Zoology 60:3127-3133.
Huckabee, J.W., S.A. Janzen, B.G. Blaylock, Yair Talmi, and JJ. Beauchamp. 1978.
Methylated mercury in brook trout (Salvelinus fontinalis ): Absence of an in vivo methylating
process. Transactions of the American Fisheries Society 107:848-852.
Jensen, S., and A. Jernelov. 1969. Biological methylation of mercury in aquatic organisms
Nature 223:753-754. !
Johnels, A.G., M. Olsson, and T. Westermark. 1968. Esox lucius and some other organisms and
indicators of mercury contamination in Swedish lakes and rivers. Bulletin of the Office
International des Epizootics 69:143 9-1452.
Mahin, D.T., and R.T. Lofberg. 1966. A simplified method of sample preparation for
determination of tritium, carbon-14, or sulfur-3 5 in blood tissue by liquid scintillation
counting. Analytical Biochemistry 16:500-509. ;
Major, M.A., D.H. Rosenblatt, and K.A. Bostian. 1991. The octanol/water partition coefficient
of methylmercuric chloride and methylmercuric hydroxide in pure water and salt solution.
Environmental Toxicology and Chemistry 10:5-8. ,
McKim, J.M., and H.L. Goeden. 1982. A direct measure of the uptake efficiency of a xenobiotic
chemical across the gills of brook trout (Salvelinus fontinalis ) under normoxic and hypoxic
conditions. Comparative Biochemistry and Physiology 72:65-74.
McMurtry, M.J., D.L. Wales, W.A. Scheider, G.L. Beggs, and P.E. Diamond. 1989.
Relationship of mercury concentration in lake trout (Salvelinus namaycush) and smallmouth
bass (Micropterus dolomieui) to the physical and chemical characteristics of Ontario lakes.
Canadian Journal of Aquatic Sciences 46:426-434. \
Olson, K. R., H.L. Bergman, and P.O. Fromm. 1973. Uptake of methyl mercuric chloride and
mercuric chloride by trout: a study of uptake pathways into the whole animal and uptake by
erythrocytes in vitro. Journal of the Fisheries Research Board of Canada 30:1293-1299.
Phillips, G. R., and D. R. Buhler. 1978. The relative contributions of methymercury from food
or water to rainbow trout (Salmo gairdneri) in a controlled laboratory environment.
Transactions of the American Fisheries Society 107:853-861.
Ill
-------
Randall, DJ. 1982. The control of respiration and circulation in fish during exercise and
hypoxia. Journal of Experimental Biology 100:275-288.
Richman, L.A., D.C. Wren, and P.M. Stokes. 1988. Facts and fallacies concerning mercury
uptake by fish hi acid stressed lakes. Water, Air, and Soil Pollution 37:465-473.
Schmieder, P.K., and L.J. Weber. 1992. Blood and water flow limitations on gill uptake of
organic chemicals hi the rainbow trout (Oncorhynchm mykiss). Aquatic Toxicology 24:103-
122.
Staley, C.S. 1980. Life history aspects of the Sacramento blackfish, Orthodon microlepidotus
(Ayres) in the Beach/Stone Lakes basin. M.S. Thesis. Biological Science, California State
University, Sacramento. 55pp.
Stiefel, R.C. 1976. Mercury hi white bass and carp. Ohio Department of Natural Resources.
Division of Wildlife. Columbus, Ohio, USA. 1-29 pp.
Stratton, J.W., D. F. Smith, A. M. Fan, and S. A. Book. 1987. Methylmercury in northern
coastal mountain lakes: guidelines for sport fish consumption for Clear Lake (Lake County),
Lake Berryessa (Napa County), and Lake Herman (Solano County). Administrative report.
Hazard Evaluation Section and Epidemiological Studies and Surveillance Section. Office of
Health Services. 29pp.
Toribara, T. Y. 1985. Preparation of Me203HgCl of high specific activity. International Journal
of Applied Radiation and Isotopes 36:903-904.
Weis, P., and J.S. Weis. 1976. Effects of heavy metals on fin regeneration hi the killifish,
Fimduhis heteroclitus. Bulletin of Environmental Contamination and Toxicology 16:197-
202.
Weisbart, M. 1973. The distribution and tissue retention of mercury-203 in the goldfish
(Carassius aurarus). Canadian Journal of Zoology 51:143-150.
Westoo, G. 1973. Methylmercury as percentage of total mercury in flesh and viscera of salmon
and sea trout of various ages. Science 181:567-568.
Zia, S., and D.G. McDonald. 1994. Role of the gills and gill chloride cells hi metal uptake in the
freshwater-adapted rainbow trout, Oncorhynchus mykiss. Canadian Journal of Fisheries and
Aquatic Science 51:2482-2492.
112
-------
SPECIATION AND BIO AVAILABILITY OF HEAVY METALS
IN SEDIMENTS OF THE LE AN RIVER, GHBVA
Lin Yuhuan1 and Guo Mingxin1
ABSTRACT
Several heavy metals in large quantity exist in the sediments of the Le An River from the
copper mining region to the Poyang lake area. A study of speciation and bioavailability of heavy
metals (Cu, Zn, Pb, Mn) in sediments was carried out in a simulated ecological system with
aquatic macrophytes and benthic organisms. The results show that two species of organisms,
benthic Cipangopaludina cathayensis and the aquatic rooted plant Scirpus triqueter are excellent
indicators of toxicity and bioavailability of heavy metals in sediments. They represent uptake of
heavy metals from sediment and from interstitial pore water, respectively. The concentrations of
metals in pore water and in geochemical phases of sediment were measured with an osmotic
technique and sequential extraction procedure separately; the regression coefficients between
bioaccumulation of metals in tissue and concentration of metals in pore water and one or several
geochemical phases were used to indicate bioavailability of metals in sediment, and the toxicity
of heavy metals was measured by growth, mortality, and reproductivity of the organism. The
methodology of sediment quality standard and criteria used in many laboratories at present was
used to compare uptake routes of metals in sediments.
INTRODUCTION
The speciation of metals in sediments plays an important role in their bioavailability and
toxicity. Toxicity of metals is different from one species to another. The sediment toxicity test
is a bioassay of increasing importance for scientists and technicians involved in testing the
toxicity and bioavailability to benthic organisms of heavy metals in sediments. There has been
little international standardization in this area, although the American Society for Testing and
Materials presented a set of recommended procedures in 1991, and the U.S. Environmental
Protection Agency has held workshops on this subject (U.S. EPA 1992). In China there are still
no standards for heavy metals in sediment of rivers and lakes, even though much pollution has
occurred in rivers and lakes caused by mining and industrial effluents.
The methods for bioassays and toxicity tests of sediments in the past decades were
mainly focused on tests of pore water and elutriate and bulk sediment with single indicator
organisms. Interpretation and extrapolation of results are often difficult due to uncertainties.
A simulated ecosystem, composed of biota, water, and sediment has been developed for bioassay
and toxicity testing of sediments with consideration of natural environmental conditions
(OECD 1992, SETAC 1993). This paper presents results of a study using this ecosystem, which
includes aquatic macrophytes, alga, benthic organisms, and a water-sediment unit to test
bioavailability and toxicity of metals in sediments.
'Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, PRC.
113
-------
Site Descriptions
Acid mine drainage and alkaline wastewater mixed together enter into the Le An River
and undergo a series of physicochemical reactions. Heavy metals in the wastewater eventually
deposit into sediments or flow downstream to Poyang Lake (Figure 1). The distribution of the
main taxonomic groups, and of species richness and individual abundance at every sampling site
along the Le An River are shown in Figure 2.
Figure 1. Research area and sampling locations.
114
-------
Algae (x 10 4/L)
Protozoa (ind/L)
Crustacean,
Biomass
A01
ACM
AOS A08 A13
A16
i _.\ ' Variations of individual abundance of
3.1-.
• plankton at six sampling stations
Algae (mg/L)
• Protozoa (mg/m *)/
Crustacean,
A04 AOS AOS A13 A16
1 b). \ Variation of biomass at six sampling stations
Speciet ricboeu
Individual abundance
species / m-
A01 A04 A05 A08 A13 A16
Distribution of species richness and individual
abundance alone Le An River
Metal content (ppm)
3900 -j-
2500
A01 AO4 AOS AOS A13 A16
,, : Di«tributionofCu2*,Pb2*and2n2+
'' in sediments in Le An River
Figure 2. Distribution of taxons and concentrations of metals in sediments of the
LeAn River, China.
115
-------
Methodology of Bioassay
A conceptual model of the working plan is illustrated in Figure 3. A simulated ecosystem
which consists of sediment, water, two kinds of macrophytes, and a group of benthic organisms,
was used for bioassay and toxicity testing (Figure 4).
eco-simulated
test , -
Selected organism?
and methods
Diversity of
community
structure, jetc.
Review of
literature
on toxicitjr
and speciatiott
Selection of test, methods
'.for- simuLated,_ecosystem
Speciation of metals
analytical procedures.
I = r^-1 -— .
- --.- [Test of. ginrmlgted' ecosystem
-I
Growth, development" ~
reproduction,
mortality,
accumulation of metals
L
ive
relationship
I
-jAaaesjaneht of'methoda]
Figure 3. Basic scheme of the research program procedure.
116
-------
//
IJ I\
^
Figure 4. Representation of a simulated test ecosystem: 1. Hydrochryseum \vilfordi;
2. Scripus triqueter; 3. Gorbiculafluminea; 4. Cristaria celtiformis;
5. Cipangopaludina cathayensis; 6. Chlorella.
Sediment samples were collected along the Le An River at Danjiangkou, an upstream
background location, and at three test locations: Caijiawan^ Hushan, and Gukou (Figure 1).
The basic properties of the sediment samples are shown in Table 1.
Table 1. Sediment characteristics.
Station Sediment
Number Sources
SI
S2
S3
S4
Danjiangkou
Caijiawan
Hushan
Gukou
Percentage of sediment
<55 (im
66.8
73.0
64.2
62.0
55-250 um
14.9
19.4
20.5
20.7
grains (%)
Content of
>250 urn Organics (%)
18.5
7.6
15.3
17.3
0.75
:2.72
2.58
2.44
CEC
i (me/lOOg)
76.8
68.8
68.1
54.9
pH
7.12
6.89
6.97
7.84
Eh
(mv)
-154.0
-91.0
-138.0
-208.0
117
-------
The pore water of sediment and acid volatile sulfides were measured by methods of Guo
(1995) and Allen et al. (1993). The test organism taxons, their numbers and size are reported in
Table 2. The simulation experiment began on February 28, 1994 and ended on August 27, 1994,
a total of 180 days spanning the winter, spring, and summer seasons.
Table 2. Test organism characteristics, numbers and size
Species
Cipangopaludina
calhqyensis
Anodonta
woodiana
Cristaria
celtiformis
Corbiciila
Jluminea
Characteristics
sediment feeder
filter feeder
filter feeder
filter feeder
Number
Benthic animals
5
3
4
&
Size (cm)
Shell 2.8-2.6
Shell 8.4-8.8
Shell 5.4-5.6
Shell 2.0-2.1
Average wet
weight (g)
4.39
95.8
16.4
2.44
Scirpus triqueter
Hydrochryseum
•wilfordi
Chlorella
Aquatic rooted plants
Some underground stems
with 15 buds
One plant
Alga
20 ml of suspended algae solution
Height 11-13
Methods for Analysis of Elements in Sediments and Water
The concentration of metals in overlying and pore water was determined by inductively-
coupled plasma emission spectroscopy (TCP) and atomic absorption spectroscopy (AAS) after
they were digested by HNO3. The sample of sediments and biotas were treated by HF and
HC1O4, HNO3 in a pressure container, after which the concentration of metals were determined
by ICP and AAS. The concentration of anion in water was analyzed by a Dionex automatic
chromatograph Model 16.
RESULTS AND DISCUSSION
During the 6 months of the test, the aquatic plants grew from bud to blossom and bore
fruit, completing a life cycle. At the end of the experiment, the weight and height of aquatic
plants were measured, the sediment was wet sieved by 280 mesh and the larva of the river snail
(C. cathayensis) were counted, the concentration of metals in the organism tissues was measured,
and the speciation and concentration of metals in the sediments and pore water were analyzed by
the sequential extraction procedure and osmotic technique.
118
-------
Growth and Development of Organisms
S. triqeter: Results of all experiments were the same in that the bud grew quickly up to
about 20 cm of stem. The differences of growth and toxicity effects of heavy metals on plants
occured after 3 months. At Station SI (background) the stem grew fresh and green, whereas new
buds grew gradually up to the water surface. However, in contrast with SI, the,stem tip of plants
at Station S2 became yellow and gray, and grew slowly. The growing situation at Stations S3
and S4 were similar to S2, in that stems became yellow and then stopped growing and withered
to death. There is a correlation between the life cycle of the plant growth and the pollutant
tendency of heavy metals.
H. wilfordi: The combined height and weight response of this plant after 10 days
exposure to heavy metals until the end of the experiment were S4 < SI < S2 < S3 (Table 3).
Station S4, where the concentration of metals is more than at other stations, provides a long-term
exposure test on the toxicity of heavy metals. The plants of Station S3 grew faster than the
plants in S1 and S2 due to control of sunlight; the reason is not clear.
Table 3. Growth and development of H. wilfordi.
Station number
SI
S2
S3
' S4
Sediment sources
Danjiangkou
Caijiawan
Hushan
Gukou
Plant height (cm)
,74
73
78
,70
Wet weight (g)
29.2
32.7
53.9
17.2
Alga: The C. chlorella grew quickly in the beginning of 5 days and then disappeared
gradually from the test ecosystem. Perhaps they were eaten by benthic organisms.
A. woodiana and C. celtiformis: Both are benthic filter feeder organisms of big size, and
have high resistance to heavy metals. They displayed robustness, low variability in response,
and low control mortality. Accumulation of metals in their tissues is a suitable indicator of metal
pollution.
C. fluminea is sensitive to heavy metal pollution. Its mortality is an appropriate endpoint
for a test. Mortality of C. fluminea hi our tests is reported in Table 4. Larva of C. fluminea were
not present in the sediment because they could not survive in static water bodies.
Table 4. Mortality of C. fluminea.
Station number
Total number
Dead number
Mortality (%)
SI
S2
S3
S4
8
8
8
8
; 0
0
6
8
0
0
75
100
119
-------
C. cathayensis is also sensitive to heavy metals and can digest sediment. Its larva is an
appropriate indicator of pollution. Survival and mortality of larva during our tests is reported in
Table 5.
TableS. Mortality of larva of C. cathayensis.
Station number
Total number
Dead number
Mortality (%)
SI
S2
S3
S4
38
46
37
24
27
36
32
22
71
78
87
92
The combined results showed that the number of larva and their mortality indicate the
pollution tendency of heavy metals in sediment. In the Station S4 experiment, the mortality
reached as high as 91%, and the field survey showed that C. cathyensis did not exist in sediments
at Gukou. The test results from Stations S2 and S3 showed that the number of C. cathayensis
was reduced gradually; this was also supported by the field survey. Snail larva are very sensitive
to heavy metals. In China, C. cathayensis is distributed in a wide range of aquatic systems.
Incidentally, in the Station S1 and S2 tests, L. hoffineisteri and Branchiura sowerbyi were found
in the sediment, possibly due to use of natural sediments for the test.
In general, the simulated ecosystem is a suitable long-term exposure instrument for
assessing heavy metal pollution. The simulated system contained multiple species of test
organisms, such as bentbic deposit feeders, benthic filter feeders, macrophytes, and algae. In this
kind of system, algae and macrophytes play an important role in sustaining the ecosystem
function and maintaining water quality. Algae also provide sufficient oxygen and food to the
other aquatic organisms, whereas C. cathayensis is the best scavenger.
Accumulation of Metals in Test Organisms
The concentrations of metals in the tissues of organisms are reported in Table 6. The
results indicated that accumulation of metals in tissue was a function of species, habits, and
uptake route of organisms in addition to environmental condition. For example, C. cathayensis,
a sediment feeder, was different from A. woodland, a filter feeder. The variation in accumulation
of the different metals hi a given species illustrates the bio-preferential effects for thesemetals.
For example, the concentration of Cu was more than that of Mn and Zn in C. cathayensis.
In contrast, the concentration of Zn in A. woodiana was more than that of Cu. The concentration
of metals varied in roots, stems, and leaves of plants. Based on these results, the selection of
organism for sediment bioassays is very important. In general consideration, the selection of
organism should satisfy the following rules: (1) ecological relevance, including representatives
of local taxons and their functional and structural role in ecosystem; (2) sensitivity to toxicants,
relative to other species and discriminatory to biological uptake routes; (3) availability through
culture and/or field collection.
120
-------
Table 6. Concentrations of heavy metals in test organisms (mg/L).
Station number
SI
S2
S3
S4
SI
S2
S3
S4
SI
S2
S3
S4
C. cath.
40.0
692
3360
1700
258
305
382
259
17.2
135
39.2
32.6
C. celti.
13.4
34.6
75.6
181
111
308
225
284
1480
2500
2620
2840
A. woodi
5.77
29.3
70.6
136
241
213
183
205
1560
3200
1400
1720
C.flumi.
Copper
15.3
256
.
-
Zinc
155
106
-
-
Manganese
12.9
22.6
-
-
S. triqui.
7.20
55.9
68.3
158
37.0
153
137
97.6
224
3570
984
458
H. w. root
39.6
564
563
906
95.9
666
293
620
709
3660
2830
782
H. w. stem
5.17
11.2
12.3
20.9
57.8
87.4
71.3
184
39.2
81.2
52.1
32.4
H. w. leaf
15.2
31.4
18.5
20.3
262
133
115
93.6
262
510
234
124
Relationship Between Bioaccunmlation and Concentration of Metals in Pore Water and
Overlying Water
The concentrations of metals in pore water and overlying water were determined and are
reported in Tables 7 and 8. Regression coefficients were obtained to determine the relationship
between bioaccumulation and metal concentrations in pore water and in overlying water
(Table 9). Our results indicate that accumulation of metals in C. cathayensis is not related to the
concentration of metals in pore water and overlying water except for Mn. However,
C. celtiformis and A. woodland can only ingest Cu from pore and overlying water. The
accumulation of metals in S. triqueter is related to concentration of pore and overlying water.
In the different part ofH. wilfordi, the concentration of metals is varied. The comparison of
concentration in pore and overlying water shows the relative extent of soluble metals taken up by
organisms. The concentration of Mn in pore water and overlying water is highest, followed by
Zn and Cu. The accumulation of metals in C. celtifomis and A. woodman is similar, but
accumulation in C. cathayensis has an opposite order. The order of accumulation in S. triqueter
and H. wilfordi is Mn > Zn > Cu.
Relationship Between Bioaccumulation and Speciation of Metals in Sediments
Regressive coefficients were obtained to determine the relationship between
bioaccumulation and speciation of metals in the sediments (Table 10). The results indicated that
the accumulation of Cu in C. cathayensis is not related to the total concentration of Cu in
sediment, but only related to the geochemical phases of carbonate and Fe-Mn oxide bound. The
accumulation of Cu in C. celtiformis and A. woodiana is related to the total concentration of Cu,
as is the accumulation of Cu in the aquatic plants.
121
-------
Table 7. Concentration of metals in pore water (mg/L).
Station
number
81
S2
S3
S4
Cu
1
0.020
0.031
0.060
0.139
2
0.003
0.013
0.024
0.014
3
0.004
0.006
0.011
0.019
Avg.
0.009
0.017
0.032
0.057
Pb
1 2
0.014 -
0.015 *
0.011 -
0.012 -
3 Avg.
- 0.005
- 0.005
- 0.004
- 0.002
Zn
123 Avg.
0.390.180.200.25
1.180.560.490.74
1.040.470.310.61
0.89 0.38 0.25 0.51
Mn
1
2.80
38.6
9.39
7.37
2
1.53
11.6
5.05
0.79
3
0.23
16.1
2.84
1.39
Avg.
1.52
22.1
5.76
3.18
Note: 10th day, 45th day, 165th day
Table 8. Concentration of metals in overlying water (mg/L).
Station
number
SI
S2
S3
S4
Cu
1
0.007
0.020
0.020
0.028
2
0.006
0.003
0.010
0.010
3
-
0.010
0.011
0.021
Avg.
0.004
0.011
0.014
0.007
Pb
1 2
0.006 *
0.011 *
-
*
3 Avg.
- 0.002
- 0.004
-
0.002 0.128
1
0.112
0.293
0.195
0.082
Zn
2
-
3 Avg.
- 0.037
0.1390.0180.150
0.1070.0420.115
••
0.070 0.27
Mn
123
0.180.120.04
5.76 0.41 0.07
0.340.120.01
0.080.040.13
Avg.
0.11
2.08
0.16
Note: 10th day, 45th day, 165th day
Table 9. Relationship between bioaccumulation and concentration of metals in pore water and
overlying water (Regression coefficients).
Water-Organism C. cath. C. celt A. woodL S. triqu. H. w. root H. w. stem H. w. leaf
Pore water 0.5471 0.9935
Pore-*-overlying 0.5691 0.9861
Pore water 0.0124 0.2245
Pore + overlying 0.4541 0.1271
Pore water 0.9980 0.2806
Pore + overlying 0.9984 0.2710
Copper
0.9998 0.9821 0.8836 0.9056 -0.0769
0.9984 0.9775 0.9136 0.9772 -0.0150
Zinc
-0.6466 0.9882 0.7265 0.0606 -0.7062
-0.5985 0.9803. 0.7027 0.2297 -0.6834
Manganese
0.9520 0.9996 0.8489 0.9611 0.9104
0.9567 0.9990 0.8405 0.9588 0.9129
122
-------
Table 10. Relationship between bioaccumulation and speciation of metals in sediments
(Regression coefficient).
Sed. Cu-Bio. Cu
Total
Soluble
Exchangeable
Carbonate
Fe-Mn oxide
Org-partial sulfide
Residual
a(l+2)
a (1+2+3)
a (1+2+3+4)
a (1+2+3+4+5)
C. cath.
0.4516
-0.0769
0.8637
0.9774
0.9004
0.3979
0.3583
0.8626
0.9667
0.9566
0.6902
C. celt.
0.9975
-0.2154
-0.0603
0.4756
0.0254
0.6561
0.9931
-0.0620
0.2421
0.1776
0.6092
A. woodi.
0.9915
-0.1252
0.0690
0.5877
0.1495
0.6906
0.9695
0.0676
0.3676
0.3040
0.6854
S. triqu.
0.9904
0.7413
-0.0589
0.5027
-0.0007
0.7958
0.9593
-0.0590
0.2532
0.1814
0.7259
H. w. root
; 0.8926
0.2992
, 0.1237
: 0.6468
0.1354
0.9439
0.8083
: 0.1258
0.4288
0.3432
0.9094
H. w. stem
0.9790
0.0352
-0.0187
0.5401
0.0303
0.7105
0.9358
-0.0183
0.2996
0.2195
0.7740
H. w. leaf
-0.0169
0.7868
-0.0872
0.0580
-0.2166
0.6663
-0.1346
-0.0798
0.5779
-0.0704
0.5239
a Note: 1 Soluble, 2 Exchangable, 3 Carbonate, 4 Fe-Mn Oxide, 5 Organic-Partial sulfide
Relationship Between Concentration in Pore Water and Speciation in Sediments
Regressive coefficients were determined between concentrations of metals in pore water
and speciation in sediment (Table 11). The results illustrate that the concentration of Cu, Zn, Mn
in pore water can be related to the concentration of metals in sediment. The contribution of
different geochemical phases to the concentration in pore Water is very complex. The source of
Cu in pore water comes mainly from soluble Fe-Mn oxides and organic-partial sulfide phases,
Zn comes from soluble carbonate, Fe-Mn oxides and organic-partial phases, and Mn comes from
soluble, exchangeable, organic-partial sulfide phases. The distribution of metals in geochemical
phases is dependent mainly on the characteristics of the elements.
Table 11. Relationship between speciation and concentration of metals in pore water
(Regression coefficient).
Sediment bioaccumulation
Total
Soluble
Exchangeable
Carbonate
Fe-Mn oxide
Org.-partial sulfide
Residual
s(l+2)
a (1+2+3)
a (1+2+3+4)
8 (1+2+3+4+5)
Cu
0.9929
-0.1389
0.0524
0.5729
0.1345
0.6819
0.9741
0.0509
0.3511
0.2879
0.6721
Zn
0.7502
0.9831
1 0.8502
0.9187
0.8409
0.9038
0.8192
0.8884
0.9157
,0.8648
1 0.8656
Mn
0.9799
0.9906
0.9978
0.8202
0.7324
0.9251
-0.4177
0.9990
0.9942
0.9939
0.9939
1 Notes: 1 Soluble, 2 Exchangable, 3 Carbonate, 4 Fe-Mn Oxide, 5 Organic-Partial Sulfide
123 :
-------
Bioavailability of Speciation of Metals and Test Organisms
Based on our results, the bioavailability of metals for a given organism is different from
one metal species to another (Table 12). Different species of organism were able to uptake one
or several species of metals in sediments. However, benthic organisms C. cathayensis,
a sediment feeder, is only one species available as a test organism. It can directly swallow
sediment and uptake metals. Aquatic macrophyte and benthic filter feeder organisms can only
uptake metals in pore water and overlying water. Due to variance of the distribution of metals in
geochemical phases, the percentage of metals hi sediments for bioavailability is different from
one metal to another.
Table 12. The content (mg/L) and percentage of metals for bioavailability in sediments.
Serial number
Cu
(for C. cathyensis only)
Zn
Mn
SI
S2
S3
S4
1.18(4.8%)
49.4 (9.0%)
198 (18.7%)
53.5 (2.0%)
10.2 (7.8%)
141(11.6%)
95.3 (15.9%)
21.6(3.4%)
17.8 (4.4%)
814(53.5%)
223 (27.8%)
57.2 (8.2%)
SUMMARY AND CONCLUSION
Bioassays and toxicity tests were applied in a simulated ecosystem to test bioavailability
of metals in sediments. The results indicate that responses (growth and development of
organism, survival and mortality, accumulation of metals in tissue) to metal pollutants of
organisms in sediment are consistent with bioavailability and toxicity of metal speciation in
sediment.
(1) The ecosystem constructed of sediment, overlying water, and test organisms, and
intended for long term tests, is an autotropic system under simulated natural conditions; it is able
to maintain the growth and development of test organisms for completing life cycles. The
selection and structure of organisms play a key role in system function; alga can supply
organisms oxygen and food; C. cathayensis and A.woodiana, C.celtiformis can clean the
overlying water; the source of nutrient for organisms comes only from within the ecosystem; the
system can form a sustainable productivity.
(2) C. cathayensis and S. triqueter are excellent bioindicators of metals for bioassay and
bioavailability of metal speciation. The discrimination effects of metal toxicity for larva of
C. cathayensis characterizes the availability of metals in sediments. The accumulation of
metals hi C. cathayensis reflects on the available portion of metals in sediments and relative
speciation. The concentration of Cu, Zn, and Mn in £ triqueters is consistent with the
concentration of metals hi pore water that comes from sediment. Two species of aquatic
organisms grow hi a wide region of the Yangtzi River basin, and have ecological relevance.
Their amenability and sensitivity to a variety of sediments recommends them as test organisms
of metal toxicity and bioassays.
124
-------
(3) The results of relationship between accumulation and speciation of metals
characterized in sequential extraction procedures indicate that bioavailable portion of metals in
sediment is some forms and phases existed in sediment other than the total amount of metals.
For example, the sediments feeder C. cathayensis can ingest mainly the species (forms) of metal
in sediment that are soluble, exchangeable, carbonate bound, and Fe-Mn oxides bound. The
filter feeder organism and rooted plants S. triqueter can only uptake the soluble form of metals,
that is, soluble, exchangeable and Fe-Mn oxide bound species. For example, S. triqueter can
only uptake the Cu, Zn, and Mn existing in pore water. However, A. woodiana and
C. celtiformis uptake Cu in pore water and overlying water that include suspended matter and
organism detritus, alga, etc. The bioaccumulation of metal is dependent highly on the uptake
route of organisms.
(4) The concentration of metals in pore water is an excellent indicator of bioavailability
of metals in sediments, considering AVS/SEM if AVS/SEM < 1. AVS has an important effect
that influences the bioavailability of metals in pore water due to precipitation of metals.
ACKNOWLEDGMENTS
We acknowledge the Natural Science Foundation Committee, China, for providing
financial support for this study. We thank the ICP Group, Research Center for
Eco-Environmental Sciences for helping us in analysis work, and we give thanks to Zhu Jiang
and Xu Muqi for their help in logical selection and classification of aquatic organisms.
REFERENCES
Allen H. E., Gongmin Fu, and Baolin Deng. 1993. Analysis of acid-volatile sulfide (AVS) and
simultaneously extracted metals (SEM) for estimation of potential toxicity in aquatic
sediments. Environmental Toxicology and Chemistry 12:1441-1453.
Guo Mingxin. 1995. Methodologies of relationships between bioavailability and speciation of
heavy metals in sediments. M.S. Thesis. Research Center for Eco-Environmental
Sciences, Chinese Academy of Sciences.
OECD. 1992a. Report of the OECD Workshop on the Extrapolation of Laboratory Aquatic
Toxicity Data to the Real Environment. OECD Environment Monograph No. 59.
SETAC (Society of Environmental Toxicology and Contamination). 1993. Guidance document
on sediment toxicity tests and bioassays for freshwater and marine environments. In The
Workshop on Sediment Toxicity Assessment, November 8-10, 1993.
U.S. EPA (U.S. Environmental Protection Agency). 1992. Proceeding from the Workshop on
Tiered Testing Issues for Freshwater and Marine Sediments, U.S. EPA, Washington,
D.C., September 11 -18,1992.
125
-------
-------
LETHAL AND SUBLETHAL EFFECTS OF COPPER ON FISHES
Edwin W. Taylor1, Patrick J. Butler1, Matthew W. Beaumont1,
Jeaniue Mair1, and Mujallid I. Mujallid1
ABSTRACT
Copper is very toxic to fishes, particularly the cupric ion in acid waters. The toxic action
is on the gills where copper displaces calcium, causing structural damage to the epithelium and
inhibiting transport enzymes. At high levels of toxicity, disruption of ionoregulation is followed
by impairment of respiratory gas-exchange and death due to circulatory failure. Sublethal levels
of copper reduce swimming performance without affecting blood oxygen and carbon dioxide
levels. Oxygen supply to the tissues may, however, have been restricted by increased blood
viscosity. Exposure to lethal and sublethal levels of copper caused a marked elevation of tissue
and plasma total ammonia, implying that copper interferes with an active ionoregulatory
component of ammonia excretion. Reductions hi swimming performance were directly
proportional to plasma ammonia levels. The basis of this relationship is currently under
investigation.
INTRODUCTION
Although traces of copper are essential constituents of some enzymes, copper is toxic to
both animals and plants at levels only just in excess of those found in many unpolluted aquatic
habitats. In many aquatic environments levels of dissolved copper have been increased from
anthropogenic origins such as mine washings or its direct application as an algicide,
molluscicide, or anti-fouling agent. Consequently, global copper emissions tripled between 1950
and 1980 (Moore and Ramamoothy 1984) and copper has been classified as one of the more
potentially hazardous heavy metals (Sposito 1986).
Copper, however, can exist in natural waters hi a wide variety of physico-chemical forms,
many of which are of low bioavailability and/or toxicity. A relatively low concentration of
organic complexing agents may effectively bind all the available copper. In natural waters, these
organic ligands include humic, fulvic, and amino acids and polypeptides, all of which can be
derived from decomposition processes and anthropogenic inputs such as sewage. Copper may
also be found in particulate and colloidal states arising from adsorption of copper by, for
example, hydrous metal oxides and clays. Little is known of the effects of these complexed
forms of. copper once they become incorporated into the food chain but, at least in the short term,
they will be non-toxic (Shaw and Brown 1974). However, the capacity of organics to form
'School of Biological Sciences, University of Birmingham, Birmingham, United Kingdom.
127
-------
complexes with copper is reduced as pH declines (Shapiro 1964), so that acidification of surface
waters will increase the availability of copper to aquatic biota. Turnpenny et al. (1987) found
that in mildly acid streams copper and zinc levels were more important determinants of the status
of a fishery than was pH.
A number of authors have produced models to predict the speciation of dissolved copper
in freshwater systems (e.g. Shaw and Brown 1974, Howarth and Sprague 1978, Waiwood and
Beamish 1978). Two factors are important, pH and bicarbonate concentrations. Below pH 5, the
contribution of the cupric ion to the total soluble copper is greatest regardless of total alkalinity
and may in fact be the only form present (Howarth and Sprague 1978). As acidity declines, the
concentration of the cupric ion also falls hi favour of carbonate and hydroxides so that at a pH of
8 to 9, Cu2* is almost entirely absent (Ibid.'). Such analysis is important since only certain copper
species have a toxic effect. Shaw and-Brown (1974) found both the concentration of Cu2+ and
CuCO3 to be determinants of toxicity. Other authors found evidence to indicate that at a given
level of copper, toxicity is best explained by the concentration of Cu2+ (Pagenkopf et al. 1974,
Howarth and Sprague 1978, Waiwood and Beamish 1978, Chakoumakos et al. 1979).
TOXIC ACTION OF COPPER
In common with many other waterborne pollutants and in particular with other heavy
metals and low pH, the primary target of the toxic actions of copper on fish is their gills (Evans
1987). Comprehensive structural damage is caused by severe copper exposure, characterized by
the collapse and even fusing of lamellae, the lifting of lamellar epithelium away from pillar cells,
and swelling of epithelial cells forming ridges in the epithelium (Kirk and Lewis 1993). The
effects of acute exposure to lethal levels of copper on gill histology were described by Wilson
and Taylor (1993a). In addition to ultrastructural damage, Kirk and Lewis (1993) observed an
increase in secretion of mucus and a concomitant swelling of the mucus layer around the gill
which also contained a significant amount of cellular debris.
Such extreme gill damage affects the ability of the gill to function effectively as an
exchange organ. In freshwater, where fishes are hyperosmotic to the surrounding water, copper
causes a net loss of ions such as Na+ and Cl" (Taylor et al. 1995). The increase in gill
permeability, leading to passive efflux of ions, is probably related to the ability of copper to
displace calcium from biological ligands (Nieboer and Richardson 1980). Calcium is known to
be important in controlling the integrity and permeability of the branchial epithelium in fishes
(Potts and Fleming 1971, McDonald and Rogano 1986). Lauren and McDonald (1985)
proposed that the stimulation of ionic effluxes by copper may be a result of increased ionic
permeability of the gill, secondary to the displacement of Ca2+ from paracellular tight junctions,
causing a breakdown hi cell adhesion. At low concentrations of copper, ionoregulatory
disturbances can occur without apparent physical damage to the gills (Ibid). The inhibition of
active NaCl uptake is most likely to be due to the high affinity of the cupric ion for sulphydryl
groups (-SH) on transport enzymes such as Na1", K+-ATPase located in the gill epithelium (Stagg
and Shuttleworth 1982). At higher levels of copper the displacement of intercellular and
128
-------
membrane-bound calcium may lead to a breakdown in branchial cell volume control and
epithelial organization. Indeed, exposure of freshwater trout to 4.9 mmol-1 copper caused severe
gill histopathologies which included cell adhesion failure, swelling, and pillar cell detachment
leading to numerous haematomas (red cells pooling in regions of lamellae) (Wilson and Taylor
1993a, Taylor and' Wilson 1994).
The structural disruption of the gill epithelium in freshwater rainbow trout (Oncorhychus
mykiss) exposed to copper has drastic effects on respiratory gas exchange (Wilson and Taylor
1993a, Taylor and Wilson 1994). As part of its attempt to: compensate for these changes by
increasing oxygen transport, the fish exhibit an increase in haematocrit to values as high as 50%.
This elevated haematocrit, ensuing from reduced plasma volume, red cell swelling and their
mobilization from the spleen, together with the increase in plasma proteins, will contribute to an
increased blood viscosity (Wells and Weber 1991). Catecholamines are released, causing a
vasoconstriction in the peripheral circulation and consequent increased resistance (Milligan and
Wood 1982), so that blood pressure is likely to rise. As a consequence of these physiological
changes, cardiac work will be increased to the extent that death through circulatory failure seems
probable. This likelihood of cardiac failure is increased by the additional stresses of reduced
oxygen supply to the tissues, including the myocardium, and impaired acid-base regulation. The
fish dies because of its integrated responses to the physiological changes induced by the toxic
environment (Taylor and Wilson 1994).
SUB-LETHAL EFFECTS OF COPPER EXPOSURE
Measures of the relative toxicity of a substance to fishes, such as the LC50 test that rely
on killing animals, when used alone as a basis for judging the environmentally 'safe'
concentrations of a pollutant are likely to lead to underestimation of its true impact. Cairns
(1966) stressed the need for caution in the application of results from acute toxicity tests and
indicated the importance of measured changes in oxygen consumption and swimming capacity as
sub-lethal measures of relative toxicity. Swimming performance is an indicator of the ability of
the fish to feed, escape predation and maintain position in a current (Beamish 1978), and is of
particular significance to fishes such as salmonids that migrate upstream to spawn. '
In a study performed by the authors, adult brown trout (Salmo truttd) (300-600 g) were
acclimated to 5 C in winter ("winter trout") and 15 C in summer ("summer trout") in
dechlorinated City of Birmingham tapwater for 2-4 weeks and then in artificial softwater
(Ca2+ 50 Aimol I/1) for another 2-4 weeks. A catheter was inserted into the dorsal aorta of each
fish while under anaesthesia (MS222) to allow blood sampling with minimum disturbance. The
fish were then exposed to a 96-hour episode of copper at pH 5. The copper levels used had been
previously determined by toxicity testing of brown trout to'be the sub-lethal copper
concentrations (SLCC) for these experimental conditions, I e., the highest dose that
caused no mortality during the exposure period. The SLCC was 0.47,umol I/1 Cu2+ at 5 C and
0.08 ^mol L'1 Cu2+ at 15 C. In addition, winter trout were exposed to the summer SLCC
-
129
-------
The effect of copper upon swimming performances was most marked in the winter trout
exposed to 0.47/^mol L"1 Cu2* (Figure la). Only one of six fish tested swam steadily at the
lowest test speed of 0.3 m s"1. The others swam only for 1 or 2 minutes in brief bursts and glides.
Those trout at 5 C that displayed no ability for aerobic exercise, seemed to retain some capacity
for anaerobic 'burst' swimming. This observation strongly suggests the effect of the copper/acid
exposure to be upon aerobic exercise. Below 0.3 m s"1, the trout were able to maintain position
simply using their outstretched pectoral fins and the friction of the bottom of the flume
(cf. Randall 1970).
UCM| (bodylengths s'1)
3 O — » — » NJ
3 In o bi o
1.1,1.1
i: 5-
b>
1 4-
E.
^ 3-
m
Q.
1 2"
X
0
OJ 1 -
3
O
T
_n
T
T
(a)
i
(b)
T
T
I
T
T
X
° I I I I I
Temp'C 5 5 5 15 15
pH 7 5 5 7 " 5
[Cu2*] nmol M - 0.08 0.47 - 0.08
Figure 1. Effects on brown trout of exposure to sub-lethal levels of copper in soft, acid water •
(pH 5): (at) maximum sustainable swimming speed (Ucrit) was about two body lengths per s"1
in control fish (pH7, no copper) at both 5 C and 15 C. Following exposure to 0.08 nmol L"1
copper tie,,., fell to between 1 and 1.5 body lengths s'1; exposure to 0.47yumol'1 at 5 C reducing
swimming speed to less than 0.33 body lengths s'1 (&) oxygen uptake increased with .
temperature and folio whig exposure to copper. (From Beaumont et cil. 1995a.) •.
130
-------
Sub-lethal copper exposure caused damage to gill ultrastructure at all copper and
temperature combinations (Mujallid, et al. In Prep.), but the changes in the winter trout exposed
to 0.47 Mmol L"1 copper were particularly marked, including hyperplasia of epithelial cells and
proliferation of mucocytes in the secondary lamellae, necrosis of chloride cells, and fusion of
neighboring secondary lamellae. Morphometric analysis showed the harmonic mean diffusion
distance to have increased by over three-fold from 3.62 ± 0.42 //m in control fish to 1 1.54 ±
1.63 yum in these fish. While there was no change in the arterial oxygen partial pressure (PaO2)
of these trout at rest, there was a significant decline during exercise, indicating an underlying
diffusional limitation in these fish.
Exposure to 0.08 jumol L"1 copper had less drastic but still significant effects upon
swimming performance, reducing maximum sustainable swimming speed (Ucrit) by 25-50% at
15 and 5 C (Figure la). Ultrastructural changes at the gill were less severe, being characterized
by large vacuoles in gill lamellae and structural deformations, including the curling over the tips
of some lamellae. There was no significant difference in the harmonic mean diffusion distance
of trout exposed to this level of copper and that of control fish. The brown trout exposed to
0.08 ^mol L"1 Cu2+ at pH 5 demonstrated no decline in PaO2 and in no group of copper exposed
fish did the concentration of oxygen in arterial blood (CaO2) fall. In fact, at 15 C the oxygen
content of copper/acid exposed trout rose significantly with exercise. This was achieved through
an increase in haemoglobin concentration ([Hb]), most probably due to the release of
erythrocytes from the spleen, since haematocrit was also elevated. Slight increases in
haemoglobin concentration may also have been responsible for maintaining CaO2 at 5 C.
Plasma lactate concentration was not elevated in fish exposed to sub-lethal levels of
copper and in these trout there is no evidence of a general systemic hypoxia or a failure of
branchial oxygen uptake. While there was no significant increase in plasma lactate in trout
exposed to 0.47 yumol L"1 copper at 5 C, red muscle lactate concentration of this group offish
was significantly elevated at rest to 9.0 ± 0.8 fjxaol g'1 from 5.0 ± 0.7 //mol g'1 at neutral pH.
It has been suggested that haemoconcentration may lead to disruption of oxygen transport and a
local tissue hypoxia (Randall and Brauner 1991, Butler, et al 1992). In addition to an increase in
[Hb], the significant loss of plasma ions may lead to an osmotic loss of plasma water. In the fish
exposed to 0.47 ^mol L"1 Cu2+ at pH 5, there was a very significant rise in plasma protein which
is indicative of an osmotic loss of plasma water. Both factors would elevate blood viscosity
(Bushnell et al. 1 992) in which case effects upon the local circulation of blood to the exercising
muscles may become significant.
Routine oxygen consumption rose in trout exposed' to the SLCC at each acclimation
temperature (Figure Ib). Waiwood and Beamish (1978) also found copper and acid exposure to
elevate the oxygen consumption offish at rest. At 12 C, exposure of fingerling rainbow trout to
0.4 ,umol L"1 Cu2+ at pH 6 increased standard metabolism by 70%. In the same study, Ucrit was
decreased by 40% in comparison to that of control trout in copper free water at pH 7.8. In the
study of Beaumont et al. (1995a), routine oxygen consumption rose by 38% in the summer trout
(Figure Ib). The Ucrit of these trout was decreased by 26 % (Figure la), in almost the same
proportion compared to the increase in oxygen consumption as found by Waiwood and Beamish
131
-------
(1978). However, this is unlikely to be a direct relationship as these relatively small increases in
resting uptake would not limit scope for exercise to a great extent, relative to active metabolism
which can be as much as 12-15 times the resting level (Wood and Perry 1985). Moreover, the
brown trout exposed to 0.08 ywmol L"1 Cu2+ at 5 C did not have an increased metabolic rate at rest
but displayed a decline in swimming performance similar to that of the brown trout exposed to
this level of copper at 15 C.
A number of pressure recordings made from cannulae inserted into the buccal cavity of
some of the winter trout indicated that respiratory frequency may increase by almost 40%
following copper exposure at pH 5. As well as an increase in respiratory frequency, Sellers et al.
(1975) also found ventilation volume to be greater following copper exposure. Jones and
Schwarzfeld (1974) showed that at rest 10% of the oxygen uptake is required to power the
respiratory muscles. This proportion is likely to rise during exercise or hypoxia, when
ventilation rises, as the efficiency of these muscles is relatively low. Thus, an element of the
elevated oxygen consumption of copper-exposed fish at rest is likely to arise from increased
ventilation.
ACCUMULATION OF AMMONIA
Exposure of freshwater rainbow trout to acutely lethal levels of copper caused a large and
progressive increase in plasma ammonia levels. Exposure of brown trout to 0.47 yumol"1 Cu2+ in
acid freshwater (pH 5) for 96 hours caused plasma ammonia to rise from 100 //mol"1 in control
fish (pH 7, no added copper) to 600 ,umol L"1 (Beaumont et al. 1995a). In the same fish, total
ammonia levels in red muscles rose from 17 mmol I"1 kg"1, to 37 mmol"1; values which illustrate
that ammonia produced by tissues is 'trapped' in the relatively acidic intracellular compartment
(Mair et al. In prep.). Exposure of brown trout to 0.08 yumol L"1 copper (pH 5) increased total
plasma ammonia levels to 470 yumol L'1 at 5 C and to 720 /^mol L"1 at 15 C. (Figure 2).
Ammonia accumulation during copper exposure has been observed previously (Lauren
and McDonald 1985, Wilson and Taylor 1993a,b). The mechanisms underlying this
accumulation of copper are unclear. Copper exposure caused no increase on O2 consumption in
seawater rainbow trout (O. myTdss) (Wilson and Taylor 1993a). If the normal relationship
between M02 and Mm3 (Brett and Zala 1975, Heisler 1984) remained constant throughout
exposure to copper, then the elevation of plasma [T^J cannot be explained by any change in
the rate of endogenous ammonia production. Indeed, exposure of common carp (Cyprinus
carpio) to copper levels between 0.22 and 0.84 yumol L"1 had no effect on the rate of ammonia
excretion into soft water at neutral pH-(de Boeek et ah 1995).
Although there is evidence of structural damage to the gills of trout exposed to sublethal
levels of copper, there was no evidence of impairment to oxygen or carbon dioxide gas exchange,
i.e. the fish were not hypoxic or hypercapnic. The branchial permeability coefficient of NH3
is similar to that of CO2 and much greater than that for O2 (Cameron and Heisler 1983).
In addition, the acidity of the test water should enhance 'ammonia trapping1, that is the
132
-------
conversion of excreted NH3, to the impermeable NH/ ions in the boundary layer outside the gills
(Wright et al. 1989). It is possible that local ammonia concentrations are quite different from
those of the bulk water. Ventilation rate increased in copper and acid-exposed brown trout so
that the unstirred boundary layer is likely to be reduced, but may well contain ammonia in a
higher concentration than that of the surrounding water. In addition the mucus layer may trap
ammonia. Handy (1989) found that, even in control conditions, rainbow trout body mucus had a
[NH4+] almost four times that of the ambient water. j
o
E
900
f 800
I 700
.2
^ 600
c 500
o
'13 400
c
CD
o
c
o
300-
200-
.2 100-
1 '
CO
o
o
Sin
|ap
oo~ M
O 3
Sin
So
o
0 Rest D Exercise
Figure 2. Total plasma ammonia concentration (wmol 1-1) of brown trout acclimated to either
5 or 15 C, exposed for 96 hours to either control conditions (pH 7, no copper) or to the stated
copper concentration at pH 5 and sampled either at rest or after exercise to (Ucrit).
(From Beaumont ef al. 1995b.)
However, it is difficult to see how ammonia excretion could be affected by mucus
without it affecting respiratory gas exchange. The conditions of the experiments of Beaumont et
al. (1995a) were seemingly ideal for the passive diffusion of NH3. Particularly as the water had
an acid pH. In contrast, there does appear to have been some ionoregulatory impairment of the
copper/acid exposed rainbow trout, since both [Na+] and [Cl"] decreased and it is known that
copper exposure inhibits ATPases (Lauren and McDonald 1985). These data indicate that
plasma ammonia accumulation during copper exposure is caused by the inhibition of an active
mechanism, which normally contributes significantly to the excretion of ammonia. This could be
active elimination of NH4+ (Wilson and Taylor 1992), though this was recently questioned by
Wilson et al. (1994). Alternatively, it could be the electrogenic proton pump described by Lin
and Randall (1991), which by acidifying the water in the boundary layer close to the gills may
promote ammonia trapping, thus maintaining the outward diffusion gradient for ammonia
(Wright et al. 1989).
133
-------
AMMONIA AND SWIMMING PERFORMANCE
Regression analysis has shown there to be a significant linear relationship between Ucrit
and plasma total ammonia concentration (Figure 3). The coefficient of determination (r2) implies
that almost 70% of the variation hi swimming performance can be explained by plasma
ammonia levels (Beaumont et al. 1995b). There is no evidence from this study that ammonia is a
normal cause of fatigue hi brown trout, as there was no ammonia accumulation in control
animals at Ucrit. Instead, what these data imply is that elevation of systemic ammonia levels due
to copper and acid exposure is the basis of reduced swimming performance, i.e., ammonia causes
fatigue to occur earlier in fish exposed to copper and low pH. A limited pilot experiment (A/=4),
in which ammonium bicarbonate was infused into the dorsal aorta of brown trout caused an
average reduction in Ucrit of 30%. These infusions took place over 24 hours, shorter infusions
were without effect, so that infusions of 96-hour duration may reciprocally produce changes in
swimming performance more similar magnitude to those that occurred following 96 hours of
copper exposure. Ammonia and ammonium ions have a number of possible biochemical and
physiological effects that may influence swimming performance either by affecting the metabolic
status of the muscle or by interfering with central or peripheral nervous activity, transmission at
the neuromuscular junction, excitation/contraction coupling or muscle electrophysiology (Taylor
etal. 1995).
2.5
2.0
_. 1.5
t
tn
s 1.0
0.0
4-
0
Y = -0.002* + 2.089
/?z = 0.670
200 400 600 800
Plasma ammonium ion concentration (|imol I"')
1000
1200
Figure 3. The relationship between ammonium ion concentration in arterial plasma and
swimming performance (U^ of individual brown trout acclimated to either 5 or 15 C and
exposed for 96 hours to either control conditions (pH 7, no copper) or to copper
concentrations of either 0.08 or 0.47 yumol-l at pH 5. (From Beaumont et al. 1995b.)
134
-------
Elevation of plasma ammonia concentration is not a phenomenon related to copper
exposure alone. For example, low pH (Day and Butler 1996), high pH (Ye and Randall 1991)
and aluminium (Booth et al. 1988) elevate plasma ammonia, and these pollutants also decrease
swimming performance (Ye and Randall 1991, Butler et dl. 1992, Wilson and Wood 1992).
If a relationship between Ucrit and ammonia can be confirmed, perhaps by direct manipulation of
plasma ammonia through infusion of ammonium salts, then this single effect may prove, at least
partly, to explain the reduced swimming performance offish arising from their exposure to
sublethal levels of a variety of environmental pollutants. Factors affecting swimming
performance are likely to have overiding effects on the survival offish populations, as discussed
earlier, so that our views on the relative toxicity of ammonia may need revision.
ACKNOWLEDGEMENTS
The authors acknowledge the technical help and advice of Norman Day.
REFERENCES
Beamish, F. W. H. 1978. Swimming capacity. In: Fish Physiology, Vol. 7. W. S. Hoar and
D. J. Randall (Eds.) New York, London: Academic Press, pp. 101-187.
Beaumont, M. W., P. J. Butler, and E. W. Taylor. 1995a. Exposure of brown trout, Salmo trutta,
to sub-lethal copper concentrations and its effect upon sustained swimming performance.
Aquatic Toxicology 33:45-63.
Beaumont, M. W., P. J. Butler, and E. W. Taylor. 1995b. :Plasma ammonia concentration in
brown trout (Salmo trutta) exposed to acidic water and sublethal copper concentrations and
its relationship to decreased swimming performance. Journal of Experimental Biolosv
198:2213-2220.
Booth, C. E., D. G. McDonald, B. P. Simons, and C. M. Wood. 1988. Effects of aluminium arid
low pH on net ion fluxes and ion balance in the brook trout (Salvelinus fontinalis). Canadian
Journal of Fisheries and Aquatic Science 45:1563-1574.
Brett, J. R., and C. A. Zala. 1975. Daily pattern of nitrogen excretion and oxygen consumption
of sockeye salmon (Oncorhynchus nerkd) under controlled conditions. Journal of the
Fisheries Research Board of Canada 32:2479-2486.
Bushnell, P. G., D. R. Jones, and A. P. Farrell. 1992. Thelarterial system. In: Fish Physiology,
Vol. 12 A. W. S. Hoar, D. J. Randall, and A. P. Farrell (Eds.) New York, London:
Academic Press, pp. 89-139.
135
-------
Butler, P.J., N. Day, and K. Namba. 1992. Interactive effects of seasonal temperature and low
pH on resting oxygen uptake and swimming performance of adult brown trout (Salmo truttd).
Journal of Experimental Biology 165:195-212.
Cairns, J. 1966. Don't be half safe—the current revolution in bioassay techniques. Engineering
Bulletin Purdue University Proceedings of the 21 Industrial Waste Conference, pp. 559-567.
Cameron, J.N., and N. Heisler. 1983. Studies of ammonia in the rainbow trout: physico-
chemical parameters, acid-base behavior and registry clearance. Journal of Experimental
Biology 105:107-125.
Chakoumakos, C., R. C. Russo, and R. V. Thurston. 1979. Toxicity of copper to cutthroat trout
(Salmo clarki) under different conditions of alkalinity, pH and hardness. Environmental
Science and Technology 13:213-219.
Day N., and P.J. Butler. 1996. Environmental acidity and white muscle recruitment during
swimming in the brown trout (Salmo truttd). Journal of Experimental Biology 199:1947-
1959.
de Boeck, G., H. de Smet, G.E. Nilsson, and R. Blust. 1994. Oxygen consumption, nitrogen
excretion, brain monoamines and brain energy metabolites in copper exposed carp.
Physiologist 37A:78.
Evans, D. H. 1987. The fish gill: site of action and model for toxic effects of environmental
pollutants. Environmental Health Perspectives 71:47-58.
Handy, R. D. 1989. The ionic composition of rainbow trout body mucus. Comparative
Biochemistry and Physiology 93A:571-575.
Heisler, N. 1984. Role of ion transfer processes in acid-base regulation with temperature
changes in fish. American Journal of Physiology 246:R441 -451.
Howarth, R. S., and J. B. Sprague. 1978. Copper lethality to rainbow trout in waters of various
hardness and pH. Water Research 12:455-462.
Jones, D. R., and T. Schwarzfeld. 1974. The oxygen cost to the metabolism and efficiency of
breathing in trout (Salmo gairdneri). Respiration and Physiology 21:241 -254.
Kirk, R.S., and J. W. Lewis. 1993. An evaluation of pollutant induced changes hi the gills of
rainbow trout using scanning electron microscopy. Environmental Technology 14:577-585.
Lauren, D.J., and D. G. McDonald. 1985. Acclimation to copper by rainbow trout Salmo
gairdneri: Biochemistry. Canadian Journal of Fisheries and Aquatic Science 44:105-111.
136
-------
Lin, H., and D. J. Randall. 1991. Evidence for the presence of an electrogenic proton pump on
the trout gill epithelium. Journal of Experimental Biology 161:119-134.
Mair, J., E.W. Taylor, and P.J. Butler. (In preparation). Tissue metabolism in brown trout
exposed to sublethal levels of copper.
McDonald, D. G., and M. S. Rogano. 1986. Physiological responses to stress in lake trout:
effects of water hardness and genotype. Transactions of the American Fisheries Society
122:1146-1155.
Milligan, C. L., and C. M. Wood. 1982. Disturbances in haematology, fluid volume,
distribution and circulatory function associated with low environmental pH in the rainbow
trout, Salmo gairdneri. Journal of Experimental Biology 99:397-415.
Moore, J.W., and S. Ramamoothy. 1984. Heavy Metals in Natural Waters. New York:
Springer-Verlag.
Mujallid, M.S.I., E.W. Taylor, and P.J. Butler. (In preparation). Changes in gill morphometrics
and fine structure of respiratory epithelial cells of brown trout on exposure to sublethal levels
of copper.
Nieboer, E., and D.H.S. Richardson. 1980. The replacement of the nondescript term 'heavy
metals' by a biologically and chemically significant classification of metal ions.
Environmental Pollution 1:3-26.
Pagenkopf, G. K., R. C. Russo, and R. V. Thurston. 1974. Effects of complexation on toxicity
of copper to fishes. Journal of the Fisheries Research Board of Canada 31 -.462-465.
Potts, W. T. W., and W. R. Fleming. 1971. The effect of environmental calcium and ovine
prolactin on sodium balance in Fundulus kansae. Journal of Experimental Biology 54:63-75.
Randall, D. J., and C. Brauner. 1991. Effects of environmental factors on exercise in fish.
Journal of Experimental Biology 160:113-126.
Sellars, C. M., A.-G. Heath, and M. L. Bass. 1975. The effect of sublethal concentrations of
copper and zinc on ventilatory activity, blood oxygen and pH in rainbow trout (Salmo
gairdneri). Water Research 9:401-408.
Shapiro, J. 1964. Effect of yellow organic acids on iron and other metals in water. Journal of
the American Water Works Association 56:1062-1082.
Shaw, T. L., and V. M. Brown. 1974. The toxicity of some forms of copper to rainbow trout
Water Research 8:377-382.
137
-------
Sposito, G. 1986. Distribution of potentially hazardous trace metals. In: Trace Metals in
Biological Systems, Vol. 20. H. Sigel (Ed.) New York, Basel: Marcel Dekker Inc. pp. 1-20.
Stagg, R. M., and T. J. Shuttleworth 1982. The effects of copper on ionic regulation by the gills
of the seawater adapted flounder (Platychthys flesus L.). Journal of Comparative Physiology
149:83-90.
Taylor, E. W., and R. W. Wilson. 1994. The physiological responses of rainbow trout to copper
and ammonia: mechanisms of toxicity and ammonia excretion. Fish Physiology, Toxicology
and Water Quality Management. Proceedings of Biennial International Symposium,
Nanjing, PRC, EPA/600/R-94/138. pp. 36-46.
Taylor, E.W., M. W. Beaumont, P. J. Butler, J. Mair, and M. S. I. Mujallid. 1995. Lethal and
sublethal effects of copper upon fish: a role for ammonia toxicity? In: Toxicology of Aquatic
Pollution. E. W. Tayler (Ed.) Society of Experimental Biology, Seminar Series. Cambridge
University Press. Vol.57: 85-114.
Turnpenny, A. W. H., K. Sadler, R. J. Aston, A. G. P. Milner, A.G.P., and S. Lynam. 1987. The
fish populations of some streams in Wales and northern England in relation to acidity and
associated factors. Journal of Fish Biology 31:415-434.
Waiwood, K. G., and F. W. H. Beamish. 1978. Effects of copper, pH and hardness on the
critical swimming speed of rainbow trout (Salmo gairdneri Richardson). Water Research
12:611-619.
Wells, R. M. G., and R. E. Weber. 1991. Is there an optimal haematocrit for rainbow trout,
Oncorhynchus mykiss (Walbaum)? An interpretation of recent data based on blood viscosity
measurements. Journal of Fish Biology 38:53-65.
Wilson, R. W., and E. W. Taylor. 1992. Transbranchial ammonia gradients and acid-base
responses to high external ammonia concentration in rainbow trout (Oncorhynchus mykiss)
acclimated to different salinities. Journal of Experimental Biology 166:95-112.
Wilson, R. W., and E. W. Taylor. 1993a. The physiological responses of freshwater rainbow
trout, Oncorhynchus mykiss, during acutely lethal copper exposure. Journal of Comparative
Physiology 163B:38-47.
Wilson, R. W., and E. W. Taylor. 1993b. Differential responses to copper in rainbow trout
(Oncorhynchus mykiss) acclimated to sea water and brackish water. Journal of Comparative
Physiology 163B:239-246.
Wilson, R. W., and C. M. Wood. 1992. Swimming performance, whole body ions and gill
aluminum accumulation during acclimation to sublethal aluminium in juvenile rainbow trout
(Oncorhynchus mykiss). Fish Physiology and Biochemistry 10:149-159.
'13*
-------
Wilson, R. W., P. M. Wright, S. Munger, and C. M. Wood. 1994. Ammonia excretion in
freshwater rainbow trout (Oncorhynchus mykiss) and the importance of gill boundary layer
acidification: lack of evidence for Na+/NH4 exchange. Journal of Experimental Biology
191:37-58.
Wood, C. M., and S. F. Perry. 1985. Respiratory, circulatory and metabolic adjustments to
exercise in fish. In: Circulation, Respiration and Metabolism. R. Gilles (Ed.) Heidelberg:
Springer-Verlag. pp. 1-22.
i
Wright, P. A., D. J. Randall, and S. F. Perry. 1989. Fish gill boundary layer: A site of linkage
between carbon dioxide and ammonia excretion. Journal of Comparative Physiology
158:627-635.
Ye, X., and D. J. Randall. 1991. The effect of water pH on swimming performance in.rainbow
trout (Salmo gairdneri R.). Fish Physiology and Biochemistry 9:15-21.
139
-------
-------
SUBLETHAL EFFECTS ON RAINBOW TROUT OF CHRONIC EXPOSURE
TO MIXTURES OF HEAVY METALS
Milda-Zita Vosyliene1 and Gintaras Svecevicius1
ABSTRACT
A study has been conducted on the sublethal effects of model mixtures of heavy metals to
rainbow trout (Oncorhynchus mykiss). Two kinds of model mixtures were formed based on data
of the amounts and relative proportions of five common heavy metals in discharges from the
cities of Vilnius and Kaunas into the Neris and Nemunas Rivers. The long-term (3 months)
effects of these two kinds of mixtures on behavioral, respiratory, hematological, and
morphological indices have been estimated in laboratory studies, and comparisons have been
made on the sensitivity of the parameters measured. The suite of parameters suggested to
determine and evaluate the presence of heavy metals in water in laboratory experiments are:
respiratory rates, tissue-somatic indices, hemoglobin, glucose and blood smear counts, and
avoidance-preference reactions.
INTRODUCTION
Heavy metals are common pollutants in inland waters of Lithuania, and these
continuously impact individual fish, their communities, and their populations. Metals are often
found in Lithuanian rivers downstream from many of the larger, and some of the smaller, cities
and towns due to industrial activities. The most common metals are chromium, copper, iron,
nickel, and zinc which are byproducts of electrodeposition processes, and these escape into waste
waters. Bioindication data obtained under field conditions often show the final effect of water
pollution on hydrobionts and ecosystems, but these data do not reveal the processes going on
within the hydrobionts under acute or chronic pollution which may later cause biocenotic
changes (Landner 1989). Toxicants in aquatic environments are frequently mixtures of various
chemicals and their derivatives, so it is impossible to create all water variables and toxicant
concentrations that are found in water bodies. Even so, a simplified but still realistic approach to
studying effects of complex pollutants on hydrobionts is to conduct toxicity tests using model
mixtures of the principal pollutants (Landner 1989, Cuvin-Azalar and Azalar 1993).
'Institute of Ecology, Vilnius, Lithuania.
141
-------
Few investigations have been performed on short- and long-term effects of heavy metal
mixtures on fishes (Spehar et al. 1976, Kazlauskiene e£ a/. 1993, Vosyliene et al. 1993). Heavy
metals may stress a fish and induce various physiological responses, or as toxicants they may
affect fish survival and growth (Eaton 1973, Finlayson and Verue 1982). A combination of
various biological activity indicators (physiological, morphological, and behavioral) can provide
information about the exposure offish to specific chemicals, and these may serve as early
warning signals of deleterious effects on organisms and populations (Adams et al. 1990).
In a previous article we reported on the acute toxiciry of two kinds of model mixtures of
heavy metals such as mentioned above (Vosyliene et al. 1994). The ratio of heavy metal
concentrations in the mixtures studied corresponded to the concentration ratio of these metals in
wastewaters of the cities of Vilnius discharged into the Neris River and of Kaunas discharged
into the Nemunas River, during 1990-91. The aim of the present study was to determine the
sublethal effects of these same model mixtures to rainbow trout (Oncorhynchus mykiss) after
long-term exposure.
MATERIALS AND METHODS
The tests were conducted during January through March 1993. Rainbow trout used for
the tests (12-16 cm total length and 25-50 g total weight) were obtained from the £eimena
hatchery. The fish were kept in 2000-L holding tanks supplied with flow-through aerated
artesian water. Chemical analysis of the water (Table 1) was performed in the Hydrochemical
Laboratory of the Environmental Protection Ministry. The test fish were transferred from the
holding tanks to 120-L aquaria in groups of 16 individuals and were kept until acclimation in the
new medium, i.e., until they started swimming freely and feeding well. The water was changed
three times in a week (on Monday, Wednesday, and Friday) after the fish were fed. Feeding and
behavior of fish were observed.
Table 1. Chemical and physical characteristics of experimental water (All values are mg/L
unless otherwise noted).
PH
Dissolved oxygen
Temperature (C)
Hardness (as CaCO3)
Alkalinity (as HCO/)
Suspended solids
BOD5
Mg2t
Ca2*
SO 2"
NH*
NO/
NO/
P043"
8.0 (7.9-8.1)
11(10.5-11.5)
10 (9.5-10.5)
284 (271-296)
244
2.7
0.000
16.0
64.0
15.0
0.13
O.OOO
O.OOO
0.032
Total phosphorus
Total nitrogen
cr
Phenols
Oil hydrocarbons
Fe
Mn
Zn
Cu
Cr
Ni
Cd
Pb
0.044
0.3
1.9
O.OOO
O.OOO
0.56
44.0
0.0109
0.0031
0.0014
0.0033
0.00004
O.OOO
142
-------
-•<' -• The two different heavy metal model mixtures studied Were a Vilnius variant and a
Kaunas variant. The formation of these model mixtures was carried out based on the available
. analytical data of the amounts of five representative heavy metals in the annual discharge of
.was|ewater from the cities qfVilnius and Kaunas .during 1990-1991. The. metals were , .
chromium, copper, iron,,nickei, and zinc. The following chemically pure substances were used
in making the model mixture solutions: K2CrO4, CuSO4-5H2O, NiSO4-7H2O, FeSO4 (anhydrous),
ZnSO4-7H2O. During the 3-month test period, respiratory and behavioral indices were registered,
and at the end of exposure, hematological and morphological parameters were measured.
Respiratory Indices. Gill ventilation frequency (N/minute) and "coughing" rate
(back-flushing of gills, N/minute) were measured at the beginning of experiments and after each
1-month period of exposure. They were measured during 3-minute periods for each fish
individually, from which the mean value for 10 fish was calculated.
Hematological Parameters. Erythrocytes (Er), hemoglobin (Hb), hematocrit (Hct) and
their derivatives, mean corpuscular volume (MCV), and mean corpuscular hemoglobin
concentration (MCHC) were determined using methods and techniques included in Svobodova
and Vykosova (1991). Lymphocyte counts were made from blood smears. No distinction was
made in counting large and small lymphocytes, and the absolute counts of lymphocytes were
estimated as described by Casillas and Smith (1977). Differences in blood smears of control and
exposed fish were noted according Golovina and Trombickij (1989). Glucose concentration was
determined using a method with "EKSAN-G" (Kulys et al. 1989).
Morphological Indices. Body lengths, weights, and tissue-somatic indices of the fish
were determined at the end of the experiments. Tissue-somatic index was calculated using the
formula K=WtAV2 • 100, where W, = weight of tissue (g), and W2 = weight offish (g).
Avoidance-Preference Reactions. Fish were tested according to the scheme described by
Svecevicius (1994). A plastic gradient chamber (1500 X 600 X 300 mm) with two paralied
water sources was used, in which a sharp boundary was formed between polluted and non-
polluted water. The total water flow through the chamber was 6 L/minute. The fish were tested
in groups of 10 individuals, and each group placed into the gradient chamber for acclimation
when water was not flowing through the chamber. The control fish were placed into the non-
polluted water, and pre-exposed fish were placed into the same concentration of heavy metal
mixture (polluted water) to which they had been adapted. After an acclimation period of
0.5-1.5 hours, fish began to swim freely exploring the chamber. This was followed by a control
period (10 minutes) during which the positions of the fish were registered every 10 seconds hi
that section of the chamber which was to be polluted during the experiment. After the control
period, water was allowed to flow through the chamber; heavy metal mixture solution was
introduced in one section of the chamber and pure water was introduced through the other.
The concentration of heavy metal mixture corresponded tq the concentration to which pre-
exposed fish had been adapted. Two variants of heavy metal mixtures were tested on control
fish. After introduction of pollutants to the chamber for 30 minutes, a test period of 10-minutes
143
-------
duration was conducted. During this test period the positions of the fish were registered in the
same way as during the control period. The character and intensity of behavioral reaction was
estimated by the following formula:
Reaction index (RI) = 50(2-^/1^,
where NT = number offish in polluted section at registration moment during test-period,
and NK = mean number offish hi polluted section during control period. The index value 100
indicated maximal avoidance, 50 indicated indifference, and 0 indicated maximal preference.
The significance of all the data obtained was determined by use of Students' t-test.
RESULTS
Respiratory Indices. The exposure of test fish to both of the model heavy metal mixtures
caused significant decrease in their gill ventilation frequency and significant increase in their
"coughing" rate during all periods of the experiment (p < 0.05) (Table 2). The mean values of
respiratory parameters did not fluctuate widely except after 60 days. The slight increase in gill
ventilation Frequency and "coughing" rate was determined after this period in both control and
pre-exposed fish.
Table 2. Effect of exposure to heavy metal modelmixtures on gill ventilation frequency and
coughing rate in rainbow trout, N/min (X±SE, N=10).
Exposition
(days)
0
30
60
90
Control
GVF
94.5±1.0
96.0±1.5
107±2.6
94.1±1.4
CR
2.510.37
0.4±0.16
2.1+0.54
0.5±0.22
Vilnius variant
GVF
94.7±1.0
80.9±1.7*
90.1±1.8*
72.0±1.7*
CR
2.3±0.33
0.8±0.20
3.5±0.50
1.3±0.15
Kaunas variant
GVF
94.3±1.0
82.7+1.8*
94.8±2.5*
73.3+1.4*
CR
2.5±0.43
0.8+0.16
4.4±0.40
1.1+0.18
GVF = gill ventilation frequency; CR = coughing rate; * = significantly different from control (P^O.05).
Hematological Parameters. The exposure of test fish to both of the model heavy metal
mixtures did not effect their erythrocyte or lymphocyte counts, or hematocrit of their blood
(Table 3). Meanwhile, reduced hemoglobin, MCHC, and increased MCV hi the test fish was
determined. Glucose concentration was increased hi the blood of pre-exposed fish. Changes
were determined in blood smears of 38.7% of the pre-exposed fish as compared to control fish.
It was established that there was an increased count of "amitotic" erythrocytes, changed forms
and sizes of erythrocytes, and an increased amount of erythrocytes of old removal forms (stages
2, 3, and 4) significantly different from control (p < 0.05).
144
-------
Table 3. Effect of 90-day exposure to heavy metal model mixtures on hematologi.cal parameters
in rainbow trout. (X+SE, N=l6)
Parameters
Erythocytes (T/L)
Hemoglobin (g/L)
hematocrit (L/L)
MCHC (L/L)
MCV (fl)
Lympohocytes (G/L)
Glucose (mmol/L)
Control
1.14+0.08
106.0±3.33
0.38±0.02
0.27910.02
330±12.0
20.51±2.04
2.5510.13
Fish test group
Vilnius variant
L06±0.02
88.8±2.84*
0:41+0.02
0.216±0.01*
389120.0*
24.7+1.9
3.85±0.13*
Kaunas variant
1.05+0.05
90.213.47*
0.41+0.02
0.222±0.01
391±15.7*
25.4+1.9
3.00+0.12*
: = significantly different from control (P<0.05).
Morphological Parameters. No significant differences were determined between total
length and body weight values of control fish and fish pre-exposed to heavy metal mixtures
(Table 4). However, exposure offish to heavy metal mixtures did cause significant decrease in
gill-somatic index. Decreased liver and spleen-somatic indexes of pre-exposed fish were also
estimated. However, significantly lower liver-somatic index was determined only in fish pre-
exposed to the model mixture of the Vilnius variant, and significantly decreased spleen-somatic
index was observed in fish pre-exposed to the Kaunas mixture variant.
Table 4. Effect of 90-day exposure to model heavy metal mixtures on length, weight, and tissue-
somatic indexes in rainbow trout. (X+SE, N=16)
Test group
Control
Vilnius variant
Kaunas variant
Length (cm)
14.4+0.3
13.9+0.3
13.610.3
Weight (g)
42.6+2.9
41.7+2.2
40.6+1.9
GUI
3.02+OI.08
2.34+0.18*
2.5010.13*
Somatic index (%)
Liver
0.92+0.03
0.7310.01*
0.8510.04
Spleen
0.1710.03
0.1410.03
0.1310.01*
= significantly different from.control (P<0.05).
The control rainbow trout significantly avoided both of the heavy metal model mixtures
(Table 5). The intensity of avoidance reaction was very high and the reaction index reached
maximal levels. Fluctuations of reaction intensity were low. Pre-exposed fish significantly
preferred both heavy metal model mixtures solutions to which they had been adapted. The
intensity of preference reaction was very high, with low intensity of fluctuations. No external
145
-------
symptoms of irritation or intoxication during contact of pre-exposed fish with heavy metal model
mixtures solutions were observed. Behavior of pre-exposed fish was normal and they showed
mean locomotor activity. The fish that were not pre-exposed reacted very intensely during their
contact with the metal mixtures solutions, their locomotor activity increased rapidly in polluted
water, and their escape from this area was accompanied with deep "coughing" movements.
Table 5. Preference-avoidance reactions of control and pre-exposed rainbow trout to heavy metal
model mixtures (Reaction indices, X±SE, N=60)
Exposition (days)
Fish test group
Heavy metal model mixtures
30
60
90
Control
Pre-exposed
Control
Pre-exposed
Control
Pre-exposed
Vilnius variant
98.5+0.3
14.0±3.6
95.3+1.3
17.5+5.5
94.5±0.8
16.2±5.0
Kaunas variant
99.1±0.2
11.4±4.2
98.2+0.2
14.7±4.4
95.3+0.2
10.1±6.6
AH values significantly different from 50 (P^O.05).
DISCUSSION
The need for sensitive and reliable bioindicators for assessing the impact of pollution on
hydrobionts has created significant interest in pollution-caused physiological and biochemical
changes in fish (Adams et aL 1990, Stein et al. 1992). In the present study, a set of responses at
various biological levels were studied in rainbow trout pre-exposed to two kinds of model heavy
metal mixtures. The toxic effect of these mixtures was established at first to the respiratory
system; a significant decrease hi the gill ventilation frequency of pre-exposed fish and an
increased "coughing" rate was registered after 1 month of exposure. Data obtained indicate high
sensitivity of respiratory indices of rainbow trout to heavy metal mixtures, and confirms the
suitability of respiratory index use for evaluation of toxic effect of heavy metals in both short-
term and chronic experiments (Drammond et al. 1973, Drummond et dl. 1974, Hughes 1986,
Drummond 1987). No significant differences were determined in the growth (total length and
body weight) of control and pre-exposed fish after 3-months exposure; all fish looked healthy
and behaved normally. However, evaluation offish tissue-somatic indices revealed toxic action
of heavy metals on various tissues.
146
-------
Gills are one of the major target-organs of pollutants due to their vulnerable external
position and large surface area. They perform the physiological tasks of respiration,
osmoregulation, and acid-base balance and therefore functional impairment by pollutants can
significantly damage the health offish (Hughes 1986, Kirk and Lewis 1993). The study of Kirk
and Lewis (1993) demonstrated that damage to gills by copper was characterized by lamellar
fusion, and the formation of ridges in the epithelium due to swelling of epithelial cells.
We assume that the heavy metal mixtures we tested induced similar changes in the gills of our
pre-exposed fish. The liver-somatic index, reflecting both the metabolic energy demands and
nutritional status, can be considered as a general health indicator, sensitive to environmental
contaminants (Adams et al 1990). Data of the present study, that gill, liver and spleen-somatic
indices of pre-exposed fish tended to be lower or were significantly lower as compared with
control fish, confirm the applicability of these parameters for the evaluation of toxic effects of
heavy metal mixtures to rainbow trout.
' !
The validity of use of the hematological status of an organism as a biological indicator
has been established and various hematological parameters have been shown as very sensitive to
intoxication by trace metals and organic microcontaminants (Everaats et al 1993). However, in
the present study only some hematological parameters as hemoglobin and derivative parameters,
MCHC and MCV, indicated the toxic effect of heavy metal mixtures on fish. MCHC expresses'
the concentrations of hemoglobin in unit volume of erythrocytes and additionally confirms the -
decreasing of hemoglobin concentration in erythrocytes; MCV increase indicates the swelling of
erythrocytes in pre-exposed fish. Changes in blood smears revealed the toxic effect of mixtures
on fish. i
Ziteneva et al. (1989) have reported inhibition of erythrocytes production, splitting of
erythrocytes cytoplasm, and an increased amount of erythrocyte "nuclei shadows" were observed
in fish from polluted water bodies. A typical response to an acute stress on fish includes a rapid
rise in glucose concentration. High glucose concentration results from glycogenolysis initiated
by catecholamines, while sustained elevation of serum glucose in long-term experiments are
maintained by cortisol-stimulated glucogenesis (Schreck and Lorz 1978). In the present study,
elevated glucose concentration was established in test fish after 3 months exposure to heavy
metal mixture, and serves as an indicator of chronic stress.
The data obtained from avoidance-preference tests Ejre in agreement with some of the
literature data concerning behavioral reactions offish pre-exposed to pollutants. Anestis and
Neufeld (1986) have reported about rainbow trout pre-exposed to sublethal concentrations of
chromium for 7-20 weeks. Control fish showed very low avoidance thresholds to chromium,
while avoidance thresholds of pre-exposed fish increased linearly with the level of pre-exposure.
Similarly, Myllyvirta and Vuorinen (1989) reported that vendace (Coregonus albuld) avoided
bleached kraft mill effluent (BKME), although fish pre-exposed to BKME for 1 week preferred
contaminated water. Lake whitefish (Coregonus clupeaformis) pre-exposed to cadmium for
3 weeks showed significant attraction to the water containing sequentially increasing sublethal
cadmium concentrations (McNicol and Scherer 1993).
147 '.
-------
The mechanism of such fish behavior is not clear. But it is no doubt that such fish
reactions are directly connected with changes in fish chemoreceptor sensitivity to pollutants. In
natural environments such fish behavior can be determined as a "physiological trap" and can
effect fish migrations and distribution if fish were pre-exposed to specific chemical substances
early in their development Thus, the study of avoidance-preference reactions offish that were
not pre-exposed may not fully reflect how natural fish populations will react to the presence of
pollutants.
Our findings suggest that a range of bioindicators including fish respiratory indices,
tissue-somatic indices, hematological parameters (hemoglobin and derivatives MCHC, MCV),
glucose, blood smears, and avoidance-preference reactions have value to determine and evaluate
the presence of heavy metal mixture in water tested in laboratory experiments.
REFERENCES
Adams, S.M., L.R. Shugart, G.D. Soutworth and D.E. Hinton. 1990. In: Biomarkers of
Environmental Contamination. J.F. McCarthy and L.R. Shugart (Eds.) Lewis Publishers,
Inc., Boca Raton, Florida, USA, pp. 333-353.
Anestis, I., and R.J. Neufeld. 1986. Avoidance-preference reactions of rainbow trout (Salmo
gairdneri) after prolonged exposure to chromium (VI). Water Research 20:1233-1241.
Casillas, E., and L.S. Smith. 1977. Effect of stress on blood coagulation and haematology in
rainbow trout (Salmo gairdneri). Journal of Fish Biology 10:481-491.
Cuvin-Azalar, L.A., and E.V. Azalar. 1993. Effects of long-term exposure to a mixture of
cadmium, zinc and inorganic mercury on two strains of Tilapia Oreochromys niloticus (L.).
Bulletin of Environmental Contamination and Toxicology 50:891-895.
Drunimond, R.A. 1987. Application of behavioral toxicity syndromes for characterizing effluent
and water quality. In: Society of Environmental Toxicology and Chemistry, Eighth Annual
Meeting. Environmental Risk: Recognition, Assessment and Management. Pensacola,
Florida, 138 pp.
Drummond, R.A., G.F. Olson, and A.R. Battermann. 1974. Cough response and uptake of
mercury by brook trout Salvelinusfontinalis, exposed to mercuric compounds at different
hydrogenion concentrations. Transactions of the American Fisheries Society 2:244-249.
Drummond, R.A., W.A. Spoor, and G.F. Olson . 1973. Some short-term indicators of sublethal
effects of copper on brook trout Salvelinusfontinalis. Journal of the Fisheries Research
Board of Canada 30:698-701.
148
-------
Eaton, J.G. 1973. Chronic toxicity of a copper, cadmium and zinc mixture to the fathead
minnow (Pimephalespromelas Rafmesque). Water Research 7:1723-1736.
Everaats, J.M., L.R. Shugart, M.K. Gustin, W.E. Hawkins, and W.W. Walker. 1993. Biological
markers in fish: DNA integrity hematological parameters and liver somatic index. Marine
Environmental Research 35:101-107.
Finlayson, B.J., and K.M. Verue. 1982. Toxicities of copper, zinc and cadmium mixtures to
juvenile chinook salmon. Transactions of the American Fisheries Society 111:645-650.
Golovina, N.A., and I.D. Trombickij. 1989. Haematology of pond fishes. Kisinev, 158 pp. (In
Russian)
Hughes, G.M. 1986 The dimension of the fish gills in relation to their function. Journal of
Experimental Biology 45:177-195.
Kazlauskiene, N., A. Burba, and G. Svecevicius. 1994. Acute toxicity of five galvanic heavy
metals to hydrobionts. Ekologija 1:33-36.
Kirk, R.S., and J.W. Lewis. 1993. An evaluation of pollutant induced changes hi the gills of
rainbow trout using scanning electron microscopy. Environmental Technology 14:577-585.
Kulys, J., V. Laurinavicius, M. Peslekiene, and M. Gureviciene. 1989. Metabolite determination
in foodstuffs by enzyme analyzer. Biotechnology and Applied Biochemistry 11:149-154.
Kumari, K., and V. Banerjee. 1986. Effect of sublethal toxicity of zinc, mercury and cadmium
on peripheral haemogram hi Anabas testudineus (Bloch). Uttar Pradesh Journal of Zoology
6:241-250. \
Landner, L. (Editor). 1989. Chemicals in aquatic environment. Advanced hazard assessment.
Springer-Verlag, Berlin, Heidelberg, 415 pp. '
Larsson, A., C. Haux, M.-L. Sjobeck, and L. Goar. 1984. Physiological effects of on additional
stressor on fish exposed to a simulated heavy-metal-containing effluent from a sulfide ore
smeltery.- Ecotoxicology and Environmental Safety 8:118-128.
Laurinavicms, V.A., S.A. Firantas, B.S. Kurtinaitiene, V.V. Gureviciene, G.J. Baltakis, G.L.
Kairys, P.S. Vasiliauskas. 1987. Estimation of glucose in blood using express-analyser
"EKSAN-G". Laboratornoje delo 11:844-849. (In Russian).
McNicol, R.E., and E. Scherer. 1993. Influence of cadmium pre-exposure on the preference -
avoidance responses of lake whitefish (Coregonus clupeafarmis) to cadmium. Archives of
Environmental Contamination and Toxicology 25:36-40.
149 :
-------
Myllyvirta, T.P., and PJ. Vuorinen. 1989. Avoidance of bleached kraft mill effluent by pre-
exposed Coregonus albula L. Water Research 23: 1219-1227.
Schreck, C.B., and H.W. Lorz. 1978. Stress response of coho salmon (Oncorchynchus kisutch)
elicited by cadmium and copper and potential use of cortisol as an indicator of stress. Journal
of the Fisheries Research Board of Canada 35:1124-1129.
Spehar, R.L., E.N. Leonard, and D.L. DeFoe. 1976. Chronic effects of cadmium and zinc
mixtures on fiagfish (Jordanellafloridae). Transactions of the American Fisheries Society
107:354-360.
Stein, J.E., T. Collier, W.L. Reichert, E. Casillas, T. Horn, and U. Varanasi. 1992. Bioindicators
of contaminant exposure and sublethal effects: studies with benthic fish in Puget Sound
Washington. Environmental Toxicology and Chemistry 11:701-714.
Sveceviclus, G. 1994. Avoidance reaction to pollutants by vimba under laboratory and field
conditions. In: Environmental Studies in the Nemunas River Basin, Lithuania. R.V.
Thurston, (Ed.) Montana State University Fisheries Bioassay Laboratory Technical Report.
94-1, Bozeman, Montana USA, pp. 115-123.
Svobodova, Z., and B. Vykusova (Eds.). 1991. Diagnostics, Prevention and Therapy of Fish
Diseases and Intoxications. Vodnany, Chechoslovakia, 270 pp.
Vosyliene, Z., G. Svecevicius, and S. Syviene. 1994. Toxicity of factory wastewaters and heavy
metal solutions to rainbow trout. In: Environmental Studies in the Nemunas River Basin,
Lithuania. R.V. Thurston, (Ed.) Montana State University Fisheries Bioassay Laboratory
Technical Report. 94-1, Bozeman, Montana, USA, pp. 107-115.
Ziteneva, L.D., T.G. Poltavceva, and O.A. Rudnickaja. 1989. Atlas of Normal and
Pathologically Changed Fish Blood Cells. Rostov-na-Donu, 112 p. (In Russian).
150
-------
DERIVING FRESHWATER QUALITY CRITERIA FOR 2,4-DICHLOROPHENOL
FOR PROTECTION OF AQUATIC LIFE IN CHINA
Hongjun Jin1, Lingwei Yu1, and Qian Tang1
ABSTRACT
Acute toxicity tests were performed on each of nine different domestic species to
determine 48-hour and 96-hour LC50 values for 2,4-dichlorophenol (DCP). Tests were also
conducted to estimate lower chronic limit (LCL) and upper chronic limit (UCL) values on four
additional species as follows: 21-day survival-reproduction test with Daphnia magna, 30-day
embryo-larval test with Carassius auratus, 60-day fry-juvenile test with Ctenopharyngodon
idellus, 30-day early life stage test with Bufo bufo gargarizans, and 96-Jiour growth inhibition
test with Sceriedesms obliqaus. In the acute tests, D. magna was observed to be the most
sensitive species tested, followed in order by Chironomus sp., Radix plicatula, and C. idellus.
The final acute value (FAV) was 2.49 mg/L DCP. In the chronic tests, survival of C. auratus
was reduced by DCP most dramatically, and survival in most tests was the most sensitive
indicator. Acute-to-chronic ratios (ACR) ranged from 3.74 to 22.5. A final chronic value (FCV)
of 0.212 mg/L DCP was obtained and a final plant value (FPV) was 7.07 mg/L DCP. Based on
FAV, FCV and FPV for DCP, a criterion maximum concentration of 1.25 mg/L DCP, and a
criterion continuous concentration of 0.212 mg/L DCP were derived. The results of this study
may provide Useful data to derive national or local water quality criteria for DCP based on
specific aquatic biota in China.
INTRODUCTION
In recent years, the impacts of industrial effluents to aquatic ecosystems have been
getting much more serious with the development of industries in China. Thus toxics control has
become an urgent issue. Approximately 7 years have passed since the revised surface water
quality standards (WQSs) and industrial discharge limits were issued by the National
Environmental Protection Agency of China (China NEPA 1988). After many years of
enforcement, the current WQSs and effluent limits seem to become forceless in water pollution
control. One of reasons is the lack of our own water quality criteria (WQC) based on aquatic
biota in China. The existing WQSs and effluent limits were established mainly by means of
'Department of Environmental Science and Engineering, Nanjing University, Nanjing, PRC.
151
-------
consulting or quoting relevant foreign criteria or standards and effluent limits. However, many
physical, chemical, and biological factors can affect the toxicity of a substance to aquatic
organisms. These factors, especially specific aquatic biota in China, could result in foreign water
quality criteria or standards that are overprotective or underprotective for aquatic ecosystems in
China. In addition, both the enforceable water quality standards and effluent limits, which are
not capable of meeting the needs of water pollution control, cover only about 30 general physico-
chemical parameters involving a few toxic organic chemicals among thousands of ones hi use
(China NEPA 1988). Under certain circumstances, therefore, even if after treatment some
effluents meet all requirements of the national effluent limits dealing with physico-chemical
parameters, toxicity to aquatic life is not significantly reduced (Jin et al. 1994). In order to
protect aquatic ecosystems hi China reasonably and adequately, it is most necessary to develop
appropriate national water quality criteria for priority pollutants based on domestic aquatic biota.
Chlorophenols have received worldwide attention hi recent years due to their high
toxicity to aquatic life, difficulty to degrade, and extreme bioaccumulation. 2,4-Dichlorophenol
(DCP) is recognized as a priority pollutant hi aquatic environment in the United States as well as
in China (Xia and Zhang 1990, Zhou et al. 1990, U.S.EPA 1991). But until now there is no
ambient water quality standard and effluent limit for this chemical in China. Few WQC
descriptions for DCP have been found through a literature search and only the lowest observed
effect level (LOEL) is available hi the U.S. Environmental.Protection Agency's Water Quality
Criteria Summary (U.S.EPA 1991).
U.S.EPA (1985) produced technical guidelines to give an objective, appropriate and
feasible way of deriving numerical national WQC. In this study, WQC were developed for DCP
based on its acute/chronic/flora toxicity data using domestic aquatic organisms selected by
referring to the technical guidelines. The toxicity data of DCP were generated by conducting
acute toxicity testing with nine species, chronic toxicity testing with four of those species, flora
toxicity testing with one species. The overall objective of this study was to provide useful data
to derive national or local WQC for DCP as well as procedures of deriving WQC for other
chemicals based on aquatic biota indigenous to China.
MATERIALS AND METHODS
Water Chemistry and Procedures
Analytical grade 2,4-dichlorophenol(HOC6H3Cl2) was purchased from Shanghai No. 1
Chemical Reagent Factory. Tap water, dechlorinated with activated carbon, was used as dilution
water for all tests. The measured quality parameters of dilution water were as follows: pH 7.0
±0.5, DO 8.0 ±0.34 mg/L, total organic carbon 0.020 mg/L, and hardness 1.86 ±0.08 mg/L as
CaCO3. The testing procedures followed the methods recommended by the relevant agencies
(APHA et al 1985, China NEPA 1986, U.S.EPA 1988).
152
-------
Test Organisms
Ten aquatic species were tested representing ten families. The following organisms were
field collected from waters in Nanjing where these organisms were abundant: Bufo bufo
gargarizans and Rana nigromaculata from Xuanwu Lake, Radix plicatula from an unimpacted
outdoor pond, Chironomus sp. from an unimpacted outdoor stream, and Limnodrilus hoffineisteri
from the Shancha River. Organisms were collected by hand or with a dip net. Great care was
taken in collection and transportation back to the laboratory. Daphnia magna a&dScenedesmus
obliqaus were obtained from an in-house culture. Carassuis auratus, Ctenopharyngodon idellus,
and Tilapia mossambica were purchased from the Freshwater Fisheries Institute of Jiangsu
Province and the Aquatic Farm of Nanjing. Upon arrival at the laboratory, all test organisms
were allowed to acclimate gradually to dilution water for a minimum of 48 hours.
Acute Toxicity Test i
Except the nonrenewal tests for Chironomus sp. and L. hoffineisteri, all tests were static-
renewal in which test solutions were totally replenished at 24-hour intervals. Standard
conditions were three replicate test containers containing 10 organisms each, with five
concentrations and a control for each test, except two replicates for R. plicatula, Chironomus sp.
and L. hoffineisteri, and six concentrations for the fish tests, Tests for fishes, amphibians, and
daphnids were under a lightdark photoperiod of 16:8 hours, while the others were under an
uncontrolled photoperiod. End points were 48-hour LC50 for daphnids, 48-hour and 96-hour
LC50 for fishes, and 96-hour LC50 for the others. Test organisms were not fed during the test
periods.
Juveniles of C. auratus (43.6 ± 4.1 mg), C. idellus (33.6 ± 34.9 mg), and T. mossambica,
(1790 ± 390 mg) were tested in beakers containing 500 ml test solution each at 21 ±3 C,
25 ±3 C, and 25 ±2 C, respectively. Tadpoles (2-5 days old) of B. bufo gargarizans and
R. nigromaculata were tested in beakers containing 500 ml test solution each at 18 ±3 C.
Neonates of D. magna (< 24 hours) were tested in 150-ml beakers at 19 +3 C. Adults of
R. plicatula were reared in dilution water in laboratory to reproduction. After the eggs were
hatched, 2- to 3-day-old juveniles were tested in 100-ml beakers at 20 ±2 C. The collected eggs
of Chironomus sp. were hatched in the laboratory. Larvae (<;24 hours old) were tested at
20 ±2 C in glass test-tubes with cotton stoppers and containing 20 ml test solution. Healthy
individuals of L. hoffineisteri with length of 1-2 cm were chosen for the test in glass test-tubes
containing 20 ml test solution at 20 ±2 C.
Chronic Toxicity Tests
The test concentrations were designed on the basis of relevant acute toxicity data.
Standard conditions mentioned above were applied to all chronic tests except 30 organisms at
four concentrations each without replicates for 60-day growth-survival test with fry-juvenile
C. idellus.
153
-------
Survival-reproduction tests (21-days) using neonates of D. magna aged no more than
24 hours were conducted in 150-ml beakers containing 100 ml test solution at 24 ±1 C in the
static-renewal mode under a lightdark photoperiod of 16:8. The daphnids were fed with green
algae (Scenedesmus obliqaus) at cell concentration of 1.0 x 105 cells/ml test solution. The test
solutions were totally renewed at 48-hour intervals. Lower chronic limit (LCL), upper chronic
limit (UCL), and chronic value (CHV) were derived by analyzing survival rate and reproduction
rate of the organisms exposed until the 21st day.
Embryo-larval tests (30-days) using newly-spawned eggs of C. auratus was performed in
150-ml beakers containing 100 ml test solution at 22 ± 1 C in the static-renewal mode under an
uncontrolled photoperiod. The test solutions were totally renewed at 24-hour intervals.
Offsprings were counted and recorded during the test. LCL, UCL, and CHV were estimated by
egg hatchability and embryo-larvae viability of the tested species.
In a 60-day flow-through test, fry of C. idellus were exposed to four test concentrations
and a control for 60 days under an uncontrolled photoperiod and at 20 ±3 C. The test was
performed in partly recycling artificial stream chambers which were 88 x 39 x 15 cm deep. The
water depth in each chamber was about 10 cm. The DCP stock solution was pumped at a speed
of 1.5 ml per minute and the dilution water was delivered through distributing pipe at a rate of
300 ml per minute into the chambers. The dilution water and DCP were mixed in the chamber.
This flow-through system yielded four complete turnovers per day in each chamber. Fish were
fed every day with 1.5 g commercial dry water flea per chamber. Periodically unused food and
waste were siphoned out of the test and control chambers, and the interior surface of the
chambers were cleaned. At the test end, length and weight of all test fish were measured and
survival rate was calculated at each concentration, from which LCL, UCL and CHV were
derived.
After the eggs of B. bufo gargarizans were hatched in the laboratory, 3-day-old tadpoles
were used in a 30-day early life stage (ELS) toxicity test in 500 ml test solution each under a
uncontrolled photoperiod and at 20 ±3 C. During the test, the tadpoles were fed every day with
cooked fresh green-leaf vegetable. The test solution was totally renewed and each container was
cleaned at 48-hour intervals. The data on survival, hind-leg development and growth (weight),
were analyzed to derive LCL, UCL and CHV.
Algae Toxicity Test
A 96-hour growth inhibition test was conducted using S. obliqaus obtained from an
in-house pure stock culture and precultured in HB-4 medium (China NEPA 1993) to the
exponential growth phase. The algae culture at the logarithmic phase of growth was diluted with
HB-4 medium to develop test algae suspension at a cell concentration of 2.0 x 104 cells/ml. The
test was performed in three replicate flasks, each containing 50 ml of test algae suspension and
50 ml of DCP solution, with five concentrations and a control for each replicate under continuous
4000 Lux illumination at 22 ±2 C. The flasks were shaken by hand several tunes a day.
154
-------
At each 24-hour interval, 0.1 ml algae-DCP mixed solution in each flask was pipetted into
20 x 20 mm plankton counting cell and direct microscopic count of algae was taken. The LCL,
UCL, and CHV values were obtained by analyzing data on inhibition of DCP to algae growth. '
Statistic Analysis and Data Integration
Either the probit program (Version 1.5, U.S. EPA 1990) or the trimmed-Spearman-
Karber (TSK) program (Ibid) was employed depending on the raw data distribution to calculate
the 48- and 96-hour LC50 values and corresponding 95% confidence intervals. Data on chronic
tests and the algae toxicity test were analyzed using the Dunnett Program (Ibid). In this way,
LCL and UCL for the most sensitive biological parameter in each test were derived as the LCL
and UCL of DCP to the corresponding species.
Integration of toxicity data was conducted according to the methods mentioned in
guidelines of the U.S. EPA. Final acute value (FAV) was derived using the FAV equation.
CHV was calculated by taking the geometric mean of the corresponding LCL and UCL. Acute-
to-chronic ratio (ACR) for a species was obtained by dividing the 96-hour LC50 (48-hour LC50
for D. magna) with corresponding CHV. Final acute-to-chronic ratio (FACR) was the
arithematic mean of all ACRs. To obtain the final chronic value (FCV), EAV was divided by
FACR. The final plant value (FPV) was a CHV obtained in 96-hour growth inhibition test with
algae. In this study, criterion maximum concentration (CMC) was FAV divided by 2 and the
minimum value between FCV and FPV was considered as the criterion continuous concentration
(CCC).
RESULTS AND DISCUSSION
Acute Toxicity
Results of acute toxicity tests using nine aquatic species (Table 1) showed that D. magna
was the most sensitive species to DCP followed by Chironbmus sp., R. plicatula and C. idellus in
that order. The 48-hour LC50 of 2.12 mg/L DCP to D. magna is consistent with the 48-hour
LC50 of 2.6 mg/L DCP reported by LeBlanc (1980). R. nigromaculata and B. gargarizans had a
similar sensitivity and were less sensitive than fishes. The least sensitive species was L.
hoffineisteri with a 96-hour LC50 of 9.89 mg/L DCP, a little higher than that of amphibians
tested. According to the toxicity of this chemical to fishes,; DCP was considered as a chemical
with high toxicity. Using the FAV equation, a FAV of 2.49 mg/L DCP was obtained.
In conducting the acute toxicity tests to derive WQC for a chemical, species selection is
an important factor that affects the results (Dobbs et al. 1994). In addition to a broad range of
taxa, availability, sensitivity, economic significance and distribution of test organisms hi China
are necessary to take into account.
155
-------
Table 1. Acute toxicity of DCP to nine freshwater species.
Rank
Species
48-hour LC50 (95% C.I.)
(mg/L DCP)
96-hour LC50 (95% C.I.)
(mg/L DCP)
I
2
3
4
5
6
7
8
9
D. magnet
Chironomus sp.
R. plicatula
C. idellus
C. auratus
T. mossambica
B. bufo gargarizans
R. nigromaculata
L. hoffineisteri
2.12(1.79-2.51)
—
—
5.84 (5.49-6.22)
10.3(9.02-11.7)
13.1 (11.0-15.7)
—
—
—
—
. 2.74 (2.27-3.30)
3.37 (2.88-3.96)
5.25 (4.83-5.70)
7.94(6.63-8.92)
8.35 (7.09-9.84)
9.46(9.11-9.83)
9.85 (8.90-10.9)
9.89(8.35-11.7)
D. magna is a standard test species in aquatic bioassays in China. It is available from
in-house cultures. The high sensitivity of daphnids to chemicals was demonstrated in this study,
so daphnids would be suitable organisms to be used in toxicity tests to derive WQC for priority
pollutants.
The amphibians R. nigromaculata and B. bufo gargarizans play an important role in
agriculture and are widely distributed in ecosystems in China. Egg masses of these species
contains thousands of eggs from the same parents. Due to their similar sensitivity to DCP, either
R. nigromaculata or B. bufo gargarizans can represent the amphibian used in toxicity tests in the
WQC derivation process.
The snail R. plicatula is a widely-distributed sensitive organism which has some
economic significance hi China as well as being important in aquatic food chains. Mature adults
of mis snail have been reported to be used in toxicity tests (Stuart 1985, Khangarot 1987,
Liu 1989). Because of higher sensitivity and less variance among individuals in early life
stages, snail larvae were used in this study. The results indicated that the larvae of R. plicatula
were highly sensitive to DCP ranking just below D. magna and Chironomus sp. in sensitivity.
Liu (1989) reported that the snails Lymnaea luteola and L. acuminata were more sensitive than
fish to phenol, pentachlorophenol (PCP), and sodium pentachlorophenolate. Stuart (1985) also
found mat the snail Gillia altitts was more sensitive to PCP than fish. In short, larvae of snail are
also ideal test organisms to be used in deriving WQC for chemicals.
Larvae of chironomids are important benthic organisms with a biomass frequently equal
to 70-80% of total benthic organisms. They act as food for commercial fishes and their activities
can affect the properties of sediment. In the past, larvae of chironomids were seldom used in
acute toxicity tests because of their commonly-recognized tolerance to pollutants. Now,
however, besides sensitivity of a species, a broad range of taxa is required in bioassays. Recent
studies have demonstrated that not all species of Chironomidae are highly tolerant to pollutants
and no given specie is able to tolerate all pollutants. Moreover, organisms at different life stages
156
-------
may have significantly different tolerance to a given pollutant. By studying the acute toxicities
often herbicides and two surfactants to larvae of C. riparius, Buhl and Faerber (1989) found that
the toxicity order of these chemicals were similar to those derived from tests with other aquatic
organisms. In this study, larvae of chironomids (Chironomus sp.) were demonstrated to be
sensitive to DCP. Studies on tests with larvae of chironomids will help to make progress on
aquatic ecotoxicology and biomonitoring.
Chronic Toxicity
Results of the 21-day test with D. magna showed that lower survival and reproduction
rates were significantly observed at 0.8 mg/L DCP compared to control (Table 2).
Table 2. Chronic toxicity of DCP to four freshwater species.
LCL/UCL (mg/L DCP)
Species
D. magna
C. auratus
C. idellus
B. bufo gargarizans
Survival
0.4/0.8
0.25/0.5
0.5/1.0
2.0/4.0
Reproduction Hatching
0.4/0.8
4.0/»
Length
1.0/2.0
Weight
1.0/2.0
0.5/1.0
Hind-leg development
2.0/4.0
At the test end, survival rate of 100% and reproduction rate of 117 ± 12.0 offsprings/adult
individual were observed in the control, satisfying the requirements for tests with this species
recommended by U.S.EPA (Weber et al. 1989) and China NEPA (1990). LCL and UCL for
survival or reproduction was 0.4 mg/L DCP and 0.8 mg/L DCP, respectively. Using CHV and
ACR equations, we derived a CHV 0.566 mg/L DCP and an ACR 3.74 for D. magna (Table 3).
The LCL, UCL and ACR derived for D. magna are close to the values described in the literature
(Gersich and Milazzo 1989).
Table 3. Acute/chronic toxicity values and acute-to-chronic ratios.
Species
D. magna
C. auratus
C. idellus
B. bufo gargarizans
LC50 (mg/L)
2.12
7.94
5.25
9.46
LCL (mg/L)
0.4
. 0.25
0.5
0.5
UCL (mg/L)
0.8
0.5
1.0
1.0
CHV (mg/L)
0.566
0.354
0.707
0.707
ACR
3.74
22.5
7.42
13.4
The results of the 30-day embryo-larval test with C. auratus showed that LCL for
survival (0.25 mg/L DCP) was less than that for hatching (4 mg/L DCP) (Table 2). Introducing
the LC50 (7.94 mg/L DCP), the LCL (0.25 mg/L DCP) and the UCL (0.5 mg/L DCP) into CHV
157
-------
and ACR equations, we obtained for this species a CHV (0.354 mg/L DCP) and the ACR
(22.5 mg/L DCP) (Table 3). In this test, lethal effect was the significant phenomenon at higher
concentrations (Table 4) which often occurred in the stages of fry/larvae development. This
finding agrees with a relevant description by McKim (1985). There are two factors that could
account for this phenomenon. First, fry/larvae are completely exposed to toxics after hatching
while the egg envelope could protect the embryo to some degree from toxic effects. Secondly,
sublethal toxic effects on the embryo could lead to deformation or death of organisms in later life
stages. Birge et al. (1979) and McKim (1985) pointed out that the embryo-larval stage was the
most sensitive in the life cycle of fishes because of critical developments involved in this stage.
In order to derive WQC, U.S. EPA (1985) allow use of data on ELS tests to estimate the chronic
toxicity of chemicals. Therefore, in this study, it was practical and feasible to employ ELS test
with fish to estimate the chronic toxicity of DCP to fish.
Table 4. Effects of DCP on hatching/survival of C. auratus.
Mean hatching/survival rate (%)
Exposure time (days)
1
2
3
4
7
30
control
2.50
50.0
87.5
95.0
92.5
90.0
0.25 mg/L
2.50
25.0
70.0
92.5
92.5
92.5
0.5 mg/L
2.50
17.5
85.0
92.5
87.5
60.02
1.0 mg/L
0.00
10.0
87.5
82.5
67.5
52.52
2.0 mg/L
0.00
0.00
87.5
65.0
55.0
22.52
4.0 mg/L
0.00
0.00
75.0
52.51
50.0
2.502
'Significant effect on hatching (a = 0.05)
'Significant effect on survival (a = 0.05)
In the tests on C. idellus and B. bufo gargarizans, although mean weight and mean length
were sensitive indicators in chronic toxicity according to Dunnett's test, survival of fry-juvenile
C. idellus was the most sensitive parameter with the lowest LCL (0.5 mg/L DCP) (Table 2). In
the ELS test with B. bufo gargarizans, the lowest LCL (0.5 mg/L DCP) meant that the mean
weight as an indicator was more sensitive than survival and normal-hind-leg development.
ACRs derived hi these two tests were 7.42 and 13.4, respectively (Table 3).
In the four chronic tests, we derived FACR of 11.8 and FCV of 0.212 mg/L DCP.
The least UCL (0.5 mg/L DCP) in these four tests is higher than the LOEL (0.366 mg/L DCP)
given in U.S.EPA's Water Quality Criteria Summary (U.S. EPA 1991). Different test species
and/or different test conditions may account for this difference.
Survival, development, reproduction (fertility, spawning and hatching), and growth
(length and weight) are the main common indicators observed in chronic toxicity tests with
single species to derive LCL and UCL values. In this study, the finding that the reproduction of
D. magna was a relatively sensitive indicator is in agreement with that reported in the literature
158
-------
(Parkhurst et al. 1981, Day and Kaushik 1987). Analyzing data in Table 2, we found that in
three tests out of four, survival was the most sensitive indicator, and only in the ELS test with
B. bufo gargarizans, growth (weight) was more sensitive than survival. Weltering has reported
survival to be a relatively sensitive indicator in 173 sublethal toxicity tests with fishes (Kooijman
1987) while 16 chronic tests out of 18 with fish demonstrated that growth was not a sensitive
indicator (Ward and Parrish 1980). In recent years, ASTM;and U.S. EPA have recommended the
highest concentration in chronic toxicity tests with fishes or daphnids to reach the level of LC50
so that survival could be guaranteed as the most sensitive parameter (Winner 1981, Day and
Kaushik 1987). Therefore, in chronic toxicity tests to derive WQCs, survival should be the first
choice as an indicator.
Toxicity to Algae
Based on results of the 96-hour growth-inhibition test with S. obliqaus, and using CHV
and FPV equations, we derived a LCL of 5.0 mg/L DCP, a UCL of 10.0 mg/L DCP, and a FPV
and CHV of 7.07 mg/L DCP. Compared to the LCL derived hi chronic tests with aquatic
animals, the LCL for S. obliqaus was much higher, which indicated that S. obliqaus was far less
sensitive to DCP than the aquatic animals tested. :
DCP Criteria Calculation
According to the equation recommended hi U.S. EPA's guidelines, CMC of 1.25 mg/L
DCP was obtained by dividing FAV (2.49 mg/L DCP) by 2. In this study, the lower value
between FCV (0.212 mg/L DCP) and FPV (7.07 mg/L DCP) was chosen as CCC=0.212 mg/L
DCP. A numerical WQC can be considered as the highest concentration of a certain substance
that would not cause any unacceptable long-term or short-term effect on the aquatic organisms or
their uses. Because aquatic ecosystems can tolerate some stress and occasional adverse effects, it
is not necessary to protect all species at all times and places. Therefore, the purpose of deriving
numerical national WQC is not to provide the same concentration at any tune for the survival and
reproduction of all species in a specific ecosystem, but to protect reasonably and adequately the
fundamental and important species and their uses hi waters at most tunes, and to avoid
overprotection or underprotection (U.S. EPA 1985).
Besides FCV and FPV, final residue value (FRV) is^also an important variable to derive
CCC. But FRV was unavailable in this study due to the lack of bioaccumulation tests and
maximum permissible tissue concentration (MPTC) issued in China. So studies on residue of
DCP remain to be conducted in the future.
159
-------
REFERENCES
APHA (American Public Health Association), American Water Works Association, and Water
Pollution Control Federation. 1985. Toxicity test methods for aquatic organisms. In:
Standard Methods for the Examination of Water and Wastewater, 16th ed. American Public
Health Association, Washington, D.C., USA pp. 689-826.
Birge, W. J., J.A. Black, I.E. Hudson, and D.M. Bruser. 1979. Embryo-larval toxicity tests with
organic compounds. In: Aquatic Toxicology, ASTM STP 667. L. L. Marking and R. A.
Kimerle, (Eds.). American Society for Testing and Materials, Philadelphia, Pennsylvania,
USA pp.131-147.
Buhl K. J., and N. L. Faerber. 1989. Acute toxicity of selected herbicides and surfactants to
larvae of the midge Chironomus ripaius. Archives of Environmental Contamination and
Toxicology 18:531-536.
China NEPA (China National Environmental Protection Agency). 1986. Standard techniques of
environmental monitoring. Sect. 4, Biomonitoring (Aquatic Environment). Beijing, China.
pp.55-64.
China NEPA (China National Environmental Protection Agency). 1988. Surface Water Quality
Standards. China National Standard GB3838-88. Beijing, China.
China NEPA (China National Environmental Protection Agency). 1990. Standard Methods for
Chemical Examination. Chemical Industry Press, Beijing, China.
China NEPA (China National Environmental Protection Agency). 1993. Manual on Aquatic
Biomonitoring. Southeast University Press, Nanjing, China p.58.
Day K., and N. K. Kaushik. 1987. An assessment of the chronic toxicity of the synthetic
pyrethroid, fenvalerate, to Daphnia galeata mendotae, using life tables. Environmental
Pollution 44:13-26.
Dobbs, M. G.5 J.L. Farris, RJ. Reash, D.S. Cherry, and J. Cairns, Jr. 1994. Evaluation of the
resident-species procedure for developing site-specific water quality criteria for copper in
Blaine Creek, Kentucky. Environmental Toxicology and Chemistry 13, 6:963-971.
Gersich, F. M., and D. P. Milazzo. 1989. Chronic toxicity of aniline and 2,4-dichlorophenol to
Daphnia magna Straus. Bulletin of Environmental Contamination and Toxicology 40:1-7.
Jin, H. J., X. Lou, Z. Zhang, and G. Wang. 1994. Ecotoxicological monitoring of major
industrial effluents in Nanjing, China. In Fish Physiology, Toxicology, and Water Quality
Management. Proceedings of the Third International Symposium, Nanjing, PRC, November
3-5,1992. U.S. EPA, EPA/600/R-94/138. pp.99-107.
160
-------
Khangarot, B.S. 1987. Zinc sensitivity of freshwater snail, Lymnaea luteola L, in relation to
seasonal variations in temperature. Bulletin of Environmental Contamination and
Toxicology 39:45-49. '
Kooijman, S. A. L. M. 1987. A safety factor for LCso values allowing for differences in
sensitivity among species. Water Research 3:269-276.
Liu, Baoyuan. 1989. The study on new material for water toxicology test. Environmental
Science & Technology 2:10-13.
LeBlanc, G.A. 1980. Acute toxiciry of priority pollutants to water flea (Daphnia magnd).
Bulletin of Environmental Contamination and Toxicology 24:684-691.
McKim, J.M. 1985. Early life stage toxicity tests. In Fundamentals of Aquatic Toxicology.
G. M. Rand and S. R. Petrocelli, Eds. Hemisphere Publishing, Washington, D.C. pp.58-95.
Parkhurst, B. R., J. L. Forte, and G.P. Wright. 1981. Reproducibility of a life-cycle toxicity test
with Daphnia magna. Bulletin of Environmental Contamination and Toxicology 26:1-8.
Stuart, R. J. 1985. Acute toxicity ofPCP to the snail, Gilliaaltills. Bulletin of Environmental
Contamination and Toxicology 35:633-640.
U.S. EPA (U.S. Environmental Protection Agency). 1985. Guideline for Deriving Numerical
National Water Quality Criteria for the Protection of Aquatic Organisms and their Uses.
National Technical Information Service Accession Number PB85227049. U.S. EPA,
Washington, D.C., USA.
U.S. EPA (U.S. Environmental Protection Agency). 1988. Short-term Methods for Estimating
the Chronic Toxicity of Effluents and Receiving Waters to Marine and Estuarine Organisms.
EPA/600/4-87/028, Environmental Monitoring Laboratory, U.S. EPA, Cincinnati, Ohio,
USA.
U.S. EPA (U.S. Environmental Protection Agency). 1990. Dunnett Program Version 1.5, Probit
Program Version 1.5, and Trimmed Spearman-Karber (TSK) Program Version 1.5,
Ecological Monitoring Research Division, Environmental Monitoring Systems Laboratory,
U.S. EPA, Cincinnati, Ohio, USA.
U.S. EPA (U.S. Environmental Protection Agency). 1991. Water Quality Criteria Summary,
Ecological Risk Assessment Branch (WH-585) and Human Risk Assessment Branch
(WH-550 D), Health and Ecological Criteria Division, U.S. EPA, Washington, D.C., USA.
161
-------
Ward, G. S., and P. R. Parrish. 1980. In: Aquatic Toxicology: 3rd Conference, ASTM STP
707, J. G. Eaton, P. R. Parrish, and A. C. Hendricks, Eds. American Society for Testing and
Materials, Philadelphia, Pennsylvania, USA pp.737-748.
Weber, C. L, W.H. Peltier, TJ. Norberg-King, W.B. Horning, II, F.A. Kessler, J.R. Menkedick,
T.W. Neiheisel, P.A. Lewis, D.J. Klemm, Q.H. Pickering, E.L. Robinson, J.M. Lazorchak,
L.J. Wymer, And R.W. Freyberg. 1989. Short-term Methods for Estimating the Chronic
Toxicity of Effluents and Receiving Waters to Freshwater Organisms, 2nd ed. EPA/600/4-
89/001, Environmental Monitoring and Support Laboratory, Office of Research and
Development, U.S. EPA, Cincinnati, Ohio, USA.
Winner, R. W. 1981. A comparison of body length, brood size and longevity as indices of
chronic copper and zinc stresses in Daphniapulex. Environmental Pollution 26:33-37.
Xia Q., and X.H. Zhang. 1990. Manual on Water Quality Standards. China Environmental
Science Press, Beijing, PRC.
Zhou, W.M., D.Q. Fu, and Z.G. Sun. 1990. LisLfor priority pollutants in waters.
Environmental Monitoring in China 6:1-3.
162
-------
A COMPUTER-ASSISTED ALTERNATIVE TO THE
USE OF FISHES IN TOXICITY STUDIES
Barbara Wigglesworth-Cooksey1 and Keith E. Cooksey1
ABSTRACT
The organism we have chosen as the model for this toxicological study is the diatom
Amphora coffeaeformis. Its choice has relevance beyond diatom toxicology. Diatoms are
eukaryotes with most of the attributes of other more developed organisms. For instance, diatom
adhesion like that in other eukaryotes requires metabolic energy, protein synthesis, glycoprotein
synthesis, the presence of sufficient calcium and cytoskeletal activity. Cells of Amphora were
attached to a microscope slide cover glass at'a defined concentration. A diffusion chamber
designed to expose cells to uniform concentrations or spatial gradients of test substances was
assembled using the cover glass with attached cells. Dark field phase contrast microscope
images were video-taped. After digitization of individual microscope fields, images were
analyzed. The following parameters were measured: number of cells, number of cells moving
greater than a predetermined speed, the paths of these cells, speed of moving cells together with
their angular velocity, acceleration and compass bearing. The analyses show that motility is
changed by chemotactic effectors. We have shown that the experimental system is capable of
detecting small behavioral changes in such measurements as speed, and should therefore be of
use as a rapid screen for compounds affecting behavior or treatments that change the
hydrophobicity of the substratum. The system responds more quickly and is less expensive than
the use offish in acute aquatic toxicology studies. ;
INTRODUCTION
Biological contamination of man-made surfaces in aqueous milieus is inevitable. All that
can be done at the moment is to slow fouling or ameliorate its effects. Recently, one of the major
weapons in the armory has been lost, since hi many parts of the world trialkyl tin compounds are
no longer considered to be environmentally acceptable components of antifouling coatings. Thus
there is need to find new antifouling strategies which must be environmentally-friendly. There
are several ways to approach this. The first depends on reducing the physicochemical interaction
of the cellular adhesive with the substratum (so-called fouling release coatings); the second
requires a slow release of a benign toxicant. This latter, seemingly oxymoronic, goal could be
accomplished if the signaling mechanism that informs the organism that is in close proximity to
a surface could be specifically confused. A molecule that could achieve this end may well be
highly specific for eukaryotic foulding organisms. However, means to screen for such
compounds are not routinely employed.
'Department of Microbiology, Montana State University, Bozeman, Montana, USA.
163
-------
Classical surface antifouling assessment techniques are slow, labor-intensive, and often
depend on the death of the organism as an end-point. They are also difficult to interpret unless
the result is dramatic. Methodology that depends on behavioral changes rather than death of the
organism might therefore be a reasonable approach to screen for toxic compounds and surface
treatment that promote fouling release.
The organism chosen as the model is the foulding diatom. Amphora coffeaeformis
(Callow 1986, Clarkson and Evans 1995). Its choice has relevance beyond diatom foulding.
Diatoms are eukaryotes with most of the attributes of other organisms of the same type. For
instance, diatom adhesion like that in other eukaryotes requires metabolic energy, protein
synthesis, glycoprotein synthesis, the presence of sufficient calcium and cytoskeletal activity
(Wigglesworth-Cooksey and Cooksey 1992). Thus compounds that affect any of these essential
metabolic processes are likely to be potentially useful as antifoulant molecules.
Most of the testing of antifoulants using algae has been related to the cell's inability to
reproduce (e.g., Callow and Evans 1981, Clarkson and Evans 1993), or their lack of success in
adhesion (Cooksey 1981, Milne and Callow 1985, Callow et aL 1986). Pennate diatoms move
by gliding, i.e., they move only when attached to a surface. The precise mechanism of the
development of the motive force is still unknown, but it depends on the secretion of an
adhesion/motility polymer that interacts with a substratum (Edgar and Pickett-Heaps 1984,
Webster et al. 1985). The physiologies of cellular motility and adhesion are closely related and
thus, it is reasonable to measure the activity of one as an indicator of the other.
MATERIALS AND METHODS
Organism
Amphora coffeaeformis (Agardh) Kutz was isolated and grown as reported previously
(Cooksey and Chansang 1976). The medium used was ASP-2 (Provasoli et al. 1957) modified to
contain 0.25 mM calcium and 8 mM Tris-HCl. In ASP-2 so modified, A, coffeaeformis, and all
other pennate diatoms tested so far, grow unattached to the vessel surface (Cooksey and Cooksey
1986). The cells were grown under constant illumination at 100-100 jiE m"2 s"1 and at a
temperature of 25 "C. Stock cultures were maintained on ASP-2 agar slants at room temperature
and lighting. The culture was recloned every 6 months and cells of the original size (25 pm)
were isolated.
Preparation of Cell Suspensions
Cells were harvested at 3.2 X 105 cells ml"1 which is in the middle of the logarithmic
phase of growth. After centrifugation for 10 minute at 3000 X g and 25 °C, the cell pellet was
washed twice in minimal medium containing 0.25 mM calcium. Minimal medium contains the
major ions of ASP-2 (NaCl, KC1, MgSO4) buffered at pH 7.8 and mM Tris-HCl, pH 7.8. In this
medium cells do not clump, are not motile and do not attach to surfaces. At the cell
164
-------
concentration at resuspension (4 X 104 cells ml"1), the spatial distribution of cells attached to the
microscope cover glass was such that cell-cell collisions were minimized.
Attachment of Cells to Glass Microscope Slide Cover Glasses
The cell suspension in minimal medium was diluted to 4 X 104 cells ml"1 in minimal
medium containing 10 mM calcium (i.e., the concentration of seawater). Immediately 18 ml of
this were added to a 20 ml beaker containing a glass microscope slide cover glass (20 mm X 30
mm) held at 45 °C to the horizontal plane. The suspension was incubated for 1-2 hours at 25 C
and 100-110 ]iE m"2 second"1 during which time all (100% ± 10%) cells above the cover glass
attached to it (Wigglesworth-Cooksey and Cooksey 1993). The cover glass was removed
carefully using forceps and assembled on the Zigmond Chamber.
Zigmond Diffusion Chamber
This 7.5 cm X 2.5 cm lucite chamber can be manufactured locally or purchased from
. Neuroprobe, Incorporated (Cabin John, MD.20818), (Zigmond 1977, Figure 1). In this work, the
chamber was modified so that the bridge was 2 mm in width. Diatoms were wiped from the
cover glass except for a strip 2 mm wide which was centered over the chamber bridge.
Assembly of the cover slide on the lucite chamber allowed the two reservoirs to be completed,
one on each side of the bridge. A chemical concentration gradient was then formed across the
bridge from one channel to the other. Details of the diffusion properties of the chamber are given
in Zigmond (1977), and Lauffenberger and Zigmond (1981).
Inverted
Cover
Glass
Figure 1. Diffusion chamber for exposure of diatom cells attached to a cover glass to uniform or
gradient concentrations of test substances (after Zigmond 1977).
165
-------
The Image Analysis System
The system, Expert vision-Cell Irak (EV) was purchased from Motion Analysis
Corporation, Santa Rosa, California, USA. The system can be used for analysis in real time or to
analyze video-taped images. It was found more convenient to do the latter. Unlike many
dynamic analysis systems which use the nearest neighbor theory to track moving cells, the EV
system depends on making a space-time worm, i.e., cells are tracked continuously in the form of
a series of over-lapping images. Thus when two cells pass close by, the computer does not
confuse their identity. It is necessary to use digitized images of single cells to enumerate the
number of pixels occupied by a cell of average size. This number is entered into the environment
file of the EV system and is used in all subsequent calculations related to that cell's behavior.
This allows the system to be tailored to examine cells of any size or a particular cell type in a
mixture of cells of different sizes, such as in a natural population.
Cells were monitored for 30 minutes after a 2-minute equilibrium period. The optical
system consisted of a Nikon Photomic microscope fitted with a 20X dark field phase objective.
Thus the images were of light cells on a dark background. During the 30-minute period, 15
randomly sleeted fields along the y-axis of the microscope stage were monitored and video taped
for a period of 2 minutes for each field. Since the normal speed of A. coffeaeformis is about 3
pm s"1, this represents tracking cells for 17-18 cell lengths.
When the computer was used to analyze the videotape, the information gathered during
two different 45-second periods from each microscopic field was digitized and processed
separately. The data from these calculations were then pooled using the MERG-Operator of the
EV-system. In this manner, data representative of the whole 2-minute period with regard to cell
number, the number motile, their direction and speed were obtained. An average of 80 cells per
microscopic field were analyzed and errors caused by cell-cell collisions were reduced. Fifteen
microscopic fields were assessed for each sample. A description of a completely manual method
to achieve similar but less complete information concerning cellular responses has been
published earlier (Cooksey and Cooksey 1988).
RESULTS
Figures 2 and 3 are examples of the output from the image analysis system when tracking
moving diatoms. Figure 2 shows the tracks of 106 cells when the results from two 45-second
periods of data collection were combined. The lengths of the tracks drawn by the computer are
proportional to the distance moved by the cells hi 45 seconds. Figure 3 shows Hie distribution of
speed (vim s"1) for all cells. Note that the line parallel to the x-axis can be set at any speed. For
these experiments cells that were moving at less than 0.25 p.m s"1 were considered "not motile",
i.e., they moved less than one cell length (25 iam). When these non-motile cells are identified in
Figure 2 they are seen as dots, e.g., track number 9. It is a simple step to calculate the percentage
of motile cells in Figure 3 (70%). Note that during any period of time (in this case 90 seconds),
nota 11 cells in the population which were capable of moving, did so.
166
-------
o^
0.25 um sec'1 during this period of data
collection [90 sec] was 70%.
167
-------
Figures 4a and 4b demonstrate that the software allows cells in a population to be
investigated individually. The tracks of cells 20 and 43 from Figure 2 were isolated
electronically. Once the tracks had been isolated, information concerning the speed of the
individual cells was calculated and if they were turning, their angular velocity. Cells that were
barely moving showed large changes in angular velocity (Figure 4b).
+ 5.9
Time [sec]
orq
ST
•i
o
n
Time [sec]
Figure 4. Angular velocity and speed of cells number 43 [a] and 68 [b] from Figure 2. Note the
fall in speed when a cell is turning [a], otherwise speed was rather constant. Angular velocity
was measured continuously from the position of the cell in the previous frame, thus only
direction of turning is given here. Counterclockwise is +. The shunting motion of a cell
moving at <0.25 vim sec"1 is shown in [b].
The presence of millimolar concentrations of calcium was shown previously to be
required for motility in marine pennate diatoms (Cooksey and Cooksey 1980, 1986). In those
experiments data gathering was a tedious process since cell movement was recorded by eye as an
"all or none" phenomenon. In Figure 5, similar information to that published previously is
presented, but now the spectrum of speeds exhibited by the population is also given. The data
were exported to a plotting program (Timetrix AXUM, Seattle, Washington) for graphical
representation. In 0.25 mM calcium in minimal medium, more than 90% of the population were
moving at speeds less than 0.5 pun second"1. In the same medium containing 10 mM calcium,
70% of cells were moving at speeds greater than this.
168
-------
0.25 mM Calcium
10 mM Calcium
*••.••••
1234
Speed (microns sec*1)
Figure 5. Speed of A. coffeaeformis in minimal medium containing 0.25mM or lOmM calcium.
Note that the cells in the lower calcium concentration were attached to a cover glass in a
solution containing the lower level of calcium immediately before assembly into the
Zigmond chamber. Measurements began immediately.
A. coffeaeformis and similar species are chemotactic to certain sugars (Cooksey and
Cooksey 1988). Figure 6 displays the direction of movement as a compass bearing for diatom
cells in a concentration gradient of 0-1 mM D-glucose, the steepness of which was 500 nM Dj-
Glucose urn"1. The gradient was linear (Zigmond 1977) and ran from 180° to 360° in an
increasing concentration. Most (79%) cells were travelling with bearings between 315° and 45°,
i.e., towards a higher concentration of glucose. Very few cells were moving at right angles to the
concentration gradient.
169
-------
WO
03
-------
Figure 7. Motility with time for A. coffeaeformis in minimal media containing [a], 0.25mM
calcium, or [b] the same medium containing ImM D-glucose.
171
-------
The results summarized in Table 1 indicate that any compound that interfered with a
major metabolic activity, influenced the motility of the cell. The lower part of the Table shows
results of the screening activity with cruce extracts of potentially bioactive compounds. The soft
coral extracts (lines 11 and 12) both caused a reduction in speed and one (Pseudopterogorgia
americana) caused negative taxis. The extract ofZostera marina (line 13) reduced diatom speed
quickly to the point that the cells exhibited a "shunting" movement similar to that seen with
tunicamyin. However, whereas the cells in tunicamyin recovered normal motility only on
washing, diatoms in Z. marina extracts recovered motility after 25 minutes without washing.
Table 1. Compounds affecting motility.
Compound concentration
Motility inhibited
Metabolic process affected
DCMU8, 2pM
CCCPb, 1.25pM
Cycloheximide, 3.6 pM
Darkness
Tunicamycin, 0.5 pg ml"1
Cytochalasin D, E, 25 pg ml"1
Colchicine, 25 pg ml"1
Podophyllotoxin, 25 pg ml"1
Ruthenium red, 6 pM
D-600C
Extract of Leptogorgia. virgulata6, 50 mg ml"1
Extract of Psaudopter-ogorgia americana ,
5 mg ml"1
Extract ofZostera marina0, methanol extract of
1 mg wet weight of tissue
0
Photosynthesis, PS II
All energy generation
Protein synthesis
Photosynthesis
Glycoprotein synthesis
Actin-based cytoskeletal activity
Tubulin-based cytoskeletal activity
Tubulin-based cytoskeletal activity
Transmembrane Ca2+ flux
Intracellular Ca2+ flux
Unknown
Unknown
Unknown
*=3-(3,4-dichlorophenyl)-l,l-dimethyl urea; b=carbonyl cyanide 3-chIorophenylhydrazone; °=a-isopropyl-a-
[(N-methyl-N-hqmoveratryl)-y amino propyl]-3,4,5-trimemoxyphenyl acetonitrile; ^extracts supplied by
Dr. N. McKeever-Targett, University of Delaware, Maryland; c=extracts supplied by Dr. R. Zimmerman, Stanford
University, Pacific Grove, California; the active principle is zosteric acid (Todd et al., 1993); 0=no effect.
DISCUSSION
The current legislative and scientific climates make the present time a particularly
importnat one in the search for novel means ot control biofouling on ships and other marine
structures. In many parts of the world, the major toxicants used in antifouling paints, the tributyl
tin (TBT) group of compounds, are no longer considered environmentally safe. This has dealt a
double blow to the industry because TBT-technology was not only antifouling, but also the basis
of a self-polishing chemistry (Milne 1991). "Self-polishing" refers to the means by which the
surface of the paint ablates continuously in seawater to leave a fresh biocide coating. Milne
(1991) has argued that irrespective of the mechanism for toxicity used in the paint, the coating
must be self-polishing for it to be a practical success.
172
-------
Whatever the development process for a new antifouling coating, the final test must be in
the marine environment. However, in selecting candidate molecules for inclusion in such
coatings, it is as well to start with a philosophy for the screening process. To some extent
potential future legislation has to be considered in designing the process to select candidate
molecules. For instance, if it is considered that in future years, no highly and generally toxic
molecules will be licensed for inclusion in antifouling coatings, then it is reasonable to screen for
other types of attributes in candidate molecules. An example could be the ability to modify
organism behavior, especially that which springs from settlement cues.
When a benign surface is placed in the marine environment, it becomes colonized with
organisms varying in size and structural sophistication from bacteria to macroalgae and
invertebrates (Costlow and Tipper 1984). Viruses are almost certainly present, but this has yet to
bew documented. An antifouling surface must take account of all these organisms if it is to be
effective, yet to utilize all organisms hi a screening process would be costly. However, a
particularly convenient assay which uses many of the groups of organisms found in a climax
community has been developed (Rittschof et ol. 1992).
A suitable model organism therefore should be chosen that is both representative and
convenient ot use. Visual inspection of a ship's hull or dock piling indicateds that the majority
of the fouling biomass is eukaryotic in nature. Furthermore, before a climax community is
achieved, biofoulding layers exposed to light are brown. This is because a large fraction of the
biomass derives from diatoms, the accessory photopigments of which are golden brown. Thus
diatoms, whicha re small eukaryotic algae, could serve as model organisms to screen for
molecules generally active against eukaryotes. It is recognized, however, that bacteria are
present in large numbers on fouled marine surfaces, and molecules active against eukaryotes may
not be effective against bacteria. :
Apart from their ability to photosynthesise and their possession of an SiO2 cell wall,
diatoms have no unusual metabolic attributes (see Table 1). They have been used successfully in
a study to discover the surface chemical factors necessary to incorporate into the design of
human plastic prostheses (Wigglesworth-Cooksey and Cooksey 1993). The results presented in
Figure 7 are an example of the similarity between diatoms (A. coffeaeformis hi this case) and
mammalian cells. The two-pool model for calcium-induced calcium release (Berridge 1991)
proposes that membrane receptor occupancy by an agonist molecule such as a hormone causes
the release of internally bound calcium which can propagate as a concentration wave through the
animal cell. This is a complicated, multi-step process that details of which are not appropriately
discussed here. However, it is proposed that the binding of a chemqtactic effector to receptors in
Amphora is a synonymous process that allows release of internally bound calcium. In this case,
the event promoted by calcium release was short-term motility. This motility was not continuous
because it was dependent on the size of the internal calcium pools, which under the condition of
the experiment, was finite.
173
-------
Bioassays using diatoms have been used by other workers. For example, Callow et al.
(1986) measured the ability of A. coffeaeformis var. perpusilla to colonize silicone elastomers in
a static adhesion test wherein the cellular attachment phas of the procedure occupied 8-16 hours.
Similar experiments (Milne and Callow 1985) were carried out under dynamic conditions using
the radial flow chamber of Fowler (Fowler and McKay 1980). The same species was used by
Clarkson and Evans (1995) to assess the efficacy of a non-leaching biocide in influencing cell
viability and growth. Thus there is a consensus that diatoms in general, and A. coffeaeformis in
particular, are reasonable organisms to use in a biocide screening system. Rittschof et al. (1992)
have discussed the steps necessary in developing a new antifouling strategy. The process begins
with laboratory searches for promising, short-lived formulations to control fouling. The protocol
described here will facilitate such searches in a rapid and objective manner.
The procedure is capable of detecting chemical agents that act as repellants as well as
inhibiting cellular metabolism. The videotape also provides a permanent record of the
experiment mat facilitates re-analysis of behavioral data. All other procedures to detect the
metabolic influences of potentially toxic materials on diatoms rely on an incubation period of
hours to days before the assessment is made. The effect can be measured by cell number.,
chlorophyll or ATP (Callow et al. 1986, Clarkson and Evans 1995). All of these measurements
take much longer than the 90 seconds used in the procedure described here. The image analysis
system also allows small differences in'behavior between cells in the same population to be
detected and measured.
The above methodology describes the use of an image analysis system to detect
behavioral changes in diatoms in response to a chemical challenge. Chemical control of
settlement and reproduction of fouling organisms is only one of the strategies in use and
undergoing further development (Milne 1991, Alberte et al. 1992). An alternative strategy to the
use of toxic materials involves a coating, the surface of which provides reduced physicochemical
interaction with the adhesives fabricated by the fouling organisms, and thus less strongly adhered
organisms (Cooksey and Wigglesworth-Cooksey 1992).
Diatoms move by the continuous secretion of a motility polymer (Edgar and Pickett-
Heaps 1984). For the diatom to develop "power" in the engineering sense, the polymer must
interact physically with the substratum on which the cell is located. Where the interaction is
small, "power" generation will be poor and the speed of the cell should reflect this, i.e., the speed
of the cells on surfaces allowing only limited interaction will be reduced. It is likely that such a
change of speed can be measured by a system able to distinguish differences of the order of 0.2
ym second"1.
Although the emphasis of this paper is the use of the image analysis system to serve anti
fouling coating research, it is obvious that it can also be used as a general test for
environmentally toxic materials. For example, it could be used to measure the success of a toxic
pollutant remediation effort. A recent U.S. Environmental Protection Agency publication (U.S.
EPA 1992) described the use of a bioremediation system for pentachlorophenol (PCP) in ground
water. The PCP load of the groundwater was reduced to 2 uM which in Daphnia magna. and
.174
-------
fathead minnows (Pimephales promelas) bioassays was considered "non-toxic". Ten pMPCP
has been found to reduce diatom motility by 81% (Cooksey unpublished) and 1.25 pM carbonyl
cyanide 3-chlorphenol hydra zone (with a mode of action identical to that of PCP) by 85%
(Table 1).
The ability to utilize small samples (100 pi) and the speed with which results can be
obtained should he useful in assessing success in experimental clean-up procedures. Further
advantages are that diatoms are far simpler to maintain than fish or other macro-organism stocks.
the test does not use animals, although it produces results that can be related to animal biology
(Table 1). Diatoms respond to mixtures of compounds and the range of chemicals that change
diatom motility is large.
The use of this image analysis in environmental toxicology is not unique. The system has
been used by Boitano and Omoto (1992) with fish sperm and Young and Bodt (1994) with rabbit
sperm. However, the time window where measurements can be made with sperm are very short
(seconds) whereas diatoms on cover glasses are motile for several hours. Diatoms have been used
in toxicological assays previously (Stauber and Florence 1985) but image analysis was not
involved in the assessment of the action of the toxicant. There is no reason that this protocol
cannot be adapted to freshwater organism.
ACKNOWLEDGMENTS
We thank the U.S. Office of Naval Research (Code 322), Defense Experimental Program
to Stimulate Competitive Research (image analysis aspects of work), and the. U.S. National
Science Foundation, Molecular and Cellular Biology program (results concerning calcium
effects) for support of this research. We also thank John Greaves, Motion Analysis Corporation,
for help in the early stages of the work.
REFERENCES
Alberte, R.S, S. Snyder, B.J. Zahuranec, and M. Whetstone. 1992. Biofouling research needs
for the United States .Navy: program history and goals. Biofouling 6:91-95.
Berridge, M.J. 1991. Cytoplasmic calcium oscillations: at two-pool model. Cell Calcium
12:63-72.
Boitano, S., and C.K. Omoto. 1992. Trout sperm swimming patterns and the role of Ca++. Cell
Motility and Cytoskeleton 21:74-82.
175
-------
Callow,M.E. 1986. Fouling algae from "in-service" ships. BotanicaMarina41:351-357.
Callow, M.E., and L.V. Evans. 1981. Some effects of triphenyltin chloride onAchnanthes
subsessilis. Botanica Marina 24: 201-205.
Callow, M.E., R.A. Pitchers, and A. Milne. 1986. The control of fouling by non-biocidal
systems, pp. 145-158 In: L.V. Evans, and K.D. Hoagland (eds). Algal Biofouling. Elsevier,
Amsterdam.
Clarkson N., and L.V. Evans. 1993. Evaluation of a potential non-leaching biocide using the
marine fouling diatom Amphora coffeaefontlis. Biofouling 7:187-195.
Clarkson, N., and L.V. Evans. 1995. Further studies investigating a potential non-leaching
biocide using the marine fouling diatom Amphora cqffeaeformis. Biofouling 9:17-30.
Cooksey, B., and K.E. Cooksey. 1980. Calcium is necessary for motility in the diatom Amphora
coefeaeformis. Plant Physiology 65:129-131.
Cooksey, B., and K.E. Cooksey. 1988. Chemical signal-response in diatoms of the genus
Amphora. Journal of Cell Science 91:523-529.
Cooksey, K.E. 1981. Requirement of calcium in adhesion of a fouling diatom to glass. Applied
Environmental Microbiology 41:1378-1382.
Cooksey, K.E., and H. Chansang. 1976. Isolation and physiological studies on three isolates of
Amphora. Journal of Phycology 12: 455 460.
Cooksey, K.E., and B. Cooksey. 1986. Adhesion of fouling diatoms to surfaces: some
biochemistry, pp. 41-53 In: L.V. Evans, and R.D. Hoagland (eds). Algal Biofouing.
Elsevier, Amsterdam.
Cooksey, K.E., and B. Wigglesworth-Cooksey. 1992. The design of antifouling surfaces:
backgrounds and some approaches, pp. 529-549 In: L.F. Melo, T.R. Boti, M. Fletcher, and
B. Capeville (eds). Biofilms: Science and Technology. Kluwer, Dordrecht. The Netherlands.
Costlow, J.D., and R. Tipper. 1984. Marine biodeterioration: an interdisciplinary study. Naval
Press, Annapolis, Maryland 384 p.
Edgar, L.A., and J. Pickett-Heaps. 1984. Diatom locomotion, pp. 47-88 In: F.E. Round, arid
DJ. Chapman (Eds). Progress in Phycological Research. Volume 3. Biopress Ltd., Bristol,
U.K.
Fowler, H.W., and A.J. McKay. 1980. The measurement of microbial adhesion, pp. 143-161
In: R.C.W. Berkeley, J.M. Lynch, J. Melling, P.R. Rutter, and B. Vincent (eds). Microbial
Adhesion to Surfaces. Ellis-Horwood, Chichester, UK.
176
-------
Lauffenberger, D.A., and S. Zigmond. 198L Chemotactic factor concentration gradients in
chemotaxis assay systems. Journal of Immunological Methods 40:45-60.
Milne, A. 1991. Ablation and after: the law and profits, pp. 139-144 In:R. Button,
T. Yakimiuk, and K. Williams (Eds). Polymers in a Marine environment, transactions of
the Institute of Marine Engineers, London.
Milne, A., and M.E. Callow. 1985. Non-biocidal anti fouling processes, pp. 229-233 In:
Polymers in the Marine Environment. Transactions of the Institute of Marine Engineers,
London.
Provasoli, L., JJ.A. McLaughlin, and M. Droop. 1957. The development of artificial media for
marine algae. Archives of Microbiology 25:392428.
Rittschof, D., A.S. Clare, D.J. Gerhart, J. Bonaventura, C. Smith, and M.G. Hadfield. 1992.
Rapid field assessment of anti-fouling and foul-release coatings. Biofouling 6:181-192.
Stauber, J.L., and T.M. Florence. 1985. The influence of iron on copper toxicity to the marine
diatom, Nitzschiaclosterium (Ehrenberg) W. Smith. Aquatic Toxicology 6:297-305.
Todd, J.S., R.C. Zimmerman, P. Crews, and R.S. Alberte. 1993. The anti-fouling activity of
natural and synthetic phenolic sulphate esters. Photochemistry 34:401-404.
U.S. EPA (U.S. Environmental Protection Agency). 1992. Biological treatment of wood
preserving SITE groundwater by Biotrol, Incorporated. EPA/540/S5-91/001, 6 pp.
Webster, D.R., K.E. Cooksey, and R.W. Rubin. 1985. An investigation of the involvement of
cytoskeletal structures and secretion in gliding motility of the marine diatom Amphora
coffeaeformis. Cell Motility 5:103-122.
Wigglesworth-Cooksey, B., and K.E. Cooksey. 1992. Can diatoms sense surfaces?: state of our
knowledge. Biofouling 5:227-238. •
Wigglesworth-Cooksey, B., and K.E. Cooksey. 1993. Final report to the U.S. National Science
Foundation Industry-University Center for Biosurfaces, Buffalo, New York. 24 pp.
Young, R.J., and B.A. Bodt. 1994. Development of computer-directed methods for the
identification of hyperactivated motion using motion patterns developed by rabbit sperm
during incubation under capacitation conditions. Journal of Andrology 15:362-377.
Zigmond, S. 1977. Ability of polymorphonuclear leukocytes to orient in gradients of
chemotactic factors. Journal of Cell Biology 45:606-616.
177
-------
-------
POLLUTION STUDIES ON PACIFIC COASTAL WATERS OF
NORTHERN BAJA CALIFORNIA, MEXICO
J. Vinicio Macias-Zamora1, Julio A. Villaescusa-Celaya1,
Efrain A. Gutierrez-Galindo1, and Gilberto Florez-Munoz1
ABSTRACT
Coastal waters of Baja California, Mexico, have experienced increased impact from human
activities in recent years. Studies have been conducted to establish present levels of several
pollutants. In particular, a study was conducted to investigate the spatial distributions and temporal
fluctuations of several contaminants for the Pacific coastal area in the vicinity of the international
border between California in the U.S., and Baja California, Mexico. Analyses were conducted for
n-hydrocarbons, polyaromatic hydrocarbons (PAHs), coprostanol, and trace metals; these were
found to be relatively low in concentration. Average concentration of n-hydrocarbon was 1.31 pg/g
for 1990 and 0.64 ug/g for 1991. Concentrations of PAHs were 0.67 and 0.59 ug/g for those same
years, and the average concentration of chromium was ca. 200 pg/g for 1991. Spatial gradients for
surface sediments are in agreement with respect to flow of materials for the area. All parameters
confims earlier findings of two main sources of materials. Trends, and origins are discussed.
Temporal trends indicate a decrease with respect to earlier studies for hydrocarbons; possible
mechanisms for this decrease are suggested.
INTRODUCTION
The Pacific coastal waters of northern Baj a California have been under the influence of an
ever growing population. Human activities, given the small degree of pollution control, have
influenced some portions of different water bodies along the peninsula of Baja California. Studies
have been conducted in the past few years in order to obtain a better idea of levels of the principal
pollutants as well as the patterns of flow of these materials, arid to investigate trends against time of
such chemicals.
The northwest area of Baja California, located hi the vicinity of San Diego, California, is an
area subjected to continuous discharges of both industrial and urban wastewater. The treatment
received by these water is usually primary only, although frequently untreated water is discharged
along the coast. Flooding associated with winter storms has produced repeated discharges of raw
sewage. The San Diego, California, wastewater treatment plant at Point Loma has been
discharging volumes on the order of 248 x 109 L per year (SCCWRP 1992). On the Mexican side
of the border, the Tijuana River receives unreported discharges of wastes from urban and industrial
areas. When winter storms force the opening of the curtain of the Rodriguez Dam, all untreated
wastewater that accumulates in different portions of the river are discharged to the sea.
Simultaneously, the Tijuana water treatment plant is releasing sewage on the beach after only
'Institute de Investigaciones Oceanologicas, Autonomous University of Baja California North, Mexico.
179 ;
-------
primary treatment in volumes of approximately 30 x 109 L/year (Macias-Zamora 1995).
In addition, the operation of the electricity generating power plant at Rosarito, which is oil driven,
represents another potential source of contaminants to the area.
In 1988, a program designed by our institution was set into place to monitor the potential
danger presented by the coastal waters to local communities. The driving force for this program
has been continual reports of closures of beaches in the San Diego area, and our lack of information
about the reasons for these closures. As a first step, it was decided that the levels of the different
contaminants should be determined and also their possible origins understood, as well as their
movements associated to sediment transport. Two kinds of pollutants were investigated: organic
pollutants (n-hydrocarbons and polyaromatic hydrocarbons (PAHs), as well as Coprostanol) and
inorganic pollutants (trace metals hi sediments). We decided to use sediments because of their
ability to contain a certain degree of history on the chemicals. The program, designated as Estudios
de Contaminacion en Baja California (ECOBAC), started in 1988 and it has been running for
several years now.
The station plan and geographical locations are shownin Figure 1. These stations have
changed as a function of available time on board our research vessel, and as a function of
availability of permits to collect on the U.S. side of the border. Typically, between 20 to 24 surface
sediment samples were collected for the different analyses during each sampling trip.
-32°30' N
^
60 A
Point Loma outfall
-32°15' N
\ t*—Tijuana Estuary
37 '38
; " - \ Tijuana
Baja California
PuntaBandera
Pacific Deean
117°15'w
Figure 1. Geographical location of the .study area and list of stations used for 1991
!mthejjrogramEGQBAG-Tv:.
ISO
-------
METHODS AND MATERIALS
n-Hydrocarbons and PAHs.
This method closely follows that reported by Burrows (1983) with some modifications.
Briefly, surface grab samples were collected in glass containers using a Van Veen grab sampler.
The samples were frozen on board and brought to the laboratory. All samples were thawed and
dried with anhydrous sodium sulfate reagent grade (J. T. Baker). Each sample was Soxhlet
extracted with dichloromethane overnight. The extracts were fractionated and cleaned by solid
phase extraction (SPE) with silica cartridges. Three fractions were obtained, and the first fraction
(F,) was eluted with hexane and contained mainly n-hydrocarbons. The second fraction (F^) was
eluted with hexane:toluene (70:30) and contained n-hydrocarbons and an unresolved complex
mixture (UCM) material if present. Fraction (F3), eluted with toluene, contained PAHs and other
toluene-soluble material.
Separation and identification of compounds was achieved with a Varian Model 3700 gas
chromatograph equipped with a 30 m capillary column (DB-5) and flame ionization detector (FID).
The oven temperature was programmed isothermally at 70 C for 2 minutes, and then it was
increased at a rate of 10 C/minute until reaching a final temperature of 270 C which was held for 6
minutes. Identification of the different compounds was done by comparison to the retention times
of pure standards. Quality control included extensive use of standards, reference materials (HS-3)
blanks, and spikes of samples with pure standards. Integration of the signal was carried out by
means of a PE Nelson model 1020X. A multiple standard external calibration was used for n-
hydrocarbons and PAHs. The limit of detection was calculated by the method described in Foley
and Dorsey (1984).
i
Coprostanol
The method, adapted to our laboratory for extraction, cleaning, and quantification of
coprostanol, was originally proposed by MacLeod et al. (1985) and later modified by Wade et al.
(1993). Briefly, it consists of addition of pregnenolone (as an internal standard) to the dry
sediment. The sample is then Soxhlet extracted with dichloromethane for 12 hours. Activated
copper fibers are placed in the extraction vessels to remove interference of sulfur. The extract is
concentrated to approximately 5 ml by evaporation in a three-ball Snyder column placed in a water
bath at 65 C. The sample is further concentrated and the solvent exchanged to hexane under
nitrogen flow. The extract (1 ml) is fractionated in a chromatography column containing 20 g of
silica gel (5% deactivated) over 10 g of neutral alumina (deactivated 1%). The column is cleaned
with pentane (fraction Fj), pentane:dichlorometane (fraction F2) and the fraction containing the
sterols is collected with methanol. The extract is evaporated to dryness and derivatized with
trimethyl chlorosilane in N,O-Bis(trimethylsylil) trifluoroacetamide (BSTFA + 1% TMCS). The
mixture is allowed to react 45 minutes at 70 C. The derivatized sample is evaporated to dryness
and a second standard (recovery standard) is added (5-p-cholestane). The sample is taken to a final
1.0ml volume.
The gas chromatograph fractionation column was a DB-5 (J & W Scientific), 30m in length
with an internal diameter of 0.53 mm. The mobile phase was H2 with 8 p.s.i. measured at 200 C.
The temperature program was from 200 to 290 C at a rate of 2 C/minute with no time holding.
This procedure was able to separate pregnenolone, 5-p-cholestane, coprostanol epicoprostanol
cholesterol and cholestanol. The injector and detector temperatures were 280 C and 320 C.
181
-------
Total Trace Metals
The metal that we are reporting was selected because it shows the general pattern obtained
for others and also, because it was the one with the largest signature. To reduce the size grain effect
on the metal distributions, we decided to utilize the analysis of the <63 fj.m size particle sediment
(Salomon and Forsmer 1984). For the analysis of the total sediment sample, we followed the
method described hi Loring and Rantala (1992). In both cases, the total metal extraction was done
according to the following procedure:
A 200-mg sediment sample was carefully weighed inside a Teflon digestion vessel
(LORRAN), to which 1 ml of aqua regia (HNO3:HC1,1:3 v/v) was added, and 3 ml .of HF.
The closed vessel was placed inside a polyethylene closed container along with a water container.
The mixture was placed inside a microwave oven and heated at full power (600 watts) during
2.5 minutes. After cooling for 5 minutes, the Teflon vessels were heated again at half power
(300 watts) for another 2.5 minutes. The samples were allowed to cool inside a water bath at room
temperature for 30 minutes. Finally, the samples were made-up to 20 ml with a 1M solution
ofH3BO3.
Trace Metal Speciation
Trace metal speciation was also performed on the samples following the scheme originally
proposed by Tessier et al. (1979), however, the fraction of exchangeable (Fj) was modified
according to Kersten and Forsmer (1986) and the residual or litogenic fraction (F2) was modified
according to Loring and Rantala (1982). The fractions Fj (exchangeable), F2 (carbonates), andF3
(Fe-Mn oxyhydroxides) were acidified to reduce re-adsorption problems. Briefly, the fractions
were obtained as follows: F, was obtained by treatment of the sample with ammonium acetate at
pH 7. Fraction F2 was obtained by treatment of the sample with a buffer of 1 M acetic acid/sodium
acetate at pH 5. The fraction F3 was obtained by treatment with NH^OH-HGl in acetic acid at 25%.
The F4 fraction is obtained after digestion of the residue with H2O2 followed by an extraction with
3.2 M of ammonium acetate. The final fraction F5 was obtained after treatment'by the procedure
described above for total metals. All metal analysis were carried out on an atomic absorption
spectrophotometer from Thermo Jarrel Ash SmithHieftje 12.
RESULTS
In a previous study (Macias-Zamora 1996), the distributions of n-hydrocarbons, PAHs and
UCM were reported for this same study area. The average concentrations for the three classes of
compounds measured were: n-hydrocarbons 15 iig/g; UCM37 y.g/g; and PAHs 0.4 /wg/g-. Those
numbers correspond to the program ECOBAC-1. Theresults'obtainedby us for ECOBAC's 3 and
4 (1990 and 1991) indicates a large change in concentration for all variables measured. The
average total concentration of n-hydrocarbons was 1.31 /^g/g for 1990 and 0.64 /ug/g for 1991. The
total concentration of PAHs did not show-such a large variation; in 1990 the-average was 0.67
-------
Table 1. Concentrations measured for n-hydrocarbons and PAHs in stations off Baja California
in 1991. :
Station
11(1)
11(2)
12
13
14
15
21
22
23
24
25
26
27
28
36
37
38
A(l)
A (2)
B
C
• E
N-Hydrocarbon Gug/g)
0.48
0.48
ND*
0.42
0.31
0.28
0.17
0.72
0.79
0.75
0.16
1.03
1.32
0.40
0.99
0.10
0.90
1.18
1.27
1.69
0.34
0.64
PAHs(/zg/g)
0.21
0.34
0.55
0.15 ,
0.88
1.40 ,
0.82 ;
0.50
1.03
0.68
0.17
1.44 ,
1.18
0.44
0.49 |
0.54
0.62
ND*
0.16
0.41
0.38
0.69
% Organic carbon (LOT)
2.24
0.38
2.18
0.49
4.64
1.73
2.31
0.22
0.55
0.26
0.42
0.56
3.52
0.55
0.26
1.25
2.70
2.31
3.01
0.39
*ND = None detected
The percent of organic carbon was obtained by the loss on ignition (LOI) according to the
method described by Dean (1974). The distributions generated by those concentrations of
n-hydrocarbons and PAHs measured in 1991 are presented in Figures 2 and 3. All distributions
were carried out based on the Kriging method which is an interpolation method that has an
unbiased approach (see for example Wehrens et al. 1993). The Kriging method was performed
with the commercial package Surfer Version 5.0.
I
The distributions for coprostanol concentrations found in 1991 are presented in Figures 4a
and 4b. Figure 4a shows the coprostanol concentration for the total sample. Figure 4b, presents the
normalized distribution with respect to the amount of organic material of the samples. Finally,
Figure 5 (a,b and c) presents the chromium distribution found for the 1991 campaign. Figure 5a
shows the total chromium concentration for the <63 IMCO. sediment fraction. This procedure
corresponds to a normalization method, designed to correct by sediment size differences. In this
work, the reactive phase has been taken as the sum of the concentrations for the first four fractions,
in other words, all metal found except the metal present in the sedimentary matrix. In this sense,
Figure 5b presents the distribution generated by the concentration of chromium in the reactive
phase. Finally, Figure 5c presents the bar distribution of the chromium concentrations found for
each separated fraction.
183 i
-------
Mexico -
Baja California
Figure 2. Distribution of the gradient concentrations (in /ug/g)
of n-hydrocarbons on surface sediments.
-32°45' N
Point Loma outfall
N
USA
SanDiego Ca|ifomja
Tijuana Estuary
Tliuana Mexico -
BaJa California
-32°15' N
Punta Bandera
Pacific Ocean
Figure 3. Distribution of the gradient concentrations (in /^g/g)
of polycyclic aromatic hydrocarbons on surface sediments.
1S4
-------
Figure 4. Spatial distribution of coprostanol (coprostanol + epicoprostanol)
in surface sediment samples: (a) in /zg/g dry weight; and (b) normalized to organic material
present in the sediment (wg/g organic matter).
185
-------
(a)
•32-30-N
S-32a15'M
V T
to
san Diego
E.U.A. -
California
"
\ L __ . _________
Oc6ano Pacifico
117°1S'W 117°00'W
(b)
|-32°45' N
-32°30' N *<#
*&
-32°15'M
E.U.A. -j
San Diego
Estuario Tijuana
Tijuana
Baja California
Punta Bandera
Oc6ano Pacifico
117*15'w
(C)
F1
F2
F3
%100
F4
F5
uu —
80 -
60 -
40 -
20 -
1
1
1
i
£
1
^
/y
ff
i
1
1
1
±*
%
\
^
^
iz
£
§
^
I I i i i . i i i i i i i i i i i i i i i
E C B A 38 37 36 28 27 26 25 24 23 22 21 15 14 13 12 11
No. de estaci6n
Figure 5. Spatial distribution for chromium: (a) Total chromium in the < 63 mm
sediment fraction; (b) Distribution of chromium Og/g) in the reactive phase;
and (c) Metal speciation according to sequential procedure.
186
-------
For quality control of metal concentration, we have used reference materials from the
National Research Council of Canada with the following results for chromium: BCSS-1 certified
123 ± 14, measured 115 ± 22.8; MESS-1 certified 71 ± 11, measured 64.4 ± 5.6.
DISCUSSION AND CONCLUSIONS
It is clear to us now that from year to year, there were large variations in the concentrations
of the different pollutants for the area. This was possibly produced by differences in the
sedimentary load moved by the local currents and also, closely related to both, the intensities of
seasonal rains for the year and the intensities of the storm-generated waves and currents
transporting materials hi the area. It appears however, that the origin for the n-hydrocarbons and
title PAHs is not the same. If their origin for these two families of compounds were common, the
dilution process produced by the sediment transport should have resulted hi similar variations in the
concentration of both types of molecules. We are inclined to think that the PAHs are introduced
mainly by the atmospheric route.
Most indications based on these analyses also suggest that the main sources of pollutants hi
the area are the wastewater treatment plants located in both Tijuana and San Diego. In particular, it
is clear that the Tijuana treatment plant, which is discharging directly on the beach, is the main
source of contaminants. The Point Loma wastewater treatment plant, discharging in the 60 m
isobat, is a more diffuse although just as permanent source of contaminants to the zone. The
circulation pattern is complicated but it appears that the main transport of sediments is towards the
south as it is reflected by Figures 1 and 3a, and not as clearly hi Figure 4a. Another indication that
the flow is southward is the fact that during the samplings of 1988,1990 and 1991, there has been
no evidence at all of a gradient hi concentration going north ;in the station closer to the Rosarito
thermoelectric power plant. This facility was originally thought to be the main concern with
respect to the oil being extensively used year around. Our distributions are consistent with the
transport flow that the angle of incidence of the waves could produce as reported by Hickey (1993).
REFERENCES
Burrows, R. 1983. Development and applications of high resolution capillary columns for gas
chromatography. Ph.D. Thesis, The Bristol University. Bristol, England 279 pp.
Dean, W.E. 1974. Determination of carbonate and organic matter hi calcareous sediments and
sedimentary rocks by loss on ignition: comparison with other methods. Journal of
Sedimentary Petrology 44:242-248.
Foley, P., and J. G. Dorsey. 1984. Clarification of the limit of detection in chromatography.
Chromatographia 18: 503-511.
187
-------
Hickey, B. M. 1993. Physical Oceanography. In: Ecology of the Southern California Bight;
A Synthesis and Interpretation. M. D. Dailey, D. J. Reish, and J. W. Anderson (Eds.)
University of California Press, US A. pp 19-70.
Kersten, M., and U. Forstner. 1986. Chemical fractionation of heavy metals in anoxic estuarine
and coastal sediments. Water Science Technology 18:121-130.
Loring,D.H.,andR.T.T.Rantala. 1992. Manual for the geochemical analyses of marine
sediments and suspended particulate matter. Earth-Science Reviews 32:235-283.
Macfas-Zamora, J. V. 1996. Distribution of hydrocarbons in recent marine sediments off the coast
of Baja California. Environmental Pollution 92:45-53.
MacLeod, W. D., D. W. Brown, A. J. Friedman, D. G. Burrows, O. Maynes, R. W. Pearce, C. A.
Wingren, and R. Bogar. 1985. Standard Analytical Procedures of the NOAANational
Analytical Facility, 1985-1986. NOAA Technical Memorandum NMFS F/NWC-92.121 pp.
Salomon, W., and U. Forstner. 1984. Metals in the Hydrocycle. Springer-Verlag, Berlin. 349pp.
SCCWRP (Southern California Coastal Water Research Project). 1992. Characteristics of
effluents from large municipal wastewater treatment facilities in 1990 and 1991. In: SCCWRP
Annual Report 1990-91 and 1991-92. SCCWRP, Long Beach, California, USA.
Tessier, A., P. G. C. Campbell, and M. Bisson. 1979. Sequential extraction procedure for the
speciation of particulate trace metals. Analytical Chemistry 51:844-851.
Wade, T. L., J. M. Brooks, M. C. Kenicutt H, T. J. McDonald, J. L. Sericano, and T. L. Jackson
1993. GERG trace organic contaminant analytical techniques. In: G. G. Lauenstein, and A. Y.
Cantillo (Eds.) Sampling and Analytical Methods of the National Status and Trends Program.
National Benthic Surveillance and Mussel Watch Project, 1984-1992, IV:121-139.
Wehrens, R., P. Van Hoof, L. Buydens, G. Kateman, M. Vossen, W. H. Mulden, and T. Bakker.
1993. Sampling of aquatic sediments. Design of a decision-support system and a case study.
Analytica Chimica Acta 271:11-24.
188
-------
WATER QUALITY OF LAKE CHAPALA, JALISCO, MEXICO
Arturo Curiel-Ballesteros1
ABSTRACT
Jalisco is one of the states in Mexico with considerable surface water, including Lake
Chapala, the largest lake in Mesoamerica. Other large lakes in Jalisco include Lake Sayula and
Lake Zapotlan. Chapala is also one of the largest lakes of the world, although with limited
depth. Normally, the lake depth has averaged 7.7 m, but by 1990 the average depth had
decreased to only 3.0 m. In water quality, the lakes in the center of the State of Jalisco have been
found to be abnormally high in total coliforms, fecal coliforms, oils and fats, and total solids
during the rainy season. The fecal coliforms have increased slightly, and concentrations of lead,
iron, phenols, dissolved oxygen, and biochemical oxygen demand are not satisfactory for human
use and aquatic life. In the last 10 years the concentration of dissolved solids has increased by
approximately 40%. The Rio Lerma is the main source of water for Lake Chapala, and because
more than 50% of the lake water volume comes from the Rio Lerma, and the supply of water to
the lake is river-limited.
INTRODUCTION
The Rio Lerma-Lake Chapala basin is the most contaminated water area of Mexico.
The basin represents 6.5% of the national territory surface, 15% of population, and 13% of the
industrial production of Mexico. In the Rio Lerma-Lake Chapala system the principal
contaminated loads received are from the Toluca Valley which has very important industrial
concentrations, including chemical-pharmaceutical plants. Also included are the industrial
complex of Queretaro, the Pemex refinery in Salamanca, and the numerous pig feed lots hi the
La Piedad region. None of the river basin towns with sewerage has an operating treatment plant
for emissions. This failure has a relationship with the current economical crisis in Mexico.
Lake Chapala, like other lakes in the State of Jalisco, is vulnerable because of continuous
changes in the receiving water and the lake's limited depth, which is decreasing. Nutrient loads
cause hyperfertilization conditions which accelerate the eutrophication process, and as a
consequence Lake Chapala finds itself in a transition phase from a healthy lake to an aging lake.
At risk are the biodiversity offish in Lake Chapala, the lake's fishing activity, and the people that
depend on its water, including Guadalajara, the second largest city in Mexico.
'Division de Ciencias Biologicas y Ambientales. Universidad de Guadalajara. Mexico.
189
-------
LAKE CHAPALA
Lake Chapala (Figure 1) is located between the co-ordinates of 20° 04' and 20° 20' North;
and 102° 35' to 103° 26' West; it has a maximum length of 86 km, an average width of 25 km,
and an approximate area of 124,000 ha. It is the largest lake of the Mexican Republic, the third
in size in Latin America, and the second in altitude in America, averaging 1,524 msnm (meters
above sea level).
.SAN NICOLAS
CHAPALA
SAM CRISTOBAL ZAPOTITLAN
SAN LUIS SOYATLAN,
TUXCUECA
TIZAPAN EL ALTO
Figure 1. Lake Chapala.
Normally, the lake average depth is 7.7 m, although this is variable. At the Jocotepec
suburbs it is 4 to 5 m deep, at the north and south shores it varies between 3.8 and 4.7 m, and at
the east shore it varies around 2.5 m. These proportions are right when the lake is in quota 97.80.
These quotas are conventional, and were fixed hi relation to a given point established on the
bridge of Rio Santiago in Cuitzeo, Jalisco, which has been assigned the quota at 100 (1526.20
msnm). The maximum quota reached by the lake is 99.38, when its storage capacity has
approximately 8,148 million m3; its "normal" capacity is approximately 6,354 million m3, and
the level and capacity are considered uncertain when it reaches the quota of 93.00, with storage
of 2,570 million m3 (Tables 1 and 2). As for administration, the lake is located in the states of
Jalisco (90%) and Michoacan (10%). In Jalisco the river cities are La Barca, Jamay, Ocotlan,
Poncitlan, Chapala, Jocotepec, Tuxcueca, and Tizapan el Alto. In Michoacan they are
Cojumatldn de Regules, Venustiano Carranza, and Briserias.
Table 1. Physical characteristics of Lake Chapala.
Characteristics
Altitude
Volume
Depth average
Area
Quota 98.00
1524 msnm
8, 100 million m3
7.2m
114,500 ha
Quota 95.00 .
1521 msnm
4,750 million m3
4.5m
103,900 ha
Quota 93 .00
1519 msnm
2,570 million m3
2.9m
90,300 ha
190
-------
Table 2. Millons of cubic meters in Lake Chapala, 1985-1995
Year
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
Minimum quota
94.19
94.14
94.29
93.45
93.20
92.20
91.91
94.41
94.53
94.55
93.89
Millions of m
3898
3844
4000
3154
2915
2020
1780
4124
4250
4271
3590
Maximum quota
95.40
95.20
; 94.98
94.71
93.47
! 93.11
I 94.66
95.56
'• 95.77
95.19
Millions of m3
5188
4970
4644
4442
3137
2830
4388
5364
5598
4959
Basically the lake serves as a regulator receptacle of the Lerma/Santiago river basin
which covers around 12,926,300 hectares and it includes the Mexican states of Queretaro,
Guanajuato, Michoacan, Aguascalientes, Jalisco, and Nayarit. The Rio Lerma is the main
water supply for Lake Chapala, providing an annual average volume of 40.6 nrYsecond; direct
precipitation and run-off in the river basin total 39.7 nrVsecond. Discharge towards the Rio
Santiago is 66.2 mVsecond, and the annual evaporation average rate is 62.8 mVsecond.
In cities neighboring the lake, more than 10% of the economically active population
depends directly on the adjacent irrigation areas which reach about 50,000 ha. These are located
mainly in the area known as La Cienega (Michoacan) and in the cities of La Barca, Jamay,
Ocotlan, Poncitlan, Jocotepec, and Tizapan el Alto (Jalisco). The most common agricultural
products are corn and sorghum in the spring-summer sub-cycle, and wheat, chick-pea, and
vegetables in winter. For the purpose of irrigation, around 1,133 million m3 are extracted from
the lake every year. The Rio Lerma originates in the state of Mexico, close to Toluca, and it
runs, before arriving into Lake Chapala, through an area of the most industrialized region of the
country, and the Bajio which, in addition to having a great number of industries, is one of the
most developed and populated agricultural area in the country. This results in the Rio Lerma-
Lake Chapala basin being the most contaminated area of the country. The relation existing
between the concentrations of nitrogen in the lake and the primary productivity process
mentioned above, has allowed the Government to establish that Lake Chapala is in a phase
between oligotrophic and mesotrophic.
Fluctuations in the lake volume may be grouped into three periods (Table 3). The most
important of all losses of water from the lake is evaporation, the annual average of which exceeds
1,400 million m3. Because Lake Chapala is one of the largest lakes of the world with limited
depth, there is a constant re-suspension of sediments. During the desiccation period of 1978
through 1983, the average depth decreased from 7.2 m to 3.4 m, but in July 1990 the average
depth was only 3.0 m.
191
-------
Table 3. Flucuations in water volume of Lake Chapala.
Rain period
Primitive
Dry
Abundant
Normal
Years
1934 thru 1944
1945 thru 1957
1958 thru 1978
1979 thru 1988
Lerma storage in million m3 annually
2,144
794
1,767
342
Quota
96.50
93.77
96.97
94.87
POLLUTION STATUS OF LAKE CHAPALA AND THE RIO LERMA
Monitoring Program
The principal pollution loads received in the Rio Lerma-Lake Chapala basin are from the
Toluca Valley in which are located very important industrial concentrations including several
chemical-pharmaceutical plants, the industrial complex of Queretaro, the Pemex refinery in
Salamanca, and numerous pig feed lots in the region of La Piedad. At the river basin of Alto
Santiago (once named Salto de Juanacatlan), 22 industrial installations are located, and none of
the river basin towns has treatment plants for sewage emissions. Therefore, the main causes of
water pollution in the Rio Lerma are: (a) discharge of residual waters from towns along the
river; (b) residual waters originating from the industry of chemical, sugar, gas, alcoholic
beverages, milk products, food, pig-farming, and others; and (c) non-point run-off waters
originating from the agricultural fields, including fertilizers, insecticide, salts, and other
agricultural byproducts.
For development of a monitoring program, 34 sampling stations were established;
25 at the lake, four at residual water outlets of towns along the river shore, three at the pumping
stations at the Cienega of Chapala, one at the Rio Lerma outflow into Lake Chapala, and at the
outflow of the Rio Santiago.
Bacteriological Analysis
The water quality of Lake Chapala from a bacteriological point of view, has been
evaluated using the technique of "Most Probable Number" for each 100 ml (MPN/100 ml),
of total coliforms as well as of faecal coliforms. The distribution of the total coliforms in Lake
Chapala is more or less uniform and concentrations detected vary between 0 and 150 MPN/
100 ml. The average number of coliforms at the lake is 118; but near to Chapala 7,500 or more
coliforms have been detected. At this tune, the lake's water has been declared hi numerous
analyses as bacteriologically undrinkable. Values from 390 thru 550 MPN/100 ml have been
measured hi areas nearest to the discharge of residual waters from the towns of Chapala,
Jocotepec, and Tizapan. Although undrinkable, the average values throughout the lake are
below the maximum allowable values of 200 MPN/100 ml for water areas that may be utilized
for recreation.
192
-------
Chemical Analysis
Fats and Oils. Regarding the Lerma chemical pollution the following is presented:
The annual load of fats and oils provided to the Rio Lerma has been in the past few years of
30,000 tons per year. Regarding chemical discharge into the lake, it represents 2,000 ton/year of
fats and oils provided also by the Rio Lerma. Said discharge has an average concentration of
29 mg/L, with a maximum of 120 mg/L, and a minimum of 0.8 mg/L, when the maximum
allowable for water is of 0.79 mg/L.
Nutrients. Nutrients, mainly from fertilizers provided by the Rio Lerma to Lake Chapala,
reach values of more than 2,000 tons per year of total nitrogen and more than 600 tons per year
of total phosphate. During the period 1972 through 1984, the annual rate of total inorganic
phosphate concentration increased from 230 g/L to almost 800 g/L. The Rio Lerma contributes
79% of this total annual load of phosphate. This load of nutrients causes hyperfertilization
conditions which accelerate the eutrophication process.
Dissolved Oxygen and Organic Matter. At Lake Chapala, the saturation concentration of
oxygen is approximately 8.2 mg/L. The annual average concentration of dissolved oxygen (DO)
in all the water mass has varied from 7.5 mg/L in earlier years to 6.8 mg/L in more recent years.
These values represent 91% and 82% respectively of saturation. Although a decrease has
occurred, the value of 6.8 mg/L is very high above the 3 or 3.2 mg/L, which is the minimum
allowable limit for healthy development of aquatic life. Values measured for biochemical
oxygen demand (BOD) vary between 1 and 3 mg/L, presenting a distribution more or less
uniform throughout the lake, except in the closest areas to discharge from the towns where the
highest values are found. The concentrations of DO at the lake surface decrease in the months of
June and July. During these months the rain is very intense and there is a lot of alluvium of clays
and tripilla (Potamogeton sp.), especially in the areas closest to the Rio Lerma outflow, which
result in a great amount of material in suspension. The highest values are recorded in March,
April and May, in which the average DO concentration may be 8.4 mg/L.
Suspended Solids and Dissolved Solids. The Rio Lerma provides Lake Chapala with
approximately 930,312 tons per year of total solids, which cbnsumes a volume in the lake of
approximately half a million'cubic meters. Two areas with different concentrations of total
suspended solids have been detected: the first with a range of 11 to 9 mg/L at the west and
central sides of the lake, and the second one located between the Rio Lerma discharge into the
lake, and the Rio Santiago discharge into the lake, as well as the southeast area of the lake, with
a concentration range of 25 to 44 mg/L. Regarding dissolved solids, they present an almost
uniform model of concentration, except for a small area located between the Rio Lerma
discharge and the Rio Santiago discharge, in which the concentration is 600 mg/L; for the rest of
the lake the dissolved solids concentration is between 450 and 500 mg/L. The tendency shown
by the suspended solids as well as by the dissolved solids over recent years is to increase. In the
last 10 years dissolved solids have increased by approximately 40%, and Lake Chapala is now a
muddy water lake. The annual average Secchi disc transparency readings for all the seasons
193 :
-------
during the period 1972 through 1984 was only 0.56 m (Limon et al. 1989), although these
readings varied considerably in different regions of the lake. For example, for 1980 variations
in the east region were within 0.2 m, compared with 0.8 m in the west portion. The average
value of total solids was 724 ppm, with the highest results corresponding to the months of June
and July. These parameters are very important because they have a direct influence over the fish
population. They may reduce the growth rate, sickness resistance, prevent the successful
development of eggs and larvae, and reduce the abundance and availability offish food.
Heavy Metals. The Rio Lerma contributes to the concentration of heavy metals in Lake
Chapala. These include lead, nickel, chromium, zinc, copper, manganese, and iron. However,
concentration of these metals in the Rio Lerma have shown considerable variation (Table 4).
Table 4. Concentration range of heavy metals detected in the Rio Lerma.
Heavy metals
Lead
Nickel
Chromium
Zinc
Copper
Manganese
Iron
ppm
0.252 - 4.37
nd-1.14
nd-2.31
0.147-2.22
0.308 - 4.06
0.302 - 2.95
1.19-149
nd = none detected
Rooted Plants. Another indication of lake pollution is the aquatic lily, Eichhornia spp.
The most abundant specie is Eichhornia crassips, whose vegetative cycle varies from 65 to
70 days, causing very quick reproduction, especially in waters of limited depth. The real
problem caused by the aquatic lily is that the depth of most of Lake Chapala is optimal for its
reproduction. The aquatic lily lowers oxygen to minimum levels, supressing life of other
species, and allowing life only to the most resistant species such as tilapia (Oreochuromis
aureits). The aquatic lily was introduced from South America, and started proliferation at the
end of the last century. It has infested many of the rivers, streams, and channels which exit into
Lake Chapala, and during 1962-1963 it covered more than 80% of the lake's area.
Phytoplankton. In Lake Chapala, 136 species of phytoplankton have been identified.
Palmer (1962), Margalef (1983), Wetzel (1983), Espinoza-Camarena (1982), and Garcia-Barrera
(1989), have mentioned that the presence of some species in.the water mass are indicators of its
characteristics. Of the indicator species of polluted water, Oscillatoria formosa, Spirogira
comunis, Scemdesmus quadricauda, and Euglena gracilis are all distributed in the east side
of the lake. Of the indicator species of clean water, Ankistrodesmusfalcatus, Amphora ovalis,
Euglena erenberghii, and E. spyrogira are located on the side of the lake where there are no
human settlements.
194
-------
Fishes. The lake ichthyofauna is formed by two kinds, five orders, nine families,
26 classes, and 53 species (Table 5). Some of the species such as the bagre de Chapala
(Ameiurus ochoterenaii) and the so called "white fish" (six species ofHirostoma) are very
strongly threatened because they are very much in demand gastronomically, and because of the
wasteful fishing method with chinchorro, which destroys a great amount of eggs. A serious
mistake was the introduction of species of carp (Cyprinus carpio), as they eat great amounts of
eggs and larvae of other fish species. The fishing activity within the lake is of great importance
considering it provides 40% of the total fish production of Jalisco, as well as 70% of all
freshwater resources. Fishing gives direct employment to one thousand fishermen, the majority
of whom belong to production units such as the Fishing Unions. The major species obtained are:
charal (Chirostoma spp.), whitefish, carp, and mojarra (Leppmis macrochiros). To a lesser
extent, frogs (Rana megapodd) are also captured. The annual production of fishes from the lake
is variable, but is approximately 1,200 tons. In 1982 the production was 1,450 tons and the
estimate for 1987 was 6,330 tons.
Table 5. Fishes of Lake Chapala.
Scientific Name
Common Name
Algansea dugessi
Algansea lacustris
A Igansea popoche *
Algansea rubescens
Algansea stigmatura
Algansea tincella
Allophorus robustus*
Allotoca digessi*
Ameiurus dugessi
Ameiurus ochoterenaii
Carassius auratus*
Ctenopharyngodon idella
Cyprinus carpio communis*
Cyprinus carpio rubrofuscus*
Cyprinus carpio specular is*
Cyprinus carpio*
Chapalichthys encaustus
Chirostoma arge
Chirostoma consocium
Chirostoma chapalae
Chirostoma diazii
Chirostoma estor
Chirostoma grandocule
Chirostoma jordani
Chirostoma labarcae
Chirostoma lucious
Chirostoma ocotlanae
Chirostoma promelas
Chirostoma sphyraena .
Dajaus monticola
Sardina
Sardina nativa
Popocha
Sardina
Sardina nativa
Sardina
Mojarrita nativa
Mojarrita nativa
Boquinete
Bagre de Chapala
Carpa roja
Carpa hervibora
Carpa conuin
Carpa barrigona
Carpa de Israel
Carpa
Godeido
Charales
Charal
Charal
Capamocho
Pescado bianco
Pescado bianco
Charal
Charal
Pescado bianco
Pescado bianco
Pescado bianco
Pescado bianco
Trucha
195
-------
Scientific Name
Common Name
Falcularms chapalae
Goodea atripinnis xaliscone
Haustor dugesii
Ictalurus dugesi
Ictalurus ochoterenae
Ictalurus punctatus*
Lebystes reticulata*
Lepomis macrochirus*
Micropterus salmoides**
Moxostoma austrimim
Neotoca biliniata
Notropis calientis
Notropis lermae
Ollentodon multipunctatus
Oreochromis aureus*
Otalia promelas
Poeciliopsis infans
Sarotherodon niloticus
Skiffla lermae lermae
Tetrapleurodon spadiceus
Xenotoca variata
Xiphophonts helleri*
Xistrosus popoche
Yuriria alta
Zoogeneticus diazil
Zoogoneticus quitzeoensis
Sardina nativa
Godeido thiro
Bagre
Bagre nativa
Bagre nativa
Bagre de canal
Gupi
Mojarra de agallas azules
Lobina
Boquinete
Godeido thiro nativa
Popocha sardina
Popocha sardina
Godeido thiro nativa
Tilapia aiirea
Blanco trompudo
Poecilidos
Mojarra
Godeido thiro nativa
Lamprea nativa
Godeido Thiro nativa
Cola de espada
Popocha
Popocha charal
Chegua
Godeido chegua
*exotic species
**extinct in the lake at this time
CONCLUSIONS
Lake Chapala is one of the largest lakes of the world, although witii limited depth.
Normally, average depth of the lake is 7.7 m, but this average depth decreased to 3.0 m in 1990.
In water quality, Lake Chapala like other lakes in the center of the State of Jalisco, have been
found to be abnormally high in total coliforms, fecal coliforms, oils and fats, and, during the
raining season, total solids. Certain water quality variables have been found to be unsatisfactory
for human use and aquatic life. These include lead, iron, phenols, DO at the lake surface, and
BOD in the areas closest to the town discharges. The tendency shown by dissolved solids these.
last 10 years has been an increase by approximately 40%. The Rio Lerma is of great importance
to the lake because it is the main water source, providing more than 50% of the lake water
volume. A problem of Rio Lerma in relation to Lake Chapala is represented by the river-limited
supply of water to the lake. The Rio Lerma-Lake Chapala basin is the most contaminated
aquatic area of Mexico. It represents 6.5% of the national territory surface; 15% of the
population; and 13% of the industrial production. In the Rio Lerma-Lake Chapala system the
196
-------
main pollution is received from the Toluca Valley, which has very important industrial
concentrations including chemical-pharmaceutical plants. This includes the industrial complex
of Queretaro, the Pemex refinery in Salamanca, and numerous pig farms in La Piedad and its
region. None of the towns in the river basin has an operating treatment plant for sewage
emissions. Lake Chapala, like other lakes in the State of Jalisco, is highly vulnerable because of
continuous changes in the quality of water discharged into ii, and its limited depth. The depth of
Lake Chapala has been decreasing in recent years, and nutrients loads being discharged into the
lake causes hyperfertilization conditions which accelerate the eutrophycation process. Lake
Chapala finds itself in a transition phase from a healthy lake to an aging lake. At risk are the
biodiversity offish in Lake Chapala, the lake's fishing activity, and the people that depend on the
water of Lake Chapala, including the residents of Guadalajara, the second largest city hi Mexico.
ACKNOWLEDGEMENTS
Manuel Guzman Arroya, Director of the Limnology; Institute, University of Guadalajara.
197
-------
-------
EFFECTS OF LAND RECLAMATION TECHNIQUES ON RUNOFF WATER
QUALITY FROM THE CLARK FORK RIVER, FLOODPLAIN, MONTANA
Frank F. Munshower1, Dennis R. Neuman1, Stuart R. Jennings1, and Glenn R. Phillips2
ABSTRACT
Persistent fill kills have been recorded in Montana's Clark Fork River below the
confluence of Silver Bow and Warm Springs Creeks. The cause of mortality was disputed until
changes in river water quality associated with a thunderstorm in 1989 were documented. After
this storm it was clear that the source of elevated metal concentrations in the dead fish was a
temporary increase in the metal levels in the river. These metals were derived from salts present
in crusts deposited on adjacent soil surfaces by evaporation ,of metal salt laden water. To prevent
the recurrence of these fish kills and to remediate the area, revegetation studies have been
conducted to assess the feasibility of vegetation establishment on the metal impacted areas within
and immediately adjacent to the floodplain. This paper presents the results of water quality
investigations associated with this and related studies in the Clark Fork River Basin. In addition,
the impact of the technologies developed hi the streambank study and implemented on
contaminated lands adjacent to the stream in which the fish kills occurred are noted.
INTRODUCTION
Periodic fish kills affecting several species have occurred hi Montana's Clark Fork River.
It was postulated that metal mining, processing, and smelting activities in the headwaters of the
river were responsible, but scientific documentation of the cause of mortality was sparse. During
a major thunder-storm in July 1989 a sequence of eVents that lead to a major fish kill was
documented. From analyses of the data collected during this event, it was clear that the fish died
because of elevated metal loads in the river water. Since these deaths have been associated with
intense summer thunderstorms, attendant soil erosion has been proposed as a major contributing
factor in the death of the fishes. While surface soils are contaminated with metals they do not,
under most conditions of erosion, provide sufficient quantities of metals to the river to cause
large acute toxicity events. Since ground water flushes of metals augmented by the storms would
not reach the river until some time after dead fish were identified, this water source is also
unlikely as a factor contributing to the fish kills. It is now recognized that soluble and very acid
metal salts transported to the river by surface water runoff caused the mortality (Phillips and
Lipton 1995). This paper presents arguments for assigning responsibility for.the elevated metal
levels to the surface salts and examines a method of remediating the movement of metals salts
into the river during precipitation events.
'Reclamation Research Unit, Montana State University, Bozeman, Montana, USA.
2Montana Department of Fish, Wildlife and Parks, Helena, Montana, USA.
199
-------
ENVIRONMENTAL SETTING
Butte, Montana, was the center of copper mining in the United States for many years.
Wastes from the processing of these pyritic ores still remain along the banks of Silver Bow
Creek, downstream from the copper mining area. Numerous waste deposits are also located in
and near the city of Anaconda, location of a now demolished world class copper smelter, and
other mining and smelling wastes are also found elsewhere along the Clark Fork River
(Figure 1). These materials range from pyritic tailings expressing very acid pH values to soils
contaminated with metals and arsenic. The mam geographic area of concern is located
immediately upstream from the confluence of Warm Springs and Silver Bow Creeks. The 1989
fish kill began below the junction of these two creeks.
Background Surface Water Quality
U.S. EPA (1986) water quality criteria values are reported in Table 1, as are typical
values measured hi water samples taken from Silver Bow Creek and the Clark Fork River
between 1985 and 1993 (Lambing et al. 1995). Acute and chronic criteria for copper and zinc
are routinely exceeded in Silver Bow Creek which is presently unable to sustain fish life.
Chronic criteria values for these metals are also commonly exceeded in the Clark Fork River but
water quality is sufficient to support trout, although numbers reported are lower than those hi
similar streams that are not impacted by metals (Hillman et al. 1995). It is during intense
thunderstorms that water quality in the streams degenerate and fish kills occur.
River Water Quality During Fish Kill
On the afternoon of July 12,1989, a severe thunderstorm occurred over the headwaters of
the Clark Fork River when the river was at a low flow (1430 L sec"1). Instream monitoring and
collection of water samples were initiated within a few minutes of the onset of the storm.
Results of monitoring are reported hi Table 2. "Within 20 minutes, the pH of the water was
reduced from 7.93 to 4.30. In terms of hydrogen ion concentration, this change represents an
increase of nearly four orders of magnitude. Alkalinity was reduced from 86 mg L"1 to
concentrations reported at or below the detection limit (< 10 mg/L), while the hardness of the
water increased by a factor of two.
Copper concentrations increased from a prestorm level of 120 ugL"1 to a maximum level
of 13,300 ug L"1 at 48 minutes. Large increases in concentrations of the other metals and arsenic
were also recorded. Concurrent with the water sampling, dead fishes were collected for
diagnostic tests. Mean concentrations (dry weight) of cadmium, copper, and zinc hi gill tissue
from the collected fish were 5.6 mg kg'1, 683 mg kg'1, and 888 mg kg'1 respectively, which
confirmed that mortality was due to toxic metal concentrations (MDFWP 1989).
200
-------
\;\ .. GARRISON *"**
BOUNDARY OF
SUPERFUND SITES
Figure 1. Silver Bow Creek and Clark Fork River in southwestern Montana.
201
-------
Table 1. Water quality criteria1 for arsenic, copper, and zinc and range of typical measurements2
of these metals in Silver Bow Creek and the Clark Fork River, March 1985 to
September 1993.
Type of criteria or
Location sampled
Metal (|J.g I/1)
Arsenic
Copper
Zinc
Acute
Chronic
Silver Bow Creek
Water quality criteria
360 18
190 12
Range of typical stream measurements (mean values in parentheses)
320
47
Below Blacktail Creek (N=8)
At Opportunity (N=8)
At Warm Springs (N=8)
Clark Fork River
At Galen (N=36)
At Deer Lodge (N=53)
At Gold Creek (N=8)
At Drummond (N=8)
11-39(18)
11-140(32)
13-29 (20)
3-60 (18)
8-200 (26)
11-31(16)
12-30 (17)
130-550 (252)
100-900 (261)
23-60 (38)
11-240(46)
12-1,500 (125)
21-150 (55)
18-140 (55)
990-1,600 (1,060)
260-1,400 (566)
40-140(85)
20-360 (79)
10-1,700 (155)
30-180 (69)
30-260 (92)
1 U.S. EPA (1986); based on hardness of 100 mg L'1 as CaCO3.
2 Lambing et al. (1995); all results reported as total recoverable.
Table 2. Water quality of river samples collected concurrent with fish kill.1
Minutes from commencement of thunderstorm
Characteristic
pH (SU)
Alkalinity (mg L~l)
Hardness (mg L"')
Aluminum (ug L"1)
Cadmium (ug L"1)
Copper (ugL'1)
Lead(ugL-')
Iron frig L'1)
Zinc (ug L'1)
0
7.93
86
216
280
3
120
2
410
140
11
6.52
• 30
285
7,870
14
2,660
11
8,570
. 3,020
21
4.30
<10
357
30,600
51
8,900
30
29,200
10,400
33
4.35
<10
392
28,100
55
9,280
10
33,400
10,000
48
4.14
<10
446
37,700
85
13,300
10
43,900
14,000
60
4.50
<10
394
22,100
58
8,300
20
29,0.0.0
9,900
113
4.98
<10
398
16,500
46
6,000
10
20.3 .
7,900
'Montana Department of Fish, Wildlife and Parks (1989).
202
-------
Surface Soil Chemistry
Metal levels in contaminated surface soils and wastes along streams in the upper Clark
Fork River Basin are reported in Table 3. The sampled materials ranged from well vegetated
soils to barren expanses of nonvegetated tailings containing highly elevated levels of metals
known to be toxic to plant growth. . :
Table 3. Metal concentrations and pH of contaminated soils and wastes adjacent to
Clark Fork River streams.1
Characteristic
pH and metal concentration range
pH(SU)
Arsenic (mg kg"1)
Aluminum (mg kg"1)
Cadmium (mg kg"1)
Copper (mg kg"1)
Lead (mg kg"1)
Zinc (mg kg"1)
2.5 - 6.7
19.3-3,140
227 - 19,700
2.6 - 108
260-11,200
82.6 - 6,477
19.2 - 22,000
'Reclamation Research Unit et al. 1989.
Metal Salts
During dry summer periods salts accumulate on the surface of soil at many sites along the"
stream. Metals are transported to the soil surface by water and accumulate because of
evaporation of the water. During intense precipitation events these salts will rapidly solubilize
and are flushed into the river as surface water runoff. The concentrations of metals in salts
collected along the banks of the Clark Fork River are reported in Table 4.
Table 4. Concentrations of metals in salts collected from surface of soils along the
Clark Fork River.1
Metal
Concentration range (dry weight)
Cadmium (mg kg"1)
Copper (mg kg"1)
Manganese (mg kg"1)
Zinc (mg kg"1)
48 - 120
70,000 - 98,500
10,000 - 19,700
22,000-31,000
'MultiTech 1987.
203
-------
Contaminant Transport
Despite the presence of tailings and contaminated soils adjacent to the river, only a
portion of the contaminants are present as water soluble components. Consequently, to achieve
the observed concentrations of contaminants in the river (e.g., 13,300 Cu L"1 or 23 g Cu sec"1 at a
flow rate of 1,727 L sec"1), the erosion of huge quantities of sediment from the floodplain would
also be required. Since the river flow rate increased by approximately 297 L sec"1 following the
thunderstorm of July 1989, this 20.7% increase demonstrates the hydrologic linkage of the river
to the storm. Using the observed stream flow conditions, known floodplain sediment chemistry
and an estimated 1 to 10% solubility of these sediments, extremely high values for total
suspended sediment (93.6 to 936 kg sediment m"3) are required to obtain the chemical
concentrations reported in the river water. The solubility of copper in tailing material is strongly
pH dependent so, typically, soluble concentrations are less than 1% of total levels for the tailing
media along these streams during pH conditions in the 4 to 5 range.
Performing the same calculations using the midpoint surface salt concentration for copper
at 84,250 mg kg"1 (Table 4) and the same 1 to 10% solubility yielded total suspended solids
(TSS) estimates from 15.7 to 1.56 kg m"3, respectively. Converting the low end of this range
(1.56 kg m"3) to standard units yields a TSS value of 1,563 mg L"1 which is similar to other values
observed in rainfall simulations performed hi support of reclamation studies conducted in the
Upper Clark Fork River Basin (RRU 1995). Using 10% solubility for the surface metal salts is a
conservative estimate, yet plausible considering the spatial distribution of surface salts which
would not comprise 100% of the floodplain immediately adjacent to the river. Sulfate salts and
hydrated sulfate salts of copper are known to solubilize readily in water.
It is postulated that these extreme concentrations of metals in readily soluble form were
responsible for increases hi river metal levels during the July storm. To prevent reoccurrences of
fish mortality, several solutions were suggested ranging from total removal of the contaminated
soils and tailings adjacent to the streams to chemical treatment of the wastes that would prevent
formation of the metal salts.
Phytoremediation Studies
Two large field investigations were initiated, one along Silver Bow Creek (RRU and
Schafer & Associates 1993) and one along the Clark Fork River (Schafer & Associates 1995).
The objectives of these studies were to test the ameliorative effects of the addition of lime
(chiefly CaCO3 and CaOH) to the areas covered with waste materials, followed by revegetation
of these areas. The lime would neutralize the acid produced from the wastes, thus reducing the
mobility and toxicity of the metals, alter the particle size distribution of the materials, and inhibit
the formation of soluble surface salts. The vegetation would reduce wind erosion, limit surface
water runoff, and harvest water in the rootzone. It was hoped that this phytoremediation
technique, along with streambank stabilization, would prevent fish mortality.
204
-------
In the Silver Bow Creek study, 100 experimental plots were implemented at five
locations along the creek banks and flood plain. The rates of lime application were determined
in preliminary laboratory soil column tests. The choice of vegetation type to be seeded into the
experimental plots was made based on greenhouse trials. One of the five field sites was a 50 ha
tailings area devoid of vegetation. The tailings material had a pH value of 4.8 and concentrations
of total arsenic, copper, and zinc were 1,671,4,579, and 7,194 mg kg'1, respectively. After
incorporation of lime, the pH of the amended tailings was in the range of 7.5 to 8.2. Soluble
metal levels within the rootzone were markedly reduced. During the third growing season
vegetation cover ranged from 55.4 to 62.0%, and the mean above ground biomass ranged from
1,977 to 2,651 kg ha"1. These values are considered to be very good for this geographical area.
Runoff Water Quality - Silver Bow Creek Study
To assess the phytoremediation effects on the quality and quantity of surface water, a
series of simulated rainfall-runoff tests were conducted. Water was applied both to the vegetated
experimental plot and to a barren area adjacent to, but separated from the plot, at the rate of
3.8 cm hr'1 for 2 hours. During the test, surface water runoff was collected from the vegetated
plot and from the barren area. The total runoff volume from the barren area was 20.15 L, while
the runoff volume from the vegetated plot was only 7.0 L. The quality of these waters was also
markedly different (Table 5).
Table 5. Runoff water quality, Silver Bow Creek study.
Initial Runoff
Characteristic
Volume (L)
pH (SU)
Arsenic (ug L"1)
Copper (ug I/1)
Zinc (ug L-1)
Arsenic mass (mg)1
Copper mass (mg)1
Zinc mass (mg)1
Total sediment (kg ha"1)
Control
0.50
4.4
140
196,000
563,000
Treated !
0.50
5.4 :
49
613
5,700
,
'
Composite Runoff
Control
20.15
4.8
51
24,500
66,400
1.08
676
1,680
478
Treated
7.0
6.3
47
387
2,710
0.33
11.6
30.5
26
'Mass calculations based on total concentrations and sediment load.
The dissolved copper concentration in the initial 0.5 L of runoff water from the barren
tailings was 196,000 ug L'1, compared to the copper level of 613 ug I/1 in the initial runoff of
0.5L from the vegetated plot. Similar reductions were evident for zinc. The masses of arsenic,
copper, and zinc transported in Hie runoff water were calculated from total concentrations and
205
-------
sediment load. Marked reductions in metal yields were demonstrated by the revegetation
treatments. The effect of phytoremediation on the total sediment yield from the rainfall-runoff
tests was significant: 478 kg ha'1 for the barren, untreated wastes compared to 26 kg ha'1 for the
treated, vegetated plot.
Runoff Water Quality - Clark Fork River Study
Runoff waters from natural precipitation events were evaluated as part of the
phytoremediation study on the Clark Fork River (Schafer and Associates 1995). Two micro-
watersheds, one located on untreated mine tailings and one located on an adjacent amended and
vegetated area, were identified. From October 1993 until July 1995,15 runoff events were
recorded from the untreated micro-watershed area. Precipitation events as small as 0.30 cm were
sufficient to cause measurable runoff. During the same time period, only one runoff event was
recorded from the vegetated area as a result of a rainfall of 1.32 cm. Water samples collected
during these runoff events were analyzed (Table 6).
Table 6. Runoff water quality, Clark Fork River study.
Characteristic
Amended, vegetated
microwatershed
Nontreated, barren
microwatershed2
pH (SU)
Total Arsenic (j.ig L"1)
Dissolved Arsenic (ug L'1)
Total Copper (ug L'1)
Dissolved Copper (ug L'1)
Total Zinc (ug L"1)
Dissolved Zinc (jig L-1)
6.2
320
270
1,360
1,330
800
790
3.87 - 4.68
2,900-21,100
13-23,000
10,4000 - 8,380,000
85,700 - 7,380,000
43,800 - 3,020,000
41,400-2,350,000
'Data from one runoff event recorded from October 1993 to July 1995.
2Range of data from 15 runoff events recorded from October 1993 to July 1995.
CONCLUSIONS
The use of phytoremediation techniques at a series of locations in Montana along Silver
Bow Creek and the Clark Fork River have demonstrated significant improvements in the amount
and quality of surface water runoff. The formation of soluble metal salts common on non-
amended, barren tailings also seems to be inhibited once the wastes have t>een treated with lime
and revegetated. Remediation of large areas of land adjacent to these streams should eliminate
fish kills in these waters. However, we caution that erosion of treated soils into streams could
potentially have adverse effects on sensitive early life stages of fishes or on adult fishes through
the food chain. Hence, lime treatment of contaminated soils should be limited to areas that will
not be captured and eroded by the river at some time in the near future.
206
-------
REFERENCES
Hillman, T.W., D.W. Chapman, T.S. Hardin, S.E. Jensen, and W.S. Platts. 1995. Assessment of
injury to fish populations: Clark Fork River NPL Sites, Montana. Report prepared for
Montana Department of Justice, Helena, Montana. 154 p.
Lambing, J.H., M.I. Hornberber, E.V. Axtmann and D.A. Pope. 1994. Water quality, bed-
sediment, and biological data (October 1992 through September 1993) and statistical
summaries of water-quality data (March 1985 through September 1993) for streams in the
Upper Clark Fork River Basin, Montana. U.S. Geological Survey. Open File Report 94-375,
Denver, Colorado. 85 p.
MDFWP (Montana Department of Fish, Wildlife and Parks). 1989. Memorandum from G.
Phillips and M. Kerr to the File, Clark Fork River Fish Kill - July 12,1989. MDFWP,
Capitol Station, Helena, Montana.
MultiTech. 1987. Silver Bow Creek Remedial Investigation. Appendix B, Part 1, pp. 3-62 and
3-63. Groundwater and Tailings Investigation. Montana Department of Environmental
Quality, Helena, Montana.
Phillips, G., and J. Lipton. 1995. Injury to aquatic resources caused by metals in Montana Clark
Fork River Basin: historic perspective and overview. Canadian Journal of Fisheries and
Aquatic Science 82:1990-1993. ;
RRU (Reclamation Research Unit), Schafer & Associates, and CH2M Hill, Inc. 1989. Final
Summary Report for Streambank Tailings and Revegetation Studies (STARS), Volumes I
and II. Phase I: Bench-scale soil column and greenhouse treatability studies and tailings
ranking system. Montana State University, Bozeman, Montana.
RRU (Reclamation Research Unit), and Schafer & Associates. 1993. Final Report for
Streambank Tailings and Revegetation Studies (STARS), Volumes I and IV. Phase III: Field
monitoring and Evaluation. Montana State University, Bozeman, Montana.
RRU (Reclamation Research Unit). 1995. Anaconda Revegetation Treatability Study (ARTS).
Phase IV Monitoring and Evaluation. Montana State University, Bozeman, Montana.
Schafer & Associates. 1995. Monitoring of the Clark Fork River Demonstration
Project. Annual Report for 1995. Prepared for Atlantic Richfield Company, Anaconda,
Montana. ;
U.S. EPA (U.S. Environmental Protection Agency). 1986. Quality Criteria for Water. EPA
400/5-86-001, Office of Water Regulation and Standards, Washington, DC.
207
-------
-------
ENVIRONMENTAL PROBLEMS AND MANAGEMENT STRATEGIES
IN LAKE BOSTEN, CHINA
Guoan Zhang1, Qing Yang1, Steve C. McCutcheon2, and Wei Zhou1
ABSTRACT
Lake Bosten is one of the largest freshwater lakes in China. It is the vital water resource
in Xinjiang Uygur Autonomous Region. In the last 30 years, the lake has been threatened by
decreasing water levels and increasing salinity and nutrients. The aquatic ecology of the lake has
changed significantly as a result of human activities and possibly natural changes. The causes of
these problems need to be investigated, and management scenarios need to be assessed. A field
survey on the water quality of Lake Bosten was carried out, A water balance model, salinity
balance model, hydrodynamic model, geochemical equilibria model, and a eutrophication model
for the lake were set up, calibrated and validated. Important factors and processes governing the
changing water level, salinity, and nutrients were analyzed with the models. The tested models.
were used to analyze management scenarios for the lake, and a summary of the key conclusions
and recommended management strategies are reported.
BACKGROUND AND OBJECTIVES
Location and Function
Lake Bosten (Figure 1) is in the south of Bohu (Bagrax) County of Bayinggolen
Autonomous Prefecture in Xinjiang Uygur Autonomous Region, People's Republic of China
(PRC). It is located in the lower reaches of the Kaidu River, between E 86°40' to 87°26' and
N 41°56' to 42°14'. It is one of the largest inland freshwater lakes in China and is the most
valuable resource in the region. Lake Bosten is one of four major reed producing regions in
China with 316,000 tons of reed (Phragmitis sp.) in reserve and annual output of 200,000 tons.
The lake contains a latent capacity of 16,916 tons offish with an annual catch of about 2, 000
tons, and is one of the two fisheries in Xinjiang. The lake is the terminal of the Kaidu River and
the source of the Peacock River. It is a natural regulating and counter- regulating reservoir for
irrigation, hydropower generation, industrial and domestic water of the drainage areas of the two
rivers.
'Xinjiang Institute of Environmental Protection, Urumqi, PRC.
2U.S.EPA National Exposure Research Laboratory, Athens, Georgia, USA.
209
-------
Yellow Ditch
Yanqi
• Heshuo
Wet Lands
Hongshaliang
Peacock River
Small Lake Area
Figure 1. Map of Lake Bosten.
Problems
In the last 30 years, serious changes have occurred in the lake as a result of human
activities, mainly farming. There are Indications that climate changes have influenced the lake
as well. There are three main problems in the lake: (1) the water level has been decreasing;
(2) the water quality is seriously degraded; (3) the aquatic ecology is largely changed.
Lake Bosten is the terminal point of other small rivers in the Yanqi Basin, in addition to
that of the Kaidu River. Since 1949, significant agricultural development has occurred in the
Yanqi Basin. Because large amounts of surface water have been diverted in the upper -
watersheds, the supply of fresh water to Lake Bosten has decreased. The Jiefang I Canal was
constructed in 1961, and it has increased the water supply to the Peacock River. The canal
diverts part of the flow from the Kaidu River directly into the Peacock River. The Kaidu River is
divided into west and east branches. The west branch flows into the smalllake area, while the
east branch flows into the large lake area. In 1958, a control gate was built on the Kaidu River to
increase the water supply in the Peacock River, and thereby greatly reducing the water elevation
level in the open lake (Figure 2).
210
-------
1049
1044
Figure 2. Water levels in Lake Bosten from 1955 to 1989.
Serious degradation of the water quality of Lake Bosten by pollutants from the Yanqi
Basin has been occurring since the 1960s. In 1958, the total dissolved solids (TDS) level was
about 385 ppm, in 1975 the concentration was about 1450 ppm, and in 1987, it was about
1860 ppm. Lake Bosten was a fresh water lake 30 years ago, but it is now slightly saline.
The contaminants consist of mineral pollutants, principally sulfate, chloride and other dissolved
solids, and moderate concentrations of the nutrients, nitrogen and phosphorus. Because of this
pollution, the water quality of Lake Bosten and its outflow, the Peacock River, is generally poor.
Associated with the degraded water quality has been the loss of local fish populations, such as
Schizothorax biddulphi and Aapiorhynchus laticeps; and th6 obvious changes in the aquatic
biotic communities of the lake (Figure 3).
The dropping water level has significantly reduced the water surface area including the
reed production area. Reeds and wetlands covered an area of 558 km2 in 1959, .495 km2 in 1981,
and about 444 km2 in recent years. In the last 30 years, the total area has decreased about
114 km2. The yield of reeds in 1965 was about 400 k-ton a year, but in 1961 it was only
316 k-ton; in recent years it is less that 220 k-ton.
211
-------
o
ci
ca
O
_a
•*-»-
c*'
100
80
60;
40
20
y_i x
/ -j. «- *
/ . >g. • ..: I
=X^
-------
OVERVIEW OF THE STUDY
Field Survey
A field survey for the water quality and ecology of Lake Bosten has been conducted since
1987 and has involved 12 sampling visits. Other investigations of the surface runoff into and
out of the lake, the amount of industrial and agriculture drainage, and the water quality has been
carried out every month. Fifty-two monitoring stations have been established for monitoring
water quality and chlorophyll-a, 17 sampling stations for ecology, and three stations for primary
production testing at the mouth of the Kaidu River. The distribution of sampling points was
selected to be representative of the physical properties of the lake water and the characteristics of
the organisms involved.
Twenty-three water quality parameters that have been monitored include temperature,
total dissolved solids, pH, dissolved oxygen (DO), chemical oxygen demand (COD), 5-day
biochemical demand (BOD5), total nitrogen (T-N), NO3-N, NO2-N, NH3-N, total phosphorus
(T- P), orthophosphate, Si, Fe, Mn, K+, Na+, Ca2+, Mg2*, Cl:, SO42', C032-, and HCO/. Species
composition, quality, and biomass have been measured for phytoplankton, chlorophyll,
zooplankton, and benthic animals.
Water Quality
The ranges and average values of water quality constituents observed during 1987-1989
are listed in Table 1. The pH value of the lake is quite high, at 8.3 to 9.0. Most of the lake is
above pH 8.5. The highest alkalinity of Lake Bosten is 8.00 milliequivalent/L, and the lowest
1.22 meq/L. That of most of the lake region is above 4.5 nieq/L Chloride ion in the main body
of the lake is 350 to 400 mg/L. Where agricultural drainage enters the lake, chloride exceeds
400 mg/L. The concentration of sulfate ion in Lake Bosten is 37.2 to 1,222 mg/L. At the river
mouth, sulfate concentrations are low; for the major part of the lake sulfate is above 650 mg/L.
At the region where highly mineralized agricultural drainage enters, the concentration is highest.
Because Lake Bosten contains so much sulfate and sodium, the salinity is 500 to 2,500 mg/L.
Phytoplankton Community
Altogether 130 species belonging to 77 genera of phytoplankton have been identified in
Lake Bosten, of which 63 species belonging to 30 genera are diatoms, 31 species belonging to
20 genera are Chlorophycea, 25 species belonging to 16 genera are Cyanophyta, three species
belonging to three genera are Chrysophyceae, three species belonging to three genera are
Dinophyceae, three species belonging to three genera are Euglenophyceae, and two species
belonging to two genera are Cryphtophyceae. Composition of species and population numbers
varies in different segments of the lake. The number of cells/L and dominant species are
reported in Table 2.
213
-------
Table 1. Water quality constituents measured during 1987-1989. Values (except pH) in mg/L.
Constituents
pH
cr
SO42"
HCO3-
Ca2*
Mg2*
Na +
TDS
Total-N
NO3'-N
NO2'-N
NH3-N
Total-P
Orthophosphate
BODS
Dissolved Oxygen
Average
8.70
382
639
234
60
124
329
1870
0.89
0.044
0.003
0.21
0.018
0.005
<2.0
7.30
Range
8.24-9.01
27.2-682
37.2 - 886
152-429
38.5 - 129
17,8-208
20.7-597
347-3070
0.24- 1.99
0.38
0.26
0.07-0-56
0.05
0.022
< 2.0 -2.8
5.99-11.6
Table 2. Distribution of population number and biomass and dominant species of phytoplankton
in Lake Bosten.
Season
Lake segment
Number of cells
(xlOVL)
Total biomass
(mg/L)
Dominant species
Spring
Summer
Autumn
Winter
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
249
241
211
299
265
100
417
885
338
378
247
643
1080
686
1000
30.0
58.7
44.3
40.8
53.3
0.825 . Cyclotella sp.
1.08 Bimdeoria tectorllm
1.96 Binulearia tectorllm
0.699 Binulearia tectorllm
1.21 Binulearia tectorllm
0.699 Filimia longiseta
2.07 Brachionus nrceus
3.14 Englypha sp.
1.60 Brachionus urceus
1.71 Difflugiasp.
0.896 Binuclearia tectorllm
1.48 Microcystis sp.
2.02 Microcystis sp.
2.06 Microcystis sp. -
' 2.26 Microcystis sp.
0.342 Scenedesmus sp.
0.361 Melosirasp.
0.490 Scenedesmus sp.
0.813 Binuclearia tectorllm
0.613 Scenedesmus sp.
214
-------
Zooplankton Community
Altogether 104 species belonging to 58 genera of zooplankton have been identified in
Lake Bosten, of which 39 species belonging to 10 genera are Protozoa, 33 species belonging to
18 genera are Rotatoria, 21 species belonging to 12 genera are Cladocera and 10 species
belonging to 9 genera are Copepoda. The composition of species, population number, biomass,
and dominant zooplankton species for different regions of the lake are reported in Table 3.
Table 3. Distribution of population number and biomass and dominant species of zooplankton in
Lake Boston.
Season
Spring
Summer
Autumn
Winter
Lake segment
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
Number of cells
(x!04/L)
625
1050
1090
284
898
272
380
570
482
748
787
819
1830
1010
1280
110
690
320
329
357
Total biomass
(mg/L)
0.181,
0.481
0.506
0.321
0.242
0.081
0.234
0.192
0.099
0.066
0.18 i
0.17
0.80
0.38
0.61 !
0.043 '
0.257
0.108
0.132
0.266
Dominant species
Tintinnidium entzii
Difflugia sp.
Englypha sp.
Microcyclops sp.
Keratella cochlearis
Filinia longiseta
Brachionus urceus
Englypha sp.
Brachionus urceus
Difflugia sp.
Difflugia sp.
Polyarthra trigla
Polyarthra trigla
Difflugia sp.
Englypha sp.
Eudiaptomus glaciloides
Eudiaptomus glaciloides
Keratella quadrata
Keratella quadrata
Keratella quadrata
Benthic Animals
Altogether 17 species of benthic animals have been found hi Lake Bosten, of which four
species belonging to two genera are Oligichaeta, 10 species belonging to eight genera aquatics,
and three species of Mollusca. Composition of species, population number, biomass, and
dominant benthic species in different regions of the lake are reported hi Table 4
215
-------
Table 4. Distribution of population number and biomass and dominant species of bentbic
animals in Lake Bosten.
Season
Spring
Summer
Lake segment
3
5
1
Number of cells
(x!04/L)
176
808
213
Total biomass
(mg/L)
0.39
0.75
0.60
Dominant Species
Sergentiasp.
Tubifexsp.
Tendipes plumosus
Autumn
2
3
4
5
1
3
4
5
212
32
200
360
184
32
71
197
1.01 . Tendipes, plumosus
0.062 Cryptochironomus digitatus
1.80 Tendipes plumosus
0.783 Tubifexsp.
0.91 Tendipes plumosus
0.03 Sergentiasp.
0.14 Tubifexsp.
0.36 Polypedilum sp.
Winter
1
3
4
128
32
312
1.35
0.18
0.51
Tendipes plumosus
Sergeniasp.
Tubifexsp.
METHODS AND CONCLUSIONS
Circulation and Mixing in Lake Bosten
The effect of lake hydrodynamics on the diffusion and transport of salt and nutrients was
investigated using a homogeneous wind-induced circulation model (Ambrose et al. 1990).
Figure 4 shows the segmentation of the lake. Only the large-scale motion, or the overall
circulation pattern was simulated. To understand the circulation in Lake Bosten, the following
wind conditions were considered:
(1) A uniform 8 m/s wind blowing from the southwest: In response to a uniform speed
and direction of wind, the system oscillates back and forth during the first few hours. Gradually
the fluctuation damps out because of the bottom friction. After 15 to 20 hours, the system
reaches steady-state (Figure 5).
216
-------
2 -- Segment
2 - Channel
Figure 4. Segmentation for Lake Bosten.
Figure 5. Flow field driven by a uniform wind speed and direction (t=24 hr ).
217
-------
(2) A sudden wind speed was increased linearly and stays constant at 8 m/s for 2 hours,
then suddenly decreases linearly: In the first few hours lake water is pushed from one end of the
lake to the other in the direction of the wind. The relative magnitude and direction of the
transport sets up a fixed pattern due to the wind-forced motion after 4 hours. This pattern
persists, merely increasing in magnitude, throughout the first 10 hours. By this time the system
reaches a steady-state (Figures 6).
10
: \ •«
\ *
X X
. vX_
\ \ • •* ~~\
r \ >•
.• -'''-;:;
Figure 6. Flow field driven by a sudden wind speed and southeast direction (t=20.0 hr ).
(3) Continuously changing wind direction and speed: In the more realistic case, fhe flow
pattern is more complex. It is also affected by the previous wind speeds and directions
(Figure?).
2.0 Cltl/5
Figure 7. Flow field driven by continuously changing wind direction and speed.
218
-------
Computation of Geochemical Equilibria in Lake Bosten
MINTEQ, a thermodynamic equilibrium model (Femly et al. 1984), was selected to
calculate aqueous speciation, adsorption, gas phase partitioning, solid phase saturatipn states, and
precipitation or dissolution. The results show that the major ionic species, and the associate
saturation indices all changed as inflowing water constituents change. The major ionic species
during different years are reported in Table 5, and the saturation indices are reported in Table 6.
Table 5. The major ionic species in Lake Bosten. (Concentrations reported in mg/L.)
. Ga2t
CaHCO3+
CaCO3 (aq)
CaSO4 (aq)
Mg2+
MgCO3 (aq)
MgHCCV
MgSO4 (aq)
1958
91.6
1.6
2.2
4.6
92.3
1.6
2.1
4.0
1975
80.7
2.0
2.1
15.3
82.1
1.5
2.6
13.8
1980
80.8
1.7
.1.4
16.1
82.1
• —
2.3
14.6
1986
76.2
1.6
1.6
20.5
78.0
1.2
2.2
18.7
1990
77.1
1.3
3.1
18.4
79.2
2.2
2.0
16.7
Table 6. Saturation indices of main precipitations in Lake Bosten.
CaCO3 (marble)
MgCO3 (dolomite)
CaCO3 (calcite)
Huntite
MgSO4 (siderite)
1958
0.386
1.309
0.541
-1.304
0.270
1975
0.521
1.908
0.676
-0.221
0.734
1980
6.686
1.834
0.840
-0.302
0.504
1986
0.457
1.321
0.611
0.391
0.812
1990
0.871
2.544
1.025
1.430
1.020
Water Quality Projections for Lake Bosten
Transport Calibration: The measured salinity pattern was used to divide Lake Bosten
into five segments, as shown in Figure 4. These five segments were used to simulate mass
transport. The modeling study was designed to simulate water circulation and salt balance.
The DYNHYD5 model (Ambrose et al. 1988) was chosen to model water movement and
WASP4 (Ambrose et al. 1988) was used to simulate the salinity distribution. DYNHYD5
accounts for the factors influencing the water balance and circulation: inflow, outflow, rainfall,
and evaporation. Information on advective flows were then provided to WASP4 for salinity and
eutrophication simulations. Dispersion coefficients were adjusted in WASP4 to describe the
mixing effects caused by small scale circulation not simulated by DYNHYD5. Data gathered
from 1987 to 1989 were used to calibrate and validate seasonally changing dispersion
coefficients. A long-term simulation from 1958 to 1982 was used to validate the model's ability
to make long-term predictions.
219
-------
Conclusions from the Mass Transport Projections: (1) DYNHYD5 and WASP4
were successfully used in modeling Lake Bosten. These models were efficient in simulating
long-term water balance and salinity changes in a large lake located in an arid region
(Figures 8 and 9); (2) The successful simulations demonstrate that evaporation from the surface
of Lake Bosten could be correctly estimated using the pan evaporation data available. Important
components of the water balance in Lake Bosten were the Kaidu River inflow, evaporation, the
pumping station outflow, and agricultural runoff. Ground water inflow or outflow do not have
decisive roles in the overall water balance of Lake Bosten; (3) Agricultural runoff is the major
reason for the elevated salinity in Lake Bosten. However, salinity in Lake Bosten only changed
slowly. Since it took a long period to elevate the salinity, it will also take a long time to reduce
lake salinity. Figure 10 shows mat salinity in Lake Bosten reached a constant level since the
agricultural return flow from the Yellow Ditch, which enters Lake Bosten at its northern edge
(Figure 1), was stopped in 1977; (4) The future of Lake Bosten depends on the management
method selected. Drastic increases in ground water use are necessary to maintain water level in
Lake Bosten and meet the water needs of the region. Salinity in the lake cannot be significantly
altered through lake management. Figure 11 and 12 show projections for the water level and
salinity in Lake Bosten under a reasonable plan for water use and salt load control; (5) Lake
Bosten is the middle segment of an inland watershed. Water and salt travel through the Kaidu
River, Lake Bosten, and the Peacock River only to disappear as groundwater recharge into the
desert. Human activities in the watersheds have changed the natural balance of water and salts.
More surface water has been used than the system can naturally supply, causing the water level
to decrease. The high water input and low drainage irrigation practice have caused accumulation
of salts in the large areas of farmland, and the flushing of this salt by irrigation practices has
caused a major salt increase in Lake Bosten. Once the salt in the river flow is elevated, lake
salinity cannot be easily controlled. Therefore, salt must be controlled at the source in future
management plans.
Simulation of Eutrophication in Lake Bosten
Modeling Approach to Eutrophication: The U.S. EPA-supported WASP4/EUTRO4
model system was selected to simulate eutrophication (Ambrose et al. 1990). The Lake Bosten
eutrophication model has been developed as a time varying, two-dimensional, vertically
averaged water quality model. The structure of the model consist of three parts: (1) transport
sub-model that provides the advective and dispersive field for each model segment from the
time-averaged output of DYNHYD5; (2) phytoplankton, DO, and nutrient sub-model describing
the various linear and non-linear interaction between the biological and chemical components of
the system; and (3) sediment flux of oxygen and nutrients as external forcing functions for the
chemical sub-model. .
Data to specify initial conditions, boundary conditions, and forcing functions (or
loadings) in time and space were readily available from field and laboratory observations. In
addition, numerous kinetic, stoichiometric and physiological parameters were needed for the
biological and physical components of the Lake Bosten eutrophication model. Where required
data were not available specifically for the Lake Bosten, data reported in the literature for other
lake ecosystems (Bowie et. al. 1985) were used to specify the various model parameter values.
220
-------
1050.0
1045.0
0 3 6 9 12 15 18 21 24
Time in Years (1958—1982)
• simulation
• measurements
Figure 8. Simulation of the long-term water balance in Lake Bosten.
25OO.O
2000.O
01
1500.0
in
TJ
10OO.O
jo
~o
500.0
O.O ""I—"—•"-*;-—r—rrnritiirTTiiiiniiiiiiifiiirinfiiiTrTiinii^iiiiiipiiiiiiiiiliiiii
03 6 9 12 15 18 • 21 24
Time in Years (1958—1982)
• simulation
measurement
Figure 9. Simulation of the long-term salinity in the middle of Lake Bosten.
221
-------
25OO.O
2000.O
1500.0
1000.0
500.0
seg.1 seg.2 seg.3
seg.4 ;seg.5
0 3 6 9 12 15 18 21 24
Time In Years (1958—1982)
Figure 10. Salinity change of no agricultural return flow in Lake Bosten.
1047.0
"£ 1046.5-
V.
Q)
"S
v»^
~S 1046.0
0)
CD
I 1045.5
1045.0' mm iiiiiiirnmmiiiiiiiimMiiiiillilliiiimiiiiMlilli inn in ilrnim ii mi nun minium
0 1 2 3 4 5 6. 7 8 9
Time in Yeras (1991—2000)
10
Figure 11. Water level under a reasonable plan in Lake Bosten.
222
-------
2500.O
0.0
Time In Years (1991—2000)
Figure 12. Salinity under a reasonable plan in Lake Bosten.
Conclusions from the Eutrophication Investigations: (1) WASP4/EUTRO4 was
successfully used in eutrophication modeling for Lake Bosten. The model was efficient in
simulating the water quality and biological responses to natural and anthropogenic point and
non-point sources of carbon and nutrients over both the short term (seasonal time scale) and
longer term (multi-year time scale); (2) Distinct seasonal variations in BOD5, DO, nutrients,
and phytoplankton occur as shown in Figure 13 and 14. Seasonal variability is due primarily to
large scale circulation forced by the wind; (3) The results of simulation and sensitivity analysis
indicates that Lake Bosten is a phosphorous limited system with respect to phytoplankton
dynamics. Future management plans should limit phosphorous sources to control the
eutrophication of the lake; (4) Future projections show that water quality and the ecological
community in Lake Bosten may not change very much under current loads. This means that
sound management is necessary to control new loads caused by the development of industry and
agriculture in the watershed (see Figures 15 and 16).
223
-------
—Simulated
o measured
0 200 400 600 800 1000 1200
Trme(days)
Figure 13. Model vs. observe chlorophyll a in Segment 1.
-Stafcted
o measured
0 200 400 600 800 1000 1200
Tkne (days)
Figure 14. Model vs. observe chlorophyll a in Segment 2.
224
-------
-Segment 1
•Segment 2
Segment 3
-Segment 4
Segment 5
o
0 200 400 600 800 1000 1200 1400 1600
Time In days (1993-1996)
Figure 15. Water simulation under the planned water supply and average loads.
15
-Segment 1
•Segment 2
Segments
--Segment 4
Segment 5
0 200 400 600 800 1000 1200 1400 1600
Thne In days ( 1993-1996 )
Figure 16. Water simulation under the planned water supply and average loads.
225
-------
SUGGESTIONS FOR FUTURE MANAGEMENT OF LAKE BOSTEN
Due to the complex nature of the environmental problems of Lake Bosten that were
revealed in this study, two types of remedial approaches are recommended. Both environmental
control technology and better water management should be considered. Water treatment
technologies should be considered to reduce or eliminate point source loads of phosphorus and
organic wastes.
Treatment of non-point sources of salt and nutrients, requires anew water management
approach for the lake and the Kaidu River drainage basin. The following four approaches are
recommended: (1) improve irrigation practices using spray irrigation, drip irrigation, and
irrigation on plastic film to reduce both freshwater withdrawals from the Kaidu River, and the
volume of contaminated return flows; (2) significantly increase use of ground water to drop the
phreatic water surface and prevent further salinization of shallow, surface soils in the Yanqi
basin; (3) use advanced methods for applying fertilizer to reduce return flow concentrations of
nitrogen and phosphorus; and (4) improve production of reeds and forests to provide fuel by
using industrial and agricultural wastewaters for irrigation.
ACKNOWLEDGMENTS
This project was conducted under the US-PRC Agreement to Cooperate in the Field of
Environmental Protection, Annex 3. It includes the field survey and water quality projections.
The authors thank Rosemarie C. Russo, U.S. Environmental Protection Agency, for assisting in
the organization of the projects, and also Zhang Chunghua, Zang Yuxiang, and Liang Sicui,
senior engineers of the P.R.C. National Environmental Protection Agency for their support.
Robert Ambrose of the U.S. EPA made valuable proposals for the analysis and monitoring of the
water quality of the lake. During the hydrodynamic and salinity modeling for Lake Bosten at the
U.S. EPA Laboratory in Athens, USA., the authors had many valuable and beneficial discussions
with-P. F. Wang, Tim Wool, Mansour Zakihkani and other members of AScI Corporation on the
eutrophication modeling of the lake. Zhu Dongwei did much of the preliminary hydrodynamics
and salinity modeling while he was a visiting scholar at the Laboratory in Athens. Several
valuable suggestions were made by Jin Xiancan, Li Xixian, Han Xiankun, Chen Jingsheng,
Fu Guowei, Wu Shenyan, professors and senior engineers from China. The measurement of
field data is the result of the group effort of many colleagues.
REFERENCES
Ambrose, R.B., T.A. Wool, J.L. Martin, J.P. Connolly, and R.W. Schanz. 1988. WASP4,
A Hydrodynamic and Water Quality Model-Model Theory, Users Manual, and Programmers
Guide. U.S. Environmental Protection Agency, Athens, Georgia, EPA/600/3-87/039.
226
-------
Bowie, G.L., W.B. Mills, D.B. Porcella, C.L. Campbell, J.R. Pagenkopf, G.L. Rupp, K.M.
Johnson, P.W.H. Chan, S.A. Gherini, and C.E. Chamberlin. 1985. Rates, Constants, and
Kinetics Formulations, in Surface Water Quality Modeling. Second Edition. U.S.
Environmental Protection Agency Athens, Georgia, EPA-600/3-85-040.
Felmy, A. R., D.C. Girvin, and B.A. Jenne. 1984. MINTEQ - A Computer Program for
Calculating Aqueous Geochemical Equilibria. Battelle Northwest Laboratory Report for the
U.S. EPA, Athens, Georgia.
Jorgensen, S. E., and R. A.Vollenweider. 1980. Guidelines of Lake Management. Pergamon,
London.
McCutcheon, S. C., D. Zhu, and G. Zhang. 1990. Water Quality Modeling in an Arid Region
Lake: Application to Lake Bosten, Xinjiang, P.R.C. American Society of Civil Engineers
Hydraulics Division, 1990 National Conference on Hydraulic Engineering and the
International Symposium on the Hydraulics/Hydrology of Arid Lands, San Diego, California.
Zhang, G. 1990. Lake Bosten in Xinjiang, The 4th International Conference on the Conservation
and Management of Lakes, Hang Zhou, PRC.
Zhang, G., D. Zhu, S. C. McCutcheon, X. Pei, and X. Zhong. 1990. Water Quality Projections
for Lake Bosten, Xinjiang, PRC. Sacramento, California.
Zhang, G., Q. Yang, and S. C. McCutcheon. 1993. Eutrophication Modeling for a Large Lake in
an Arid Region -- Lake Bosten, Northwest China. China-Canada Workshop on
Eutrophication of Lakes and the Relevant Water Treatment Technologies, Beijing, PRC.
Zhang, G., N. Xu, S. C. McCutcheon, Q. Yang, Y. Li, and W. Zhou. 1994. Lake Bosten in
Xinjiang. In: Lakes in China. Chinese Marine Science Publishing House, Beijing, PRC.
227
-------
-------
RTVER WATER QUALITY MODELING IN POLAND
Marek J. Gromiec1
ABSTRACT
Processes of intensive urbanization, growth of population, intensification of agriculture,
and the growth of industry in Poland have resulted in the deterioration of surface water resources.
Water pollution control has become one of the most important environmental problems
throughout Poland, since the majority of industry is situated near the origins of the country's
river systems. The main rivers are heavily used for municipal and industrial water supply,
agricultural irrigation, cooling purposes for power plants, and navigation. At the same time,
these rivers also receive discharges of wastewater with only varying degrees of treatment.
Mathematical modeling of river systems has become an integral part of water resources planning
and water-quality management in Poland, and various computerized models have been applied
for water quality simulation hi the Polish rivers. This paper presents selected examples of
various applications of these models.
BACKGROUND: WATER QUANTITY AND QUALITY ANALYSIS
Water Resources and Water Demands
The Republic of Poland covers 312,520 km2 and is divided into 49 provinces
(voivodships). The total water surface of Poland is 5,000 kin2 or 1.6% of the country. There are
about 9,300 lakes with a total area of 3,200 km2. The Masurian Lake District has 1,063 lakes, the
largest of which cover approximately 11,000 hectares. Lakes and artificial reservoirs in Poland
have a total capacity of 33 km3, and a large number of ponds hold an additional 1 km3. The two
most important rivers in the country are the Vistula River with a basin area of 194,000 km2, and
the Odra River which has a basin area of 110,000 km2.
The water balance during a normal annual cycle in Poland is presented below. The
average annual amount of rainfall is 597 mm, equivalent to 186.6 km3 of water per year over the
whole country. Since tributaries from outside Poland yield an additional 5.2 km3 of water
annually, the total input of water is 191.8 km3. Underground water resources have been
estimated at 33 km3 per year for an area of 272,520 km2, since the remaining 13% of the total
area is waterless. The annual dynamic underground water resources have been evaluated at
9.2 km3. However, rivers and streams discharge only about 58.6 km3 of water into the Baltic Sea
Dept. of Water Management, Institute of Meteorology and Water Management, Warsaw, Poland.
229
-------
during a mean low-flow year, and about 34 km3 in a mean dry-weather year. Obviously, only a
portion of this volume is available. About 10 km3 is necessary as a minimum flow to maintain
biological life and for sanitary reasons. Therefore, the available flow is only 24 km3 of water.
Poland belongs to the group of European countries most deficient in water resources,
ranking 22nd overall. Average annual water resources in Poland, estimated on the basis of
atmospheric inputs and the number of population, amount to 1,600 m3 per inhabitant compared
with 2,800 m3 per inhabitant for Europe. In 1992 the total water consumption in Poland was
12.5 km3, with 22.5%, 66.5% and 11% for municipal, industrial and agricultural purposes,
respectively. Most water for agriculture is taken during the summer months. However, available
water volume compares unfavorably witiht the water demand anticipated in the future, particularly
during dry seasons.
Legislation and Administrative Aspects
The present basis for legal action hi the field of protection against water pollution is the
Water Law Act issued by the Polish Parliament in 1974. In 1975, on the basis of the Water Law,
the Council of Ministers announced regulations concerning classification of waters and
determination of effluent standards, as well as financial penalties for effluent discharges that do
not meet the requirements specified in the regulations. The following classes of surface water
quality were established: (1) Class I waters are those used for municipal and food processing
supply purposes, and for salmon fish growth; (2) Class II waters are intended for use as
recreational waters, including water sports and swimming, and for growth offish other than
salmonidae; (3) Class III waters (the lowest class) are only used as industrial water supplies and
for irrigation purposes.
Water quality standards are tailored to meet appropriate use of surface waters. In
addition, the following provisions were laid down by the Water Law: (1) Industrial plants and
other operations which discharge wastewaters to water or to land are obliged to construct,
maintain, and utilize wastewater treatment facilities; (2) Without simultaneous operation of
wastewater treatment systems, no industrial plant or any other plant from which wastewater is
discharged can initiate operation; and (3) A permit is required to maintain wastewater discharge.
In addition, the two most important decrees include: (1) decree of the Minister of
Environmental Protection, Natural Resources and Forestry (1991) on classification of waters and
on the conditions which must be fulfilled when the wastewater is discharged into waters or .
ground; and (2) decree of the Minister of Health and Social Welfare (1990) concerning -
conditions which must be met by drinking water and by water for industrial purposes.
In 1991 the Polish Parliament passed a bill concerning ecological policy which
determines the general rules, amis, and directions of future actions in Poland. Currently a new
version of the Water Law Act (Parliment of Poland 1994) is under preparation. The new system
of water resources management introduces subdivision of the country's water system into river
catchment-based areas under responsibility of River Basin Water Authorities.
230
-------
Water Quality Problems
Processes of intensive urbanization, growth of population, intensification of agriculture,
and growth of industry have resulted in deterioration of surface water resources, despite the
introduction of certain measures for water pollution control. Therefore, the majority of the major
rivers have experienced serious degradation of water quality (Figure 1).
1964-1967 1971-1973 1978-1983 1990-1992
1968-1970 1974-1977 1984-1988 1993
Years
III
I overolass
Figure 1. Water quality classification: physical and chemical criterion.
Also lakes, especially in the northern region, are threatened by eutrophication with
phosphorus as the primary cause. Currently, different degrees of treatment are being applied to
about 70% of all wastewaters which need treatment (Figure 2), although a substantial number of
the existing municipal wastewater treatment plants are overloaded.
Water pollution has become one of the most important environmental problems in
Poland, and this is magnified since the majority of industry in Poland is situated in the south near
the origins of the country's river systems. In addition, the main rivers, the Vistula and Odra, are
heavily used for municipal and industrial water supplies, agriculture irrigation, cooling purposes
for power plants, and navigation, and they also receive surface water runoff and discharges of
wastewater with varying degrees of treatment. These multiple uses impose competing demands
on waters, and water resource management must protect many desirable uses.
231
-------
1975 1980
mechanical treat. 1
1985 1990 1991 1992 1993
| chemical treat. p^v\ biological treat. I I no treatment
Figure 2. Wastewater treatment.
The principal water-quality problems in Poland are, therefore, related to: (1) effects of
municipal and industrial wastewater, including saline discharges from coal mines; (2) influence
of non-point sources, such as agriculture and urban stormwaters; (3) effects of dams and other
water management structures resulting from phenomena associated with impounded water; and
(4) effects of power plants, since the discharge of heat from cooling operations is considered to
be specific pollutant.
Monitoring System and Data Analysis
Poland has monitoring activities in rivers, lakes, coastal waters, and in the Baltic Sea.
The river water quality surveillance system is composed mainly of conventional monitoring
stations, and the country is covered by a network of stations at established cross-sections. The
sampling frequency depends on the purpose for which data are recorded, ranging from a
minimum of bimonthly sampling up to daily sampling at some points. The sampling of water is
performed simultaneously with the rate of flow measurements.
The river monitoring network provides a large number of data observations. These data
are analyzed by a statistical method based on the assumption that, at a given cross-section, some
correlation exists between the pollutant concentration and the rate of flow. The shape of the
curve depends on many factors, such as the degree of water pollution, the type of pollutant,
hydrological characteristics of the river, its self-purification capacity, the distance between
monitoring stations, and others. From these relationships between stream flow and
232
-------
concentrations of water-quality constituents, so-called indicative concentrations (1C) for a design
flow are established. The mean low streamflow (MLQ) has been selected as the design
streamflow at each site, based on the assumption that higher streamflows will result in higher
dissolved oxygen (DO) concentrations and better water quality. In other countries, a similar
approach has been taken. For example, in the United States the design flow is the minimum
average 7-day consecutive flow expected once every 10 years. However, this is an extremely
low streamflow which is exceeded more than 99% of the time. The 1C values are plotted along
the river for various water quality constituents. Final interpretation is based on these
hydrochemical profiles, and the overall river classification is performed after all measured water
quality constituents are compared with standards. A compendium of hydrochemical profiles for
major rivers and streams is prepared each year by the Institute of Meteorology and Water
Management (IMWM) which serves as an overall river classification system.
WATER QUALITY MODELING
Early BOD-DO Models
Mathematical modeling of river systems in Poland has become an integral of water
resources planning and water-quality management. These models can be used to aid water-
quality surveillance and to predict future water quantity/quality conditions (Gromiec et al. 1983).
Various computerized models have been applied for water quality simulations in the Odra and
Vistula Rivers. As an example, a Streeter-Phelps model (Streeter and Phelps 1925) and QUAL-I
model (Masch and Associates 1971) were used by IMWM to evaluate concentrations of
biochemical oxygen demand (BOD) in the Vistula River reaches. The first model is designed to
simulate the spatial and temporal variations in BOD under various conditions of flow and
temperature. The second model is capable of routing BOD, DO, and temperature through a one-
dimensional, completely mixed branching river system. These early BOD-DO models are
representative of non-conservative coupled models. It should be stressed that the predictions
obtained from these models are only as reliable as the input data, proper measurement, and
estimation of the various model parameters.
An Overview of the QUAL2E and QUAL2E-UNCAS Models
The Stream Water Quality Model QUAL2E (Brown and Barnwell 1987) is a steady state
model for conventional pollutants in one-dimensional streams and well-mixed ecosystems. The
conventional pollutants include conservative substances, temperature, bacteria, BOD, DO,
nitrogen, phosphorus, and algae (Figure 3). The model is widely used for simulation of water
quality and for waste load allocations and discharge permit determinations in the United States,
and it is a proven, effective analytical tool (Barnwell et al. 1987).
A major problem faced by the user when working with a complex model such as
QUAL2E is model calibration and determination of the most efficient plan for collection and
calibration data. This problem can be addressed by application of principles of uncertainty
analysis. QUAL2E-UNCAS is a recent enhancement to QUAL2E which allows the user to
233
-------
perform uncertainly analysis on the steady-state water quality simulations. The above models are
available from the National Exposure Research Laboratory, U.S. Environmental Protection
Agency (U.S. EPA) at Athens, Georgia.
ATMOSPHERIC
HEAERATION
Figure 3. Major constituent interaction in QUAL2E.
The U.S. - Poland Joint Project
A variety of microcomputer models for water quality simulation are supported by the
Center for Exposure Assessment Modeling (CEAM) at the U.S. EPA laboratory in Athens,
Georgia. The Athens laboratory, and the Institute of Meteorology and Water Management
(IMWM), have jointly selected a set of CEAM-supported computer models that address surface
234
-------
water quality problems of mutual interest. The QUAL2E and QUAL2E-UNCAS models have
been applied to tributaries of the Vistula river in Poland (Gromiec et al. 1994). About 30 major
rivers from the upper, mid and lower part of the Vistula river basin have been chosen for this
study. A summary of river characteristics is reported in Table 1.
Table 1. Summary of hydraulic characteristics of simulated rivers.
River
Total length (km) Studied length (km) Range of flow (m3/s) Number of reaches
Biebrza
Brda
Bug
Dlubnia
Drweca
Dunajec
Ilownica
Kamienna
Mala Wisla
Narew
Nida
Pilica
Poprad
Przemsza
Raba
Radomka
Radunia
Rudawa
San
Skawa
Sola
Suprasl
Tanew
Wapienica
Wda
Wieprz
Wislok
Wisloka
Wkra
155.3
238.0
587.21
38.0
207.2
247.1
26.5
138.3
97.1
448.01
151.2
319.0
62.61
87.6
131.9
107.0
104.6
. 27.7
442.1 .
96.4
88.9
93.8
113.0
21.1
198.0
303.2
204.9
163.6
249.1
102-0
161-74
580-36
34-0
185-1
200-0
15-0
128-6
96.5-3.5
384-57
116-22
281-1
62-0
86-6
116-2
92-2
75-18
16-0
304-4
82-4
81-2
84-0
70-0
11-0
62-0
269-62
168-6
145-3
210-18
1.37-5.70
4.93-22.5
13.00-81.7
0.18-0.49
1.70-28.94
17.10-50.5
0.31-1.28
: 0.10-5.99
0.54-9.64
4.75-91.30
0.49-13.4
: 1.16-27.8
13.30-17.6
0.14-17.80
0.49-15.30
0.14-4.97
2.25-4.97
0.72-1.50
28.60-81.90
0.16-6.72
0.32-11.70
0.12-5.50
2.74-5.67
0.14-0.37
6.61-11.58
,1.13-12.90
0.95-9.34
0.88-16.20
0.84-9.88
3
7
14
4
12
142
52
9
202
292
6
15
4
11
8
7
5
2
342
7
8
6
4
• 2
5
11
11
14
13
'On Polish territory
2Number of reaches involves reaches of tributaries ,
For water quality simulation of rivers in Poland the following conditions were chosen:
steady-state model and water quality constituents such as: temperature, DO, BOD, nitrogen
compounds (organic nitrogen, ammonia, nitrite, nitrate), phosphorus compounds (organic and
dissolved), coliforms, chlorophyll a, conservative constituents (chlorates and sulfates), and
arbitrary non-conservative (chemical oxygen demand).
235
-------
The water quality database, which has been used for preparing input files, contains
results from field measurements. Sampling points were located at important stream cross-
sections with respect to tributaries, point loads, and withdrawal points. These points were
located upstream and downstream from these cross-sections, and on tributaries just upstream of
the junction. Data considering characterization of municipal and industrial wastewater have been
taken from the Central Statistical Office (CSO 1992). Hydraulic data (flow, velocity, depth, etc.)
and meteorological data (dry and wet bulb temperature, cloudiness, wind speed, atmospheric
pressure and elevation) were taken from hydraulic and climatological annual reports (IMWM
1992a, 1992b). Many simulations with the QUAL2E model were performed. In general, most of
the results obtained from simulation of water quality constituents, which are transformable by
biochemical, chemical, or biochemical processes, were similar to measured values. However, the
temperature results obtained from simulation were lower than actual measured temperatures.
There was also a problem with mineral, constituents, since in the QUAL2E model only dilution is
considered and there is no process that allows mineral concentrations to decrease through
settling, for example. An example, of the simulation of various water quality constituents in the
Dunajec River is shown in Figure 4.
There are three uncertainty analysis options available in the QUAL2E-UNCAS model:
sensitivity analysis, first order error analysis (FOEA), and Monte Carlo simulation. Sensitivity
analysis gives the relative change in the value of each output variable resulting from the changes
in the value of the input variable. FOEA represents the percentage changes in the output variable
weighted by the variance hi each input. The Monte Carlo simulation gives summary statistics
and frequency distributions for the variables at specific locations in the system.
HO 66
OUNAJEC RIVER, km
Conforms «lm. + Conforms rmos.
20
Figure 4a. Simulation by QUAL2E of water quality in the Dunajec River -
Coliforms.
236
-------
12
11
10
9
8
7
6
5
4-
3
2
1
0
196
186
de ios
110
DUNAJEC RIVER, km
20
DO *lm.
DO
BOD slm.
BOD nvjas.
Figure
4b. Simulation by QUAL2E of water quality in the Dunajec River
Dissolved oxygen and biochemical oxygen demand.
12
11
10
9
8
7
6
5
4-
3
2
1 -
186
196
Ho
OUNAJEC RiS€R. km
COD «Im. •«• OOO
r
6$
20
Figure 4c. Simulation by QUAL2E of water quality in the Dunajec River •
Chemical oxygen demand.
237
-------
3
2.8
2.6
2.4
2.2
2
1.8
1.6
1.4-
1.2
1
0.8
0.6
0.4-
0.2
19J3 186'
136
118 806
110
DUNAJEC RIVER, km
38
20
NH4- *)m.
NH+ nvsas.
Figure 4d. Simulation by QUAL2E of water quality in the Dunajec River •
Ammonia.
2
1.9
1.8
1.7
1.6
1.5
1.+
1.3
1.2
1.1
1
0.9
0.8
0.7
0.6
0.5
0.4-
0,3
0.2
0.1
0
20
« N02*lm.
DUNAJEC RIVER, ten
A N03«lm. x HOJrmoe.
Figure 4e. Simulation by QUAL2E of water quality in the Dunajec River •
Nitrite and nitrate.
238
-------
0.2
0.19
0.18
0.17
0.16
0.15
0.1 +
0.13
0.12
0.11
0.1
0.09
0.08
0.07
0.06
0.05
0.0+
0.03 -
0.02 -
0.01 -
0
1 Sib 1 86
196
I
156
ds I os
110
DUNAJEC RIVER. Ion
PO4- *lm. •*• PO+ meas.
|
66
38
|
20
Figure 4f. Simulation by QUAL2E of water quality in the Dunajec River •
Dissolved phosphorous.
60
50
+0
30
20
10
196
o Chlorides si
1
T?W
«7
if
66
DUNAJEC RIVER, km
Chlorides m«os. •» Sulfoto* dm.
20
Sutfot** moos.
Figure 4g, Simulation by QUAL2E of water quality in the Dunajec River
Chlorides and sulfates.
239
-------
In these studies the first order error analysis with 5% input perturbation was chosen for
all input variables (hydraulic, water quality constituents, algae content). For each river in the
Vistula River basin an uncertainty analysis was completed. At least five locations were selected
in each river basin in performing the uncertainty analysis. The criteria for locations included
water quality at the beginning of each system, major tributaries (treated as river or point load),
wastewater discharges, and reservoirs. For incremental characteristics and headwaters, the
variance coefficients were 10%, and for the point loads they were from 2 to 5% greater. If
tributaries were considered, the variance coefficients for incremental, headwater, and point loads
had the same values. In general, first order error analyses have been performed for all of the
input variables (hydraulic, reaction coefficients, and point load forcing functions) for input
perturbation equal to 5%. The results of calculated relative standard deviation (RSD) for water
quality constituents are summarized in Table 2. The RSD values for all of the constituents are in
the range of low and typical values. At some points, for point loads or for upstream locations,
the RSD values are higher.
Table 2. Summary of relative standard deviation (%) for analyzed rivers.
River
Biebrza
Brda
Bug
DIubnia
Drwcca
Dunajec
Ilowmca
Kamienna
MalaWisla
Narew
Nida
Pilica
Przemsza
Raba
Radomka
Radunia
Rudawa
San
Skawa
Sola
Suprasl
Tanew
Wda
Wieprz
Wislok
Wisloka
Wkra
DO
2-6
1-3
2-11
2-3
2-5
2-4
2-3
2-3
3-4
2-45
2-4
2-3
0-32
2-3
2-9
3
2-3
2-4
2-6
2-3
2-5
2
2-4
3-30
2-4
2-3
2-3
BOD
7-8
7-8
8-10
8-10
7-12
7-21
7-10
5-8
7-9
6-12
6-7
8-?
8-11
8-11
8-10
7-10
7-10
6-10
7-8
8-10
7-9
9-12
7-10
8-11
' 6-8
7-24
8-10
N(org)
8-1 1
9-11
11-12
10-12
7-9
7-11
9-20
10-12
7-10
7-10
10-20
11-15
7-10
NH3
8-12
7-11
10-21
12-15
9-18
7-14
9-11
8-9
10-14
9-32
8-13
9-13
7-13
8-12
8-14
10-15
7-10:
IO.-14
7-12
.9-11
8-12
13-14
8-12
7-16
10-41
8-23
NO2~
12-15
10-17
11-20
, 0-22
8-25
12-35
12-16
13-15
14-17
9-15
14-20^
11- IS
13-95
12-13
12-17
17
6-10
10-20
10-15
12-26
10-20
17-20
12-18
8-11
' 8-15
10-20
NO3"
7-11
8-10
7-12
11-15
7-8
6-9
7
8
8-9
7-10>
6-8
8-9!
9-11
7-8
8-12
7-11
7-9
7-10
7-10
9-10
' 7-8
7-10
8-15
8-10
6
7-9
8-13
240
-------
Water Quality Management Models
In addition to prediction of water quality by simulation, mathematical models have served
as the basis for determination of investment policies in the construction of wastewater treatment
plants. For example, a water quality management model has been applied to the Klodnica River.
The main goal was to determine, for the given system of wastewater treatment plants, the level of
efficiency necessary to achieve the required standard of water quality at the least cost. The
quality of the Klodnica River catchment at the respective stages of the water pollution reduction
program has been determined. It should be stressed, however, that these management models are
only tools in assisting management decision-making processes. Final decisions are usually not
made solely on the basis of their predictions. Additional data, including socio-political factors,
are taken into account. Still, in spite of their limitations, these models are the only reasonable
means presently available for the prediction of water quality.
Table 2. (continued)
River
Biebrza
Brda
Bug
Dlubnia
Drweca
Dunajec
Ilownica
Kamienna
Mala Wisla
Narew
Nida
Pilica
Przemsza
Raba
Radomka
Radunia
Rudawa
San
Skawa
Sola
Suprasl
Tanew
Wda
Wieprz
Wislok
Wisloka
Wkra
P(org)
10-14
12-15
9^16
7-14
8-20
7-10
8-10
10-25
10-14
13
12-20
P043
7-12
8-10
6-10
10-14
8-11
7-9
8-9
5-8
8-13
8-9
8-12
8-9
8-9
7-8
8-12
10-14
7-9
7-10
7-10
8-10
7-9
9-15
7-10
7-10
8-10
7-8.
CM-1
2-3
2
2-5
2-3
2
2-3
2-3
3-6
2-3
2-4
2-3
3-5
2
3-5
2-3
2-3
2-4
2-3
2-3
2-3
2-3
3-5
2-11
. 2-3
3
2-3
CM-2 •
2-4
2
2-3
3-4
3-9
: 2-3
2
: 2-3
3-4
2-3
. 2-4
2-3
2-3
2-4
2-3
2-3
2-5 .
2-4
2-3
2-3
, 2-3
3-5
: 2-3
2-3
2-3
Coliforms
10-14
14-39
12-28
15-21
11-35
12-39
12-21
8-33
10-38
12-36
14-19
11-17
16-20
11-19
11-15
12-48
12-17
12-16 .
13-20
14-15
12-22
15-142
19-67
COD
6-8
7-8
7-10
7-10
6-8
7-12
7-10
7-10
8-9
7-12
6-8
7-8
7-9
7-8
8-11
6-9
7-9
6-10
7-9
8-15
8-10
6-10
7-10
7-10
6-8
7-21
8-10
241
-------
The stream water quality models, QUAL2E and QUAL2E-UNCAS, are now available in
Poland for wasteload allocation studies. A number of such applications is currently performed.
These computer programs can be used as an aid for solving conventional and non-conventional
pollution problems in Poland.
SUMMARY AND CONCLUSION
On the whole, Poland is a water-poor country where various conflicts over water resource
use and development are increasing. Water resources are unevenly distributed among different
parts of the country, water supplies now appear inadequate in quality, and demand is growing in
many regions. In the past, environmental policies in Poland did not adequately protect water
quality, with the result of serious deterioration. The negative consequences of this fact are borne
by all water users. A number of constraints make traditional environmental strategies less
effective than in the past. In addition, many traditional strategies for solving water quality
problems are extremely costly. Under the present economic situation financial resources are
inadequate to upgrade the whole treatment capacity in Poland at once. Therefore, a new strategy
is needed, how to invest these available resources so that the environmental benefit is optimal.
The investment program for water pollution control hi 1995 alone requires about 2.1 billion
zlotys (100 million USD).
In view of this commitment, the watershed protection approach, water conservation and
pollution prevention approaches, and the restructuring of industry are clearly needed to assure
use of the most cost-effective solutions to water resource problems in Poland. Water quality
improvement programs should also be aimed at improvement of treatment methods,
implementation of advanced treatment processes, recovery and water reuse in industry, and
encouraging the production of biodegradable detergents and pesticides along with use of dry
technologies.
The watershed protection approach is an integrated strategy for more effectively restoring
aquatic ecosystems and protecting health. This approach is focusing on hydrologically defined
drainage basins rather than on areas arbitrarily defined by administrative boundaries, and it
places greater emphasis on all aspects of water quality, including physical and chemical, habitat,
and biodiversity. The watershed protection approach provides comprehensive methods for
implementing trie solutions to water quality problems in Poland.
The most urgent water quality problems will be solved only if clear goals are established
and efficient ways to achieve each goal are identified. Setting priorities often involves difficult
choices. In many cases, priorities are pre-treatment of industrial wastewater, where heavy metals
or toxic chemicals threaten the quality of surface waters. Appropriate investments are necessary
to reduce and treat toxic discharges from industrial facilities, as well as to treat saline waters and
other discharges from mines. From a domestic perspective, wastewater investments should focus
on up-stream water. However, in regional terms, strategies should focus on the overlap between
local and the transboundary impacts of pollution reduction measures. This requires directing
financial resources at reducing flows of nutrients and emissions of harmful substances from
municipal, industrial, and agricultural sources, in order to achieve an overall reduction of
pollutants discharged into the Baltic Sea.
242
-------
The priority setting requires basic data about the extent and type of water pollution.
Monitoring water quality parameters in drainage basins-watersheds is essential for assessing the
status, the trends, and the causes of water quality conditions. However, the best method of
conducting such an assessment is not easily determined. The difficulty is due to a number of
considerations: the multiplicity of water quality constituents, natural water quality variations in
time and among locations, and the high cost of collecting and analyzing samples.
Mathematical modeling of river systems has become an integral part of water quality
management. These models can be useful to provide water quality surveillance and to predict
future water quantity/quality conditions, but they should be: easily applicable and usable for
planning and management by water authorities. There are two general types of water quality
models available. One results from a desire to achieve a more comprehensive understanding of
the physical, biochemical, and ecological processes that take place in water ecosystems receiving
potential pollutants, and the other type is directly oriented toward planning and management
and/or real-time control. The best selected model for planning and management will depend on
the information needed, which will differ for various water ecosystems, on management
alternatives, and on possible institutional objectives and constraints. Only through case studies
can we learn more about how to select appropriate model complexity and how to improve the
quality of information derived from models for the planning process.
ACKNOWLEDGMENTS
This paper is based on work sponsored by the Polish-American Maria Sklodowska-Curie
Joint Fund II in cooperation with the Polish Ministry of Environmental Protection, Natural
Resources and Forestry and the U.S. Environmental Protection Agency.
REFERENCES
Barnwell, T.O., L.C. Brown, and R.C. Whittemore. 1987. QUAL2E - A case study in water
quality modeling software. In Systems Analysis in Water Quality Management, Pergamon
Press, Oxford, England. ;
Brown, L.C., and T.O. Barnwell. 1987. The Enhanced Stream Water Quality Models QUAL2E
and QUAL2E-UNCAS: Documentation and User Manual, EPA/600/3-87/007, U.S. EPA,
Environmental Research Laboratory, Athens, Georgia, USA.
CSO (Central Statistical Office). 1992. Environmental Protection and Water Management.
Warsaw, Poland. 490 pp. (In Polish)
Gromiec, M.J., D.P. Loucks, and G.T. Orlob. 1983. Stream quality modeling. In Mathematical
Modeling of Water Quality: Streams, Lakes, and Reservoirs. G.T. Orlob (Ed.). John Wiley
and Sons, Chichester, England.
243
-------
Gromiec, M.J., R.C. Russo, T.O. Barnwell, M.D. Bielasik-Rosinska, and P. Bielobradek. 1994.
Final Report on Innovative Microcomputer Applications of Water Quality Models, Project
MOS/EPA-89-17, Institute of Meteorology and Water Management, Warsaw, and U.S. EPA,
Environmental Research Laboratory, Athens, Georgia, USA.
IMWM (Institute of Meteorology and Water Management). 1992a. Climatological Annual
Report. IMWM, Warsaw, Poland. (In Polish)
IMWM (Institute of Meteorology and Water Management). 1992b. Hydrological Annual
Report. IMWM, Warsaw, Poland. (In Polish)
Masch, F.D. and Associates, and the Texas Water Development Board. 1971. Simulation of
Water Quality in Streams and Canals. Theory and Description of the QUAL-I Mathematical
Modeling System. Report No. 128, PB 202975, Austin, Texas, USA.
Minister of Environmental Protection, Natural Resources and Forestry. 1991. Decree 11.05 on
Water Quality Classes and Requirements Set for Waste Water Discharge to Water Bodies or
to Soils. Dz.U. 1991, No. 116, Item 503.
Minister of Health and Social Welfare. 1990. Decree 05.04 on Conditions for Drinking Water
and Water for Industrial Purposes. Dz.U. 1990, No. 35, Item 205.
Parliament of Poland. 1974. Water Law Act. Dz.U. 1974, No. 38, Item 230.
Streeter, H.W., and E.B. Phelps. 1925. A Study of the Pollutions and Natural Purification of the
Ohio River. Public Health Bulletin No. 146. U.S. Public Health Service, USA.
244
-------
PATTERNS OF RESPONSE OF ZOOPLANKTON POPULATIONS TO TOXICANTS:
A MODELING STUDY
Yuri M. Plis1 and M. Craig Barber1
ABSTRACT
An approach to evaluate ecological exposure and effects based on equations describing
the transport of plankton population density and biomass, as well as aging and changing
physiological characteristics of individual organisms is presented. The developed model takes
into account means and variances of dependent variables. The model is applied to simulations of
seasonal dynamics of zooplankton population in the lake polluted by PCBs.
INTRODUCTION
The methods of ecological risk assessments were developed to evaluate the effects of
toxicants on aquatic ecosystems. Analysis of ecological risk, includes evaluations of the spatial
and temporal distributions of a stressor, its contact with the ecological component of concern,
and quantifications of the adverse effects elicited by a stressor. Mathematical models have
proven to be effective tools for analyzing different aspects of these evaluations.
Individual-based population models allow us to include a description of the actual
mechanisms and processes that determine the vital rates of the different classes of individuals
that make up the populations. These models can greatly increase our ability to estimate the
responses of a population to toxicants and other environmental stressors.
Our approach employs characterizations of exposure;and ecological effects based on
equations derived to describe the transport of population characteristics (density and biomass
composition), as well as aging and changing physiological characteristics of individual
organisms. The developed model takes into account the following physical and biological
processes: hydrodynamics, advective-diffusive transport, chemical and biological transformation
of pollutants, bioaccumulation, individual organism's life-history and population dynamics, and
lethal and sublethal impacts of toxicants (Plis and Lassiter 1996). The model is enhanced by
accounting for not only mean characteristics of aquatic populations but also the variances of their
dynamic and spatial distributions. In this work we demonstrate that natural fluctuations of
plankton populations can be simulated using a physiologically and spatially structured population
model that describes the life-cycle and Interactions of individual organisms composed of a
population. This aspect should increase the utility of the developed model for assessing the
ecological risks posed by toxicants or other physico-chemical alterations in aquatic communities.
'U.S. Environmental Protection Agency, Athens, Georgia, USA.
245
-------
The model is used to simulate zooplankton populations in the Twelve-Mile Creek area of
Lake Hartwell (South Carolina). Zooplankton are important and abundant animals in an aquatic
ecosystem and primary consumers of plant materials and detritus. Like fishes, zooplankton are
integrators of contaminants in aquatic environments, and are, therefore, effective indicators of
such stress. To collect field data about plankton population temporal-spatial distributions in
polluted water bodies is, however, very difficult, expensive, and time-consuming.
The Twelve-Mile Creek area is polluted by polychlorinated biphenyls (PCBs) that were
detected at all levels of the food chain. The results of simulations of zooplankton population
dynamics allow us to obtain the information that can assist to develop some decisions preventing
damages to aquatic environment from toxicant pollution.
MODEL OF PLANKTON POPULATION DYNAMICS
Basic Population Equations
Independent variables of a plankton population model consist of time t, spatial
coordinates AT, physiological properties of an individual plankton organism m, and individual's
age a. Introducing a distribution function I(t, x, m, a), such that / dxdmda is the number of indi-
viduals of the plankton population in the volume of a linear space of our independent variables
between x, m, a and x+dx, m+dm, a+da, we can describe the transport / by the classical
advection-dispersion equation combined with the equation of physiologically structured
population dynamics (Plis and Lassiter, 1996). To simplify this equation we integrate /over m
and derive the equation for an age-structured plankton population.
dN/dt+ dN/da + LN =- DN + Q,
CO
N= f Idm, .LN= Va (va
(1)
where va is the covariant derivative; v" is the contravariant component of a velocity of water
currents; D"p is the tensor of coefficients of turbulent diffusion; D and Q are, respectively, a
mortality rate, and rate of external sources of the population organisms.
To specify the equation (1) we use a mathematical model describing the age dynamics of
a plankton individual. In general, such dynamics can be described by a systemL of nonlinear
equations for elements of the m
= F (m, tr a) ,
(2)
where F, is a rate function, describing input and output flows of mass characteristic m.j
246
-------
The birth process for the population is described by a boundary condition
A
I(t, x,m, a=0) =f f(3 (trx,mr\f) I(tfxfmry) dmdy,
(3)
0 m
where A is a limiting age; /?$ jc, /M, j^) is a birth function that represents the number of neonates
born to an individual of age 7 and mass-characteristics m at time t.
i
In an effort to preserve the information about mass diversity between individual
organisms with the same age we consider mass characteristics m as a set of random variables.
Their diversity originates from the fact that individual organisms with different age and
corresponding mass characteristics have offspring of which mass characteristics differ also.
Let m = MJ be one of the components of the vector m for an individual organism at age a,
which was born by an individual at age y, and p is the probability of this event. We can
introduce following expressions for the mean m and variance a2 of m:
flz(t,x, a) =1 p(t, x, a, Y) m(t, x, a, Y) c*Y,
o
a2 (t, x, a) =f p(t, x, a, Y) [m (t, x, a) -m (t, x, a, y) ] 2 dy .
0 :
(4)
The probability/? should describe the mass distribution between individual organisms of
the same age. At time t and age o=0 this function is
p(t, x, 0,Y)=P(t,x,Y) [/ 3(t,x, y)N(t,x,
J
o
(5)
As result of the assumption about the homogeneity of the mortality D for individual
organisms of the same age we can obtain a condition of the invariance of p with the age a as
p(t+a, x, a, Y) = p(t,xf 0, Y) • (6)
ZOOPLANKTON INDIVIDUAL MODEL
For this research we used the model describing the life history and bioenergetics of an
individual daphnid and toxicant bioaccumulation that was developed by Hallam et al. (1990).
An individual daphnid is represented as consisting of two body components: structure ms, and
lipid mL. Its dynamics are depicted by the rates of change of these two components, as controlled
by energy supply and demand. The balance of toxicant mT in an organism is formed under the
impact of the toxicant diffusion from ambient waters and inflow with feeding food. The
equations of the individual growth and toxicant bioaccumulation are discussed in details by Plis
and Lassiter (1996).
247
-------
Following Hallam et al. (1993), sublethal effects of nonpolar narcotics on growth of
individual organisms can result in reducing the growth rates of the lipid and structure
components of the individual daphnid model by the factor (1-r), where
r=1.378 (C -0.0147 9 K
ow
(0.0395K
0
(7)
Kow is the octanol-water partition coefficient, and CA is the concentration of toxicant in the
aqueous phase of an organism and satisfies lO'1-83 Kow~l z CA < lO'0'8 KOWA. This representation
of sublethal effects results in the growth reduction occurring continuously as a response to
arising the internal toxicant concentration. Lethal effect occurs at a CA of 10'°-8 Kow~l. The
coefficients of equation (7 ) were obtained on basis of experimental studies (Hermens et al.,
1985).
ALGORITHM OF SOLUTION
Advection-Diffusion Transport
Using the assumption about the ability of plankton to migrate vertically, the spatial
differential operator of equation (1) was written using curvilinear orthogonal in the horizontal
plane coordinates (Plis 1992). The advantages of applying these coordinates to plankton
population modeling are discussed by Plis and Lassiter (1996).
Numerical Algorithm
The operator splitting method has been applied as an approximate technique to integrate
equation (1) over an arbitrary time interval At. This integration can be written as three separate
integrals, i.e.,
t+At t+At _ t+At
t t a t
Each of these integrals uses solutions of the previous iteration as initial conditions. The same
procedure should be applied to integration of equations describing dynamics and spatial
distribution of populational: lipid M,=NmL, structure M2=Nms, and accumulated toxicant
M3=Nmr
The finite difference approximations of the first integral in equation (8) and equations for
Mh M2 and M3 were obtained using the QUICK scheme originally developed by Leonard (1979).
These solutions were used for deriving the rates of advective-diffusive transport of individual
organisms in the algorithm for calculations of populational mass-characteristics M.
The second integral in equation (8) is solved by implementing the 'coherent' cohort
model analytical solution (Brewer 1989). The third integral is solved by a second-order explicit
Runge-Kutta method. For the plankton population structured by A age groups we should apply
the same procedure for each of these groups.
248
-------
Age-Dynamics of m and o2
Let m(m,t,a) be a solution of equation (2) at the initial condition m(m,t-a,a)=m(t-a,a) and
m(m, t, a, y) =m(m, t, a) +e (m, t, a, y) , (9)
here e supposed to be small comparatively with m.
Applying the Euler-Cauchy approximation method to equation (2) and using the Taylor-
series expansion ofF(m,a) about m = m we can obtain with the order of magnitude O(e2) that
m(t+Aa,a+Aa)=m(t+Aa, a+Aa) ,
a2(t+Aa, a+Aa) =J][ g2 (t-a+iAa, iAa) a2 (t-a, 0) .
(10)
i=0
where k Aa = a, g(a) = 1 + (dF/dm) m=*(a).
Spatial Transport of m and a2
The transport of the mean E and variance Var of the populational mass-characteristic Mat
(ij)-th cell of the grid approximating a water body area can be calculated as
1 1
t, j+m ffii+Jc, j+m '
£ £
l — X K— J. J71—~*X
(11)
where p is a covariance coefficient;
-------
APPLICATION AND CONCLUSIONS
Lake Hartwell is a well-documented case of a xenobiotic impacting an ecosystem. PCBs
contamination in the surface water and sediment in the upper portion of the Twelve-Mile Creek
arm of the lake has been attributed to discharges from a capacitor manufacturing plant. The plant
used several varieties of PCB-containing dielectric fluids. The PCBs were primarily Aroclors
1242,1254, and 1016. The transport of PCBs absorbed on suspended solids represents the
primary pathway for the downstream transport of PCBs in Twelve-Mile Creek and Lake
Hartwell. A significant mass of PCBs has been transferred to fish and other biota in Hartwell
Lake: this is a critical component of the contaminant transport processes in the lake.
In 1993 Bechtel Environment Inc. issued the remedial investigation report for surface
water and sediment in Twelve-Mile Creek area of the lake. The investigation includes the wide
variety of field and modeling results. Some results of these studies were used as the initial
conditions, parameters and scenarios for our model of zooplankton population dynamics.
The simulations gave us the information about tendencies of seasonal dynamics of spatial
distributions of zooplankton's populational and individual physiological characteristics: total
biomass, the ratios of lipid to total biomass, mass of the accumulated toxicant, number of
organisms, and age-mass distributions in polluted and unpolluted areas of the lake. Numerical
densities of zooplankton populations depend strongly on available food resources
(phytoplankton) and tract one another; increasing during warmer and decreasing during cooler
seasons. Physiological conditions of a low body lipid are unfavorable for reproduction, and this is
depicted by the low numerical density that exists during winter-spring period. During summer-
fall period the reproductive potential of zooplankton population is high, and small, young organ-
isms dominate the daphnid numerical density distribution. The seasonal dynamics of the
populational relative lipid content (a lipid/biomass ratio) generally follow the dynamics of the
zooplankton numerical density and phytoplankton biomass distributions. It shows that for
population and for an individual organism, the lipid fluctuates more in response to environmental
factors than does body structure or total biomass.
An important characteristic for the zooplankton population is the distribution of age at
which individual first reproduce. In our individual model, this characteristic depends on the lipid
and structure contents and decreases from the unfavorable winter-spring seasons to the summer
season.
The toxicant inhibition is more significant during winter-spring period unfavorable for
population growth. The patterns of spatial distributions of population numerical density and total
biomass are very sensitive to values of toxicant concentrations. Under some scenarios the
maximum values of these populational characteristics were relocated from the areas with
increased food densities to areas with decreased concentrations of toxicant.
250
-------
The approach described here provides both diagnostic and prognostic information beyond
that available from traditional approaches to the evaluation of ecological effects. These equations
and algorithms should be effective practical tools for exploration and evaluation of ecological
risk to plankton populations, and can produce much valuable information at a level of resolution
not previously considered feasible.
REFERENCES
i
Brewer, J.W. 1989. Spreadsheets, PC's, and the finite-difference solutions for ecological
distribution. Ecological Modeling 47:65-83.
Hallam, T.G., R.R. Lassiter, J. Li, and L.A. Suarez. 1990. Modelling individuals employing an
integrated energy response: application to Daphnia. Ecology 71: 938-954.
Hallam, T.G., G.A. Ganziani, and R.R. Lassiter. 1993. Environmental Toxicology and
Chemistry 12: 947-954.
Hermens, J., E. Broekhuyzen, H.Canton and R. Wegman. 1985. QSARs and mixture toxicity
studies of alcohols and chlorohydrocarbons: Effects on growth of Daphnia magna.
AquaticToxicology 6:209-217.
Leonard, B.P. 1979. A stable and accurate modelling procedure based on quadratic upstream
modelling. Computational Methods in Applied Mechanical Engineering 19:59-98.
Plis, Y.M. 1992. An approach to calculating wind-driven currents and transport of substances in
unstratified water bodies using curvilinear coordinates. Water Resources
Research 28 : 83-88.
Plis, Y.M., and R.R. Lassiter. 1996. An approach to modeling the dynamics and spatial
distribution of a structured zooplankton population hi a toxicant polluted lake.
Nonlinear World 3:129-150.
251
-------
-------
GENETIC ALGORITHMS FOR CALIBRATING WATER QUALITY MODELS
Linfield C. Brown1 and Ann E. Mulligan2
ABSTRACT
Genetic algorithms (GAs) are a class of computer techniques which have been found to
perform well on mathematical optimization problems, particularly those that are difficult for
more conventional techniques. GAs are investigated for their potential use as a calibration tool
in water quality modeling. The steady state Streeter-Phelps dissolved oxygen model is used to
model water quality in a riverine system. Synthetic data with and without error are used initially
to investigate the GA. Estimation was aided by the inclusion of multi-response data such as
CBOD and NDOD concentrations. Constraints are successfully included in the GA search using
either a penalty function or a special decoding operation. However, the GA with the penalty
function outperforms the GA with the decoder. For ill-posed problems, the GA provides several
parameter estimates, all performing equally well mathematically. The GA is found to be a useful
water quality model calibration tool because it is not constrained by the mathematical nature of
the model and it searches from a population of points, thereby providing useful information
about the parameter space.
INTRODUCTION
This paper presents the initial results of a project to investigate the use and applicability
of genetic algorithms (GAs) to the calibration (i.e., parameter identification) problem in water
quality modeling. GAs are computer-based programs originally created to simulate the process
of evolution by natural selection and genetic inheritance. They were first described by John
Holland in his seminal book entitled Adaptation in Natural ;and Artificial Systems (Holland
1975). Subsequent theoretical and empirical research on GAs (De Jong 1975, Brindle 1981,
Booker 1982, Goldberg 1983) has proven their applicability in complex optimization and control
applications (Goldberg 1989). The relationship between the optimization terminology and the
biological analog of the genetic algorithm is described in Table 1.
GA's have met with remarkable success in optimization problems in the fields of biology
(Grosso 1985, Sannier and Goodman 1987), computer science (Gerardy 1982, Rendell 1985),
engineering and operations research (Goldberg 1983, Glover 1986, Minga 1986), image
processing and pattern recognition (Gillies 1985), and social sciences (Smith and De Jong 1981,
Axelrod 1985). In the water resources and environmental engineering literature applications
'Department of Civil and Environmental Engineering, Tufts University, Medford, Massachusetts, USA.
Department of Civil and Environmental Engineering, University of Connecticut, Storrs, Connecticut, USA.
253
-------
have been described by Wang (1991) for calibrating rainfall-runoff models; by McKinney and
Lin (1994) and Ritzel et al. (1994) for groundwater management problems such as water supply,
groundvvater remediation, and hydraulic control; and by Simpson et al (1994 ) for sizing water
distribution networks. GA applications to water quality model calibration have yet to appear in
the literature.
Table 1. Terms used to describe a genetic algorithm.
Real space optimization
Biological analogue
Objective function
Decision variable
Plays role of environment, value of the objective function at any point
is that point's fitness
Gene (often encoded in binary strings)
Vector of decision variables
Chromosome - genes strung together
Encoded chromosome: genotype (here we only use haploid chromosomes,
but more generally, the genotype is the entire genetic package)
Decoded chromosome: phenotype
Model calibration has been recognized as the single most difficult step in the process of
modeling water quality (Barnwell and Brown 1986). Calibration is an optimization task, often
employing heuristic trial and error procedures. Such procedures may result in suboptimal and
unrealistic input parameter sets in order to fit the model to the data, and rarely account for spatial
and cross correlation among the parameters.. GAs conduct efficient searches, provide
information about the search space (parameter correlation) for statistical inference, and result in a
mathematically defensible, physically realistic set of input parameters.
Traditional optimization techniques, such as random search (trial and error) or gradient
search, have been used to regress field data and identify water quality model input parameters.
Although the research literature is replete with examples of these methods, they have not found a
niche in the state-of-the-practice. Gradient search algorithms require a differentiable objective
function and are susceptible to getting trapped in suboptimal regions of the search space. GAs
have been demonstrated to outperform traditional optimization methods in many applications,
particularly those with noisy, discontinuous, or multinodal objective functions characteristic of
multivariate water quality models (Grefenstette 1986).
The short term objective of the project is to demonstrate the applicability of genetic
algorithms to the calibration of water quality models. Preliminary studies at Tufts University
(Mulligan 1995) have provided strong evidence for the effectiveness of GAs in water quality
model calibration. Our objective is to build on this initial work to solidify the position of GAs as
a credible method in the modeler's arsenal of calibration tools.
254
-------
The longer term project goals are: (1) to develop procedures whereby water quality
modelers can use GAs in routine calibration scenarios, and (2) to incorporate these methods into
a broader base of model calibration and forecasting tools, i.e. expert systems and advisors, for
improving the models used in managing and protecting the; quality of the Nation's surface waters.
The eventual goal is to incorporate GA methods into model codes, such as was done by Brown
and Barnwell (1987) in adapting uncertainly analysis methods into the water quality model
QUAL2E. . ;
GENETIC ALGORITHM METHODOLOGY
The GA methodology involves a search process which is initiated by randomly
generating a population of potential solutions to an optimization problem (Figure 1). Each
individual in the population is evaluated by the objective function and then ranked according to
performance (or fitness). In water quality modeling applications, fitness is typically assessed by
comparing observed and predicted values of water quality state variables. A second generation
of potential solutions evolves by selecting individuals from the previous generation. Those
individuals exhibiting the best performance (highest fitness) will have a higher probability of
being selected into the next generation. Following selection, genetic operators such as crossover
and mutation are applied randomly to the new generation (Figure 2). Crossover serves as an
information exchange process between individuals while mutation adds variability to the
population and guarantees that all locations in the search space have a chance of being searched.
Once the second generation is created, the new individuals are evaluated with the objective
function and the entire process is reiterated.
First Generation
(n = 0)
Initialize Population
Evaluate Fitness
of all Population
Members
Select Individuals
for Next Generation
»- End
Figure 1. Diagram of a genetic algorithm.
255
-------
Generation t
• Crossover
One-Point Crossover
00:
1.1 1.1
0000
Two-Point Crossover
• Mutation
i i
1.111 11.1.1
00
00!
0.0
1 1 1:1.1 1
Mutation point
Generation (t + 1)
1 1OOaO
0101 1,1 1
110011
0.0:1:1.0.0:
111 01 1
Figure 2. GA crossover and mutation reproduction methods.
After numerous generational cycles, the population in the GA evolves towards solutions
of unproved performance. Because GAs evolve entire populations, much information can be
gleaned about the complex search space. In other words, the population as a whole contains
information concerning the sensitivity of the objective function to the decision variables (in this
case the water quality model input parameters and variables). This is a significant advantage
because search techniques based on point-to-point search (such as trial and error and gradient
searches) contain no such knowledge.
MODEL SELECTION AND METHODOLOGY
Preliminary studies (Mulligan 1995) conducted to assess the applicability and utility of
GAs hi water quality model calibration are very encouraging. The context of these simulation
experiments is the steady-state Streeter-Phelps dissolved oxygen (DO) model (Thomann and
Mueller 1987):
D(X) =
L,*Kd\
Kr Kd\
— exp
where: D(x) = dissolved oxygen concentration at downstream distance, x, Lo - initial BOD
concentration, Kd = first-order BOD decay rate, Kr = reaeration rate, x = distance downstream
from point source, U= average stream velocity, and Do = initial DO deficit. The GA used in this
application, GENESIS, was developed by Grefenstette (Grefenstette et al. -1991). Reproduction
is done using Baker's stochastic universal sampling technique, which is a simple algorithm
256
-------
designed to implement reproduction with zero bias and minimum spread (Baker 1987). Two-
point crossover and mutation are the only genetic operators and are applied at rates of 60% and
3%, respectively. No attempt was made to optimize the genetic algorithm in these initial
experiments. While GAs can easily accommodate a wide variety of objective functions, the least
squares criterion was used so that GA results could be easily compared to non-linear regression
methods (Marquardt 1963).
Experiments were performed to address the following issues: (1) to demonstrate that the
GA works in the calibration of simple single and multiple reach systems, (2) to investigate
techniques for incorporating input parameter constraints (correlation) in the calibration
procedure, and (3) to examine the ability of GAs to estimate parameters from multi-response
data.
PRELIMINARY RESULTS
Single and Multiple Reach Systems
Initial experiments were performed on a one-reach, one point source model, using
synthetic field data created by adding randomly generated error to the exact DO sag curve. Both
the non-linear estimator based on the Marquardt algorithm and the GA were able to find
comparable optimum parameter values for each data set. This finding provided the first direct
evidence that GAs could actually be used in calibration. For simple, relatively smooth response
surfaces, the GA converged to parameter values that were essentially the same as those obtained
from an established regression procedure.
After successful implementation of this simple case, subsequent experimentation
included a more complex riverine environment with multiple point-sources and multiple reaches
as shown in Figure 3. Given initial conditions in the stream and point source effluent data, the
GA was used to estimate two parameters (reaeration rate and BOD rate) for each reach in the
stream, using synthetic data points located throughout the system. Results of these simulations
(Figure 3 a) show excellent agreement between values of the parameters estimated by the GA and
the true values.
The GA was also used to estimate joint confidence regions for the parameters estimated
by the GA. The entire population of the GA can then be scanned for individuals which define
the confidence region. Figure 3 (panels b through e) presents a plot of GA individuals (each
individual represents a set of potential input parameter values) for which the sum of square
residuals is within 5% of the value defining the 90% joint confidence region. As can be seen, the
confidence region is well defined in the parameter space, and exhibits some interesting
characteristics. There is a significant correlation between different parameters (panels b and c) in
the same reach and between the same parameter in different reaches (panels d and e).
257
-------
a. DO Sag Estimated by the GA
10
8 -•
Saturated DO concentration
Reach 0 Reach I
Ka Kd Ka Kd
GAEstimale 3.49 0.604 0.684 1.42
TiueVahres 3.50 0.60 0.70 1.50
100 150 200
Distance from head of Reach 0 (km)
250
300
-*3.75 •
I
£
s
<3 n c
§
1 '
o
C£
0.
b. Reach 0: 90% Confidence Region
o
o 0)-, o
o "^bS °°
I 1 t I 1 I I
5 0.55 0.6 0.65 0.
BOD Rate, Kd, (day1)
7
Figure 3. DO sag and parameter correlations for a multiple reach estimation problem.
258
-------
0.5
c. Reach 1: 90% Confidence Region
1.5 1.75
BOD Rate (day1)
d. Reaeration Rate: 90% Confidence Region
3.4 3.6 3.8
Reach ORearation Rate (day1)
e. BOD Decay Rate: 90% Confidence Region
§1.6
1.2
rr^*
OSJ
039 Q£l ' OA3
Rod) 0 BOD Rale (day1)
FigureS. Continued.
259
-------
The evolution (over 200 generations) of the best parameter set within the problem context
is illustrated in Figure 4. At the beginning of the simulation (generation 0) all estimates are
randomly generated. The best initial individual (parameter set), while capturing the overall shape
of the data, is unable to predict observed data accurately. By the 10th generation (1000
individuals have been sampled), the best estimate has improved considerably in both reaches.
The best estimate at the 100th generation (half-way through the simulation) is very close to the
optimum. Finally, only slight improvement is seen over the second half of the simulation.
Parameter Constraints
The pattern of spatial (reach-to-reach) correlation of a single parameter led to an
investigation of using the GA to do a constrained calibration. Parameter constraints were
investigated using synthetic data from a four reach, two point source, system shown in Figure 5.
In this system, the CBOD rate coefficient kd was constrained from reach to reach in the following
manner: k^ < kdl> k^ k^. Five data points were given for reach 0 and two data points for
reaches 1 and 3 , and only one data point for reach 2. There are therefore two sources of
variability hi the parameter estimates: Stochastic nature of the GA leads to different estimates
when the GA does not fully converge to the pptimum, and the small number of data points in the
three downstream reaches makes the solution in those reaches nonunique.
GA performance was investigated for three estimation scenarios:
Unconstrained: The GA was allowed to proceed as if there were no constraints.
Penalty Function: Individuals that did not satisfy the constraints were penalized by
reducing their fitness by an amount proportional to the deviation from the constraint boundary.
Decoding GA: Downstream parameters are mapped as functions of the upstream
parameters and an independent gene indicating the degree of correlation between the parameters.
The results for the range of estimated parameter values for each of the constraint options
are shown in Figure 6. The unconstrained GA always performed most efficiently, but often
resulted in fits that did not satisfy the parameter constraints. The decoding GA always
performed most poorly. Because of the mapping structure in the decoding GA, the real number
value of a parameter in reach 3, for example, results from the interaction of four upstream genes.
Any slight change hi an upstream gene will propagate downstream and influence all downstream
parameter values. This complicates the search and leads to inefficiencies in convergence to a
solution that will satisfy the parameter constrains, and provide good fitness. The penalty
function GA worked the best in that it produced sets of parameter estimates similar to the
unconstrained GA, and did so while meeting all the constraints and converging to the same best
performers very nearly as rapidly as the unconstrained GA. .
260
-------
<{/Iffl)
(I/3™)
261
-------
u
to
a
CO
I
1
1
o
o
I
CO
a
4
o
GO
,262
-------
I-
o o i
S I i
JO1I3 y, - J31W1QS3
i i
'S
i-
i § § . i s
1-
SS
«
8 §
*_^H ttQ
to >->
D X
!-l O
O 1>
O
-------
-M Hi
s s
••1
2
it
= .8
f
:0
"O
o
§ o
cfci Q
oo d
O
•0)
H
ro
5
.m
264
-------
Multiple Response Estimation
In addition to the experiments described above for estimating parameters from a single
DO response, several experiments were conducted to investigate the impact of including the
CBOD responses in the estimation problem. Although the two objective functions (one for DO
and the other for CBOD) are not directly comparable, an evaluation the range of parameter
estimates is possible, and enlightening. The range of parameter estimates obtained when both the
DO and CBOD responses are simultaneously fit to the data are shown in Figure 7. Clearly the
range of fit individuals (good parameter estimates) is nearly an order of magnitude smaller in the
mulitiresponse case, when compared to the fits from the single (DO) response shown in Figure 6.
The unconstrained GA still includes individuals that do not satisfy the parameter constraints and
the decoding GA again converges inefficiently. Thus we conclude that the GA with penalty
function constraints for spatial correlation in the parameters is potentially well suited for multi-
response estimation and the results are precise.
SUMMARY
The genetic algorithm presented here has been shown to accurately find an optimum set
of parameter values in a parameter estimation (calibration) problem. Although slower than
traditional non-linear estimation techniques, the GA provides a significant advantage in its ability
to define confidence regions in the parameter space. When calibrating a model it is both useful
to know the optimum parameter values and important to understand both the correlation between
parameters and the joint confidence regions of all parameters. The penalty function GA appears
to have the capability of incorporating parameter correlation in a constrained search for
appropriate parameter values. Multi-response estimation clearly provides a set of calibrated
parameter values that are more precise than those from a single response estimation.
FUTURE WORK
This preliminary study indicates that GA applications in model calibration may be very
useful and should be further studied. Because the definition of confidence regions and parameter
correlations is so important to model calibration, the next steps in experimenting with the GA
should include:
1. Further testing and evaluation of these findings by extending GA applications to
realistic case studies. Issues to investigate include identifying and correcting potential GA
deficiencies that may arise when: (a) increasing the number of stream segments and process
parameters to more realistic levels, and (b) increasing the number of state variable responses to
be fitted (e.g. nutrients, chlorophyll-a, and toxics).
2. Improving GA efficiency (i.e., convergence and run time) by investigating such
techniques as variable genetic operators (as proposed by Davis 1987, and Srinivas and Patnaik
1994) and local fine tuning.
265
-------
3. Investigating a dual objective function which seeks to optimize the parameter
estimates in terms of their capability to estimate observed response values as well as to agree
with observed or measured input parameter values.
The long range goal is to develop a calibration methodology that provides a computerized
data base of sets of site specific input parameters and variables that yield good fits to the
observed data. Relying solely on best fit parameters is troublesome, given the multitude of errors
and professional judgement involved hi model calibration. Initial experiments have
demonstrated the potential feasibility of this application. GAs can perform as well as or better
than traditional non-linear regression and have the added advantage of supplying a wealth of
knowledge about the search space. The potential of GAs to accomplish this long range goal is
very appealing.
REFERENCES
Axelrod, R. 1985. Modeling the evolution of norms. Presented at the American Political
Science Association Meeting, New Orleans, Louisiana.
Baker, J. E. 1987. Reducing bias and inefficiency in the selection algorithm. In: Genetic
Algorithms and Their Applications: Proceedings of the Second International Conference on
Genetic Algorithms, J.J. Grefenstette (Ed.). Lawrence Erlbaum Associates, Publishers,
Hillsdale, New Jersey.
Barnwell, T.O., Jr., and L.C. Brown. 1986. Development of a Prototype Expert Advisor for the
Enhanced Stream Water Quality Model QUAL2E. Internal Report, U.S. EPA Environmental
Research Laboratory, Athens, Georgia.
Booker, L.B. 1982. Intelligent behavior as an adaptation to the task environment. Doctoral
dissertation, Technical Report No. 243,University of Michigan, Ann Arbor, Logic of
Computers Group. Dissertation Abstracts International 43(2), 469B, University Microfilms
No. 8214966 (as referenced in Goldberg 1989).
Brindle, A. 1981. Genetic algorithms for function optimization. Ph.D. Thesis, Univ. of Alberta,
Alberta (as referenced in Baker 1987).
Brown, L.C., and T.O. Barnwell, Jr. 1987. Documentation and User Manual for the Enhanced
Stream Water Quality Models QUAL2E and QUAL2E-UNCAS. U.S. EPA Environmental
Research Laboratory, Athens, Georgia.
Davis, L., (Ed.). 1987, Genetic Algorithms and Simulated Annealing. Pitman Publishing Co.,
London, U.K.
266
-------
De Jong, K.A. 1975. An analysis of the behavior of a class of genetic adaptive systems, Doctoral
dissertation, University of Michigan, Ann Arbor. Dissertation Abstracts International 36(10),
5HOB, University Microfilms No. 76-9381 (as referenced in Goldberg 1989).
Gerardy, R. 1982. Probabilistic finite state system identification. International Journal of
' General Systems 8: 229-242.
Gillies, A.M. 1985. Machine learning procedures for generating image domain feature detectors.
Unpublished Doctoral Dissertation, University of Michigan, Ann Arbor.
Glover, D.E. 1986. Experimentation with an adaptive search strategy for solving the keyboard
design/configuration problem. Unpublished Doctoral Dissertation, University of Michigan,
Ann Arbor.
Goldberg, D.E. 1983. Computer-aided gas pipeline operation using genetic algorithms and rule
learning. Doctoral Dissertation, University of Michigan, Ann Arbor. Dissertation Abstracts
International 44(10), 3174B. University Microfilms No. 8402282.
Goldberg, D.E. 1989. Genetic Algorithms in Search, Optimization, and Machine Learning.
Addison-Wesley Publishing Company, Inc., Reading, Massachusetts.
Grefenstette, JJ. 1986. Optimization of control parameters for genetic algorithms. IEEE
Transactions on Systems, Man, and Cybernetics, SMC-16, no.l, 122-128.
Grefenstette, J. J., L. Davis, and D. Cerys. 1991. Genesis & OOGA: Two Genetic Algorithm
Systems. The Software Partnership, Melrose, Massachusetts.
Grosso, P.B. 1985. Computer simulation of genetic algorithms: parallel subcomponent
interaction in amultilocus model. Doctoral Dissertation, University of Michigan, Ann
Arbor. University Microfilms No. 8520908.
Holland, J.H. 1975. Adaptation in Natural and Artificial Systems. The University of Michigan
Press, Ann Arbor.
Marquardt, D.W. 1963. An algorithm for least-squares estimation of nonlinear parameters.
Journal of the Society of Industrial and Applied Mathematics 11: 431 -441.
McKinney, D. C., and M-D. Lin. 1994. Genetic algorithm solution of groundwater management
models. Water Resources Research 30: 1897-1906. ;
Minga, A.K. 1986. Genetic algorithms in aerospace design. Presented at the AIAA Southeastern
Regional Student Conference, Huntsville, Alabama.
267
-------
Mulligan, A.E. 1995. Genetic algorithms for calibrating water quality models. M.S. Thesis,
Department of Civil and Environmental Engineering, Tufts University, Medford,
Massachusetts
Rendell, L.A. 1985. Genetic plans and the probabilistic learning system: synthesis and results.
Proceedings of an International Conference on Genetic Algorithms and their Applications, p.
60-73.
Ritzel, B. J., J. W. Eheart, and S. Ranjithan. 1994. Using genetic algorithms to solve a multiple
objective groundwater pollution containment problem. Water Resources Research 30: 1598-
1603.
Sannier, A.V., II and E.D. Goodman. 1987. Genetic learning procedures in distributed
environments, genetic algorithms and their applications. In: Proceedings of the Second
International Conference on Genetic Algorithms. J.J. Grefenstette (Ed.). Lawrence Erlbaum
Associates, Publishers, Hillsdale, New Jersey.
Simpson, A. R., G. C. Dandy, and L.J. Murphy. 1994. Genetic algorithms compared to other
techniques for pipe optimization. Journal of Water Resources Planning and Management
120:423-443.
Smith, T., and K.A. De Jong. 1981. Genetic algorithms applied to the calibration of information
driven models of U.S. migration patterns. Proceedings of the 12th Annual Pittsburgh
Conference on Modeling and Simulation, p. 955-959.
Srinivas, M., and L.M. Patnaik. 1994. Adaptive probabilities of crossover and mutation in
genetic algorithms. IEEE Transactions on Systems, Man, and Cybernetics, 24: 656-667.
Thomann, R. V., and J.A. Mueller. 1987. Principles of Surface Water Quality Modeling and
Control. Harper Collins Publishers Inc., New York, New York.
Wang, QJ. 1991. The genetic algorithm and its application to calibrating conceptual rainfall-
runoff models. Water Resources Research 27: 2467-2471.
268
-------
HYDRODYNAMICS AND WATER QUALITY MODELING
OF A RIVER DELTA WITH LIMITED AVAILABLE DATA
Marina G. Yereschukova1
ABSTRACT
This report is concerned with a model that can be put into a framework for the modeling
of a river with several branched channels, and for which detailed data are not available. Because
water quality processes are affected by water flow and sediment transport, the present model
simulates hydrodynamic characteristics of a water system, and uses the results of previous
calculations for water body simulation. The model is based on the equations of continuity and
momentum in integral form. The algorithm for the branched channel system is described. This
model was applied to a simulation of the lower Don River, and some of the results of this study
are presented hi this report.
INTRODUCTION AND THEORY
Water quality assessments usually require data of observations gained with high
frequency hi time and in space. When detailed data are not available, the application of
mathematical models becomes very important, although modeling in turn also needs information
about the object of study. Water quality processes are effected by water flow and sediment
transport. So one way to model a system is to simulate the hydrodynamic characteristics of that
water body, recalculate the sediment transport, and then use the results of previous calculations
for water quality simulation. This report is concerned with a model that can be put into a
framework for the modeling of a river with branched networks or channels. The model is based
on the following equations of continuity and momentum in the integral form (Koutchment 1980):
at
n2Vabs(V)
Equation (1)
R
4/3
•) +
where Q = water discharge; Y = water depth; A = cross-sectional area; V = water velocity;
q = lateral inflow; g = acceleration of gravity; i0 = bottom slope; n = Manning roughness
coefficient; R = hydraulic radius; W = wind speed; a = angle between channel direction and
wind speed; Cst = wind stress coefficient. • . ;
'Hydrochemicial Institute, Rostov-on-Don, Russia.
269
-------
Equation (1) can be solved with respect to water discharges and water depths. For this purpose
the cross-sectional area and wetted perimeter (or hydraulic radius) is approximated by the
functions of water depth (Amein and Fang 1970):
A = -ao
P = b0
brxY
b2
Equation (2a)
Equation (2b)
where % and bt are constants specified before simulations. To obtain these constants the
algorithm of non-linear regression is used.
To simulate the flow dynamics hi branched rivers and/or hydraulic networks these
equations are solved numerically using the finite-differences method. For this purpose the
following conceptions about model structure and solving procedure are made. The river reach
and its branches to be modeled are divided into segments, the physical properties of which may
be assumed uniform. The intersections among segments or points where there are external
boundaries are called junctions. For using the implicit approximation scheme the four variables
(water discharges and depths at each and of the segment) are put into correspondence to each
segment. For a river reach divided into N segments, 2*N equations may be written by
discretizing the domain into distinct segments and applying Equations (1) with Equations (2a)
and (2b) substituted hi each. Another 2*N equation can be obtained from the boundary
conditions specified for each junction.
Following the approach of Schaffranek et al. (1981), the boundary conditions between
segments can be deterrnined for the water depths and discharges:
Equation (3)
Yk =
= 0
where: Qj} Yk, Yj = discharge, and water depths at the ends of the segments respectively;
Ij = set of indices corresponding to junction i.
If L segments meet at a junction, then L-l linear independent equations can be written for
the water depths and an equation for the water discharge. Hence, at this junction, L conditions
can be written. Another set of L conditions can be obtained for the other end of the segments at
the junction in the same way. At the external boundary one condition can also be specified. The
boundary conditions for the portion of the branched network may be determined in different
ways. There are different opportunities depending on available information for each boundary:
the time variant values of water depth or discharge given as a table, the time variant depth given
as harmonic function of tune, and functional relationship between discharge and depth. Thus,
2*N equations can be written, and altogether 4*N equations and unknowns may be obtained.
270
-------
To find approximate solutions for Equation (1), the four-point implicit scheme on a
time-space grid network is adopted. This scheme is defined by the following formulas:
^- « —(f""1
dt 2AtUn+1
dx
y "
/m , s-m+l rm \
n+1 Jn ~ J n/
1 ft
rm+1 \ . J. - Is
Equation (4)
Ax
f rm _ fm,
(J n+1 J »/
n*l
f"- \
J n/'
where: 0 = weighing parameter; At = time step; Ax = space step corresponding to the
segment length; fnn = function value at the n-th point in space and m-th lapse in time.
The choice of parameter 0 allows the use of different approximation schemes from
explicit (when 0 = 0) to fully implicit (when 0 = 1) and their combinations ( when 0 < 0 < 1).
After approximation, the continuity and momentum equations in the integral form by the scheme
expressed in Equation (4) and inclusion of conditions in Equation (3), 4*N non-linear equations
are obtained. For their simultaneous solutions the Newton-Raphson iterative procedure (Amein
and Fang 1970) is used.
To simulate sediment transport in a river's branched networks the same approach can not
be applied. The reason is the difficulty in specifying the internal boundary conditions between
segments for sediment. Following Hosseinipour and Martin (1991), the basic equations for the
sediment transport simulation is the equation of continuity:
0 . VB +
dt
dx
= Is
Equation (5)
where: Z - bed elevation; \ = porsoity of bed material; ys = sediment specific weight;
Qs = sediment volume flux; qs = sediment lateral inflow; B = hydraulic width.
271
-------
After replacing the derivatives by their explicit finite-different analogues, the equation
for determining the increment of bottom elevation due to advection-deposition processes can be
obtained. To solve this equation the sediment flux needs to be determined. This flux is
calculated by the formula of Yang (1973):
W
= 5.435 - 0.284\og— - 0.
Vis
(1.799 - 0.409logWoverVis - 0.
~*S VcrS .
W
W
Equation (6)
where: Vis = the kinematic viscosity of water; Vcr = critical velocity for incipient motion;
W = particle fall velocity; S = surface water slope.
The values of Vis, ¥„, W and U* are calculated according to the values of sediment
characteristics, morphometric parameters of the river segments, water depths, discharges and
hydraulic radius. The latter can be obtained as a result of simultaneous solution of Equation (1).
To apply these equations to the branched networks, some particular assumptions have to
be made. The bottom slope at both ends of the segments are considered to be equal and to be
determined by the bottom elevation of segments ends. The critical incipient and shear velocities
and sediment flux corresponds to the end of each segment. The average changes of bottom
elevation correspond to each segment. The changes of bottom elevation at the junction supposed
to be caused by the influence of all sediment fluxes connected with that junction and their effects
are considered to be independent. After these assumptions, the sediment yield can be calculated
and the results can be used in the sediment continuity equation for determining bottom elevation
changes due to deposition-degradation processes. The increment of bottom elevation at each
junction is calculated as the algebraic sum of bottom elevation increments of segments connected
with current junction.
The sediment flux at the boundary cross-sections can be specified if such information is
available, but in many cases it is quite difficult or even impossible to obtain this information.
So, the values of sediment fluxes at the external boundaries can be calculated with Equation (6)
by using the values of water depth and discharge and hydraulic radius, which can be given as the
boundary conditions for water flow or can be obtained through calculations.
272
-------
APPLICATION TO THE LOWER DON RIVER
The model was applied to the lower Don River, Russia. The Don is a relatively large river
approximately 1,870 km in length and has a watershed area of about 422,000 km2. Tsimlyansk
Reservoir is situated in the end of the middle section of the: river. The stretch of the Don River
from Tsimlyansk Reservoir to the Taganrog Gulf is about 313 km and is referred to as "the lower
Don River system." The river width varies from 400 to 600 m in the lower part, and the average
depth during low-water regime is in the range of 4-6 m in the main channel and decreases to
0.7 m within the shoals. There are about 96 shoals on the lower Don River. The mouth of the
Don River lies downstream from Rostov-on-Don, and the watershed of this cily covers an area of
about 340 km2. The river mouth includes several branches and creeks, the biggest of which are
Stary Don and Bolshaya Kalancha. The scheme of this portion of river is presented in Figure 1.
Razdorsiaya
Melikhouskaya [2) ? '* *
T- cross-sectioh
(nj- S8^n9iit.
number
Azov
9 8765432l
Figure 1. The scheme of the lower Don River system.
Model testing and validation were conducted on the; basis of data gained from the routine
monitoring system of the Russian Federal Service for Hydrometeorology and Environmental
Monitoring. The model parameters were identified on a set of data corresponding to the period
of time July 1-10,1979 with discharges changing from 900 to 1,000 mVsec. After that, the model
was run with initial and boundary condition corresponding to other periods of time. One of the
data sets covers the period of September 1-10,1979 with discharges changing from 616 to 660
nrVsec. The results of the simulation are presented on the Table 1.
273
-------
Water stages were measured with respect to Baltic Sea system. Water depths changed
from 3 to 6 m. Hence the error of simulations did not exceed 5% of the absolute values of water
depth, which is less than the guaranteed accuracy of observation.
Table 1. Water depth (m) at the upper portion of the lower Don River system,
September 1-10,1979.
Melikchovskaya
Bagaevskaya
Starocherkassk
Aksai
Time (date) Sim1
Obs.
Sim.
Obs.
Sim.
Obs.
Sim.
Obs.
1
2
3
4
5
6
7
8
9
10
2.31
2.33
2.25
2.22
2.23
2.25
2.24
2.22
2.19
2.25
2.33
2.39
2.35
2.33
2.33
2.35
2.33
2.32
2.21
2.31
1.33
1.28
1.15
1.11
1.12
1.14
1.15
1.15
1.17
1.22
1.36
1.32
1.16
1.10
1.12
1.16
1.21
1.22
1.20
1.22
0.78
0.50
0.30
0.26
0.24
0.29
0.39
0.40
0.47
0.54
0.81
0.53
0.26
0.18
0.17
0.22
0.41
0.33
0.47
0.52
. 0.72
0.31
0.12
0.18
0.19
0.20
0.22
0.24
0.32
0.46
0.68
0.22
0.08
0.04
0.06
0.18
0.26
0.20
0.25
0.36
'Simulated
'Observed
Another example corresponds to the period of time February 5-15,1979, when discharges
changed from 916 to 2,080 nrVsec. The water stages at the part of the Don River from
Razdorskaya to Rostov-on-Don are presented on the Figure 2. The graphs correspond to the
moments of time when discharge at the upstream boundary equaled 2,080 and 1,700 mVsec. It
might be noticed, that the absolute values of error increase comparatively with the previous table,
but the values of water depths increase also up to 5-9 m. Thus the errors do not exceed the
guaranteed accuracy of observations, which is about 10% of the measured values.
The testing of sediment transport simulation was conducted on hypothetical data because
of lack of observational data. It is known that the main parts of river bed material are sands with
different particle size from 0.07 to 0.20 mm. The results of simulation showed that the river
capacity for sediment transport is very uneven from segment to segment. This capacity is higher
at the marginal segments and lower at the middle segments of the river portions considered.
These results conform to the fact that shoal formation is proceeding intensively in the lower Don
River system.
Sediment transport capacity depends of the size of particles. Sand with particle diameter
Of about 0.07 mm are transported with the discharges from 0.0021 up to 0.036 mVs. For particles
with diameter size about 0.20 mm, discharges are sufficiently less and vary from 0.0004 to 0.005
mVs. The branches have different sediment transport capacity, and sediment discharges at the
more shallow branches are greater than at the deeper ones.
274
-------
Q = 2080 mVs
tr-
io
1
calculated
obsecvad
Calculated
Raedorstaya Kelilhovsiaya Basaevstoya surochertesst Atsal Rostov-on-Don
Q = 1700 m3/s
01
tn
cd
Observed
Calculated
Raidorstaya
Bagaersfcaya Staroctertasst Atsal Bostor-Oa-Bc
Figure 2. Water stages in the lower Don River system with respect to Baltic Sea level.
275
-------
As noted above, the last cross-section where discharges were measured is near
Razdorskaya, so the water discharge simulation at the mouth of the river is very important for
water quality modeling on this portion of the river. Results of water discharge simulations from
May to October 1983 are presented in Figure 3.
Rostov-on-Don
Hay
June July August Septenber October
AZOV
-------
The results of these simulations were used for water quality simulation using the WASP4
modeling package (Ambrose et al. 1988), which was aimed both to detect the crucial parameters
affecting the river ecosystem, and to determine more precisely the trophic level of this water
body. The model used for this purpose has to contain the stated variables for the description of
algal biomass, and organic and inorganic nutrients. Some of the results of this water quality
simulation are shown on the Figures 4-6.
0.7
.0.6
JO. 5
go.4
•H
4-3
£0.3
•P
Total Inorganic P
J L
Organic P
m .
_L
J I L
_L
May
June
July
August September October
Ammonia Kitraftot InorganicOiBjanic PAmQonia Nitraftot InorganKQ rBjanic P
MeasuredMeasured Measured Measured
Figure 4. The dynamics of nutrients near Rostov-on-Don.
277
-------
a °-4
•H
rt 0.3
§ 0.2
u
s °-1
Hitrate
Total- Inorganic P
flnmonla
gany
'i i i —r
i i
May
June
July
i i
August September October
Ammonia NitratfEot InocganlcQfganic F
Ammonia Nitraraot InorganicO^antc FMeasured Measu):ed Haasuted Measured
Figure 5. The dynamics of nutrients near Dugina.,
Q.7
tn
a
J-5
(d
"S
O)
u
o
o
0.4
0.3
0.2
0.1
*
5K
Nitrate
viy
TP: -
Total Orqanic P
. i i i
. I ...l._ J
-5K
May June
Aaaonla HitrafEot InorganioO£ganic E
July August September October
Ammonia HitraCEot InorganioOJE^anic P.
Measured Measured Measured Measured
'
Figure 6. The dynamics of nutrients near Azov.
'278
-------
ACKNOWLEDGEMENTS
This study was conducted as part of bilateral research by investigators from the Athens
Environmental Research Laboratory, U.S. Environmental Protection Agency, Athens, Georgia,
USA, and the Hydrochemical Institute, Russian Federal Service for Hydrometeorology and
Environmental Monitoring, in Rostov-on-Don, Russia. The author is grateful to Rosemarie C.
Russo, Robert B. Ambrose, and Zia E. Hosseinipour for their collaboration.
REFERENCES
Ambrose, R.B., T.A. Wool, J.P. Connoly, and R.W. Schanz. 1988. WASP4, A Hydrodynamic
and Water Quality Model - Model Theory. User's Manual, and Programmer's Guide, US
EPA/600/3-87/039, Athens, Georgia, USA. :
Amein, M.M., and C.S. Fang. 1970. Implicit flood routing hi natural channels. Journal of
Hydraulics Div., Dec., pp. 2481 -2499.
Hosseinipour, E.Z., and J.L. Martin. 1991. RIVMOD - A One-Dimensional Hydrodynamic and
Sediment Transport Model. Model Theory and User's Manual.
Koutchment, L.S. 1980. Models of River's Flow Forming Processes, Leningrad,
Hydrometeoizdat, 144 p. (in Russian)
Schaffranek, R.W., R.A. Baltzer, and D.E. Goldberg. 1981. A Model for Simulation of Flow in
Singular and Interconnected Channels. U.S. Geological Survey Techniques of Water
Resources Investigations, Book 7, Chapter C3.
Yang, C.T.., 1973. Incipient motion and sediment transport! Journal of Hydraulics Div., 1973,
Oct., pp. 1679-1701. ;
279
-------
-------
THE SUSPENDED PARTICLE SURFACE: A NANOSCALE AND HOLISTIC VIEW.
George W. Bailey1, Sergey M. Shevchenko27, Y. Shane Yu3, and Huamin Gan4
ABSTRACT
Surfaces of suspended particles in receiving waters control the speciation, sorption,
mobility, bioavailability, transport, toxicity, and exposure concentration of inorganic and organic
pollutants. In order to predict a pollutant's fate (which is the outcome of various biogeochemical
processes), we must understand the relationship between surface properties ~ structure,
morphology and chemical reactivity — and the property of the pollutant speciation. Scanning
probe microscopy (SPM) provides an excellent analytical approach to characterize in situ the
structure, morphology, and chemical reactivity of suspended particle surfaces -- mineral, humic
substances, and microbial. SPM (including scanning tunneling and atomic force microscopy) was
used to image the surface of vermiculite, graphite, and the three micas. Similarly, images were
taken of the reaction products of Fe(IH), Cr(III), La(III), and Pb(II) on muscovite at different
pHs. In a similar manner, SPM images were taken of fulvic acid, humic acid, and 1,10-
phenanthroline on muscovite and graphite surfaces. The technique was suitable to image natural
weathering products on the surface of vermiculite. Morphological properties of the phyllosilicate,
humics, and oxide reaction product surfaces were determined by a variety of image analysis
techniques -- autocovariance, bearing ratio, fractal dimension, power spectral density, roughness,
and sectional analysis. Virtual reality visualization techniques were employed and found to be
useful in more easily comprehending the morphology, structure, and "complex landscapes" of
organic and inorganic environmental surfaces.
Mica, vermiculite (in the absence of natural weathering products), and graphite all had
nearly flat surfaces even viewed from the atomic perspective. Metal hydrolysis and subsequent
precipitation reaction products occurred and were observed on mica surfaces, the morphology
being related to the nature of the metal and the pH. Both linear aliphatic chains and aromatic
rings were noted in the morphological fabric of peat, fulvic acid, and humic acid residing on
muscovite and graphite surfaces. The valence of the saturating cation bound on the phyllosilicate
mineral determined the morphology of such organic macromolecules bound on mineral surfaces.
'Ecosystems Research Division, National Exposure Research Laboratory, U.S. Environmental Protection Agency,
Athens, Georgia, USA.
2National Research Council, c/o U.S. Envioronmental Protecton Agency, Athens, Georgia, USA. Currrent address:
Department of Chemistry, University of Victoria, Victoria, British Columbia, Canada.
3DynCorp-TAI, Inc.,. Athens, Georgia, USA; Current Address: Equifax, Inc, Alpharetta, Georgia, USA.
"•National Research Council, c/o U.S. Environmental Protection Agency, Athens, Georgia, USA. Current address:
Engelhard Corporation, Gordon, Georgia, USA.
281
-------
A combined application of all the image analysis techniques permitted a more definitive
description of surface roughness, the distribution of surface height, and the periodicity of
molecular arrangement of the surface. SPM shows great promise in defining the morphology ~
micrometer level and nanoscale levels of atomic resolution — of composite environmental
surfaces Computational chemistry — molecular mechanics and molecular dynamics — will play a
major role in understanding and predicting the energetics of environmental surface-mediated
reactions..Sorption energies can be calculated as well as conventional Kds.
The composite environmental surface — not the mineral, the organic matter or the
microbial surfaces — is viewed as the "real" environmental surface. It is this surface in the "real"
world that controls pollutant speciation, sorption, mobility, bioavailability, transport, toxicity,
and exposure concentration.
INTRODUCTION
Suspended particles in surface waters exert both a positive and negative influence on
water quality, and on fish growth and reproduction. These particles may bind inorganic (metals
for example) and organic contaminants (pesticides, toxic and hazardous chemicals) lowering their
equilibrium concentration in water and/or their bioavailability and therefore lessening their adverse
impact on both fish and their crucial sources of food in the food web. The term "environmental
surface" will be used to indicate both the physical and chemical properties of the surface, e.g.,
mineral surface and the location where the speciation process is occurring. Similarly, these
surfaces may bind pathogenic organisms again lowering their availability to fish. Conversely, a
high load of suspended particles may raise turbidity to levels that would lessen light penetration
and adversely impact a crucial step in the food chain, or "silt over" fish spawning beds reducing
successful spawning. Also high suspended particle loads, particularly those high in clay-size
particles (< 2 urn in equivalent spherical diameter), may cause physical damage to the fish gill
physiology/oxygen transfer mechanisms or cause desorption of aluminum ions/ aluminum
hydroxy-polymers or transition metal species and enter into the fish causing an adverse
toxicological reaction.
To optimize fish protection and water quality management, we must understand and
predict the physical, chemical, and mineralogical properties of "real" environmental surfaces, the
nature of the physical and chemical reactions (both from a thermodynamic and kinetic
standpoints) occurring at these surfaces, the effect of "structured water" on contaminant
properties, and the effect of environmental factors on these interactions. What is meant by the
term "real" environmental surface(s)? Traditionally, we have said that there are three-types of
environmental surfaces — (a) mineral, (b) organic matter, and (c) microbial. Mineral families
include silicates, oxides, sulfides, carbonates, and phosphates Their structure and properties are
treated in several different references (van Olphen 1963, Greenland and Hayes 1978, Dixon and
Weed 1989). Organic matter can be subdivided into humic substances (humic acid, fulvic acid,
and humins ), polysaccharide, proteins, lignins, waxes, and organic acids. Several good reviews
cover the nature and properties of these components of organic matter (Schnitzer and Khan 1972,
282
-------
Gjessing 1975, Thurman and Malcolm 1983, Stevenson 1994, Greenland and Hayes 1978, Hayes
et.al. 1989, Suffet and MacCarthy 1989, Shevchenko and Bailey 1996). Families of
microorganisms occurring in aquatic and terrestrial ecosystems include bacteria, fungi,
actinomycetous, and algae. With the exception of Mollicutes, all bacteria are bounded by a cell
wall (Beveridge and Koval 1981). Metal ions can react with the anionic surface of bacteria,
which serves as a template or nucleation sites for epitaxial development (process of
biomineralization) from metal cations complexed by the cell wall (Beveridge et al. 1983,
Beveridge and Doyle 1989, Hoyle and Beveridge 1989). We have inferred that the mineral
surface is mono mineralogic, has a homogeneous surface (i.e., it is free from surface defects), and
does not have an organic coating on part or all of that surface. We also infer that a microbe is
not bound on the mineral surface and that the mineral is not affixed to a microbial surface. The
former occurs when the mineral particle is sand-or silt-size and the latter occurs when the particle
is not sand-or silt-size. :
We believe that in the "real" world the "real" environmental surface is a composite surface
— a composite of mineral, organic matter, and a microbial cell surface. In the case of the mineral
surface, it is not a pure mineral surface but it is: (a) multimineral phasic — the principal mineral
phase could be a phyllosilicate like montmorillomte, but the other mineral phase(s) e.g., Fe, Al, or
Mn oxides(s) can occur as clusters, or islands, or as an epitaxial growth on the parent mineral
surface or edge and be either crystalline or amorphous; (b) .heterogeneous in surface structure
and electronic properties; and (c) coated with organic matter (humic acid, fulvic acid, or a
combination of both) of variable thickness and indefinite structure. Organic matter may also be
present in the particulate form, "dissolved" form (probably colloidal in size and properties) as
well as coating or partially coating the mineral surface.
What we must do is examine the morphology (macro-scale structure) and the atomic
scale, i.e., the nanoscale of this "real" environmental surface (morphology and atomic scale
resolution of resultant surface(s), processes by which one mineral phase may be bound to
another), determine how organic matter is bound to mineral surfaces(s) and determine how a
contaminant is bound to the nearly ubiquitous organic coating or microbial surface. Mechanisms
for binding organic matter to a mineral surface include ( a) cation bridging between exchangeable
cation (particularly Al3+, Fe3+, Cr3+, Ca 2+, and Mg 2+) of phyllosilicate minerals to carboxylate
groups on the organic matter, (b) dangling metal bonds on oxide surfaces or phyllosilicate edges
to carboxylate groups on organic matter, (c) formation of a metal-ligand complex between an
exchangeable transition metal on the mineral's exchange complex and N-S-P ligand of the
organic matter, (d) hydrogen bonding, and (e) "hydrophobic bonding."
Depending on the size of the organic-mineral composite particle, the microorganism may
be bound on the composite surface influencing the nature and rate of binding reaction with the
contaminant. The microorganism will limit these reactions due to the physical coverage of the
reaction sites on the composite surface as well as coagulation reactions. This applies to sand-and
silt-size particles. The opposite is true for medium and fine clay-size particles. These particles
will be bound on various surface aspects of the microorganism.
283
-------
The invention of scanning tunneling microscopy (STM) (Binnig and Rohrer 1982) and
atomic force microscopy ( AFM) (Binnig et al. 1986) has revolutionized the field of surface
science in the last decade. STM and AFM may be the two greatest analytical advances made in
surface chemistry permitting the characterization of the morphology, electronic properties and
atomic structure of humic substances. STM and AFM are potentially the most powerful analytical
techniques for understanding structures and morphologies/topographies of macromolecules.
AFM reveals the magnitude of attractive or repulsive forces on a variety of surface types
including nonconductors, with subangstrom resolution, essentially allowing the imaging of surface
atoms and molecules (Golovchenko 1986; Albrecht and Quate 1987, Hansma and Tersoff 1987,
TersofFand Hamann 1983, Hansma et al. 1988, Drake et al. 1989, Hartman et al. 1990). Due to
the high resolution characteristics of scanning probe microscopy, along all three crystallographic
axes, the structures and reactivities of mineral surfaces can be better understood, thus overcoming
some weaknesses of low-energy electron diffraction (and scanning electron microscopy)
techniques. STM and AFM analyses have been performed on solid surfaces in vacuum, in air, and
under water and other fluids (Hochella et al. 1989). Therefore, SPM is a powerful tool for
providing nanometer - and micrometer - scale images of "real" environmental surfaces — from
minerals to environmental polymers to microorganisms
Molecular mechanics (Wipff 1994, Burkert and Allinger 1982) provide the theoretical
capability to simulate and evaluate alternative conformations of high molecular weight compounds
as well as "macro conformations" of molecular aggregates, and find out the most favorable
conformation with respect to the mineral or organic surface. Coupling SPM and computational
chemistry provides a deeper insight into the nature of chemical processes that proceed on
environmental surfaces and that inevitably depend on the morphology of those surfaces. Bailey et
al., (1997) demonstrated the feasibility of coupling these two techniques to simulate the solvation
and energetics of a model humic substance, citric acid, on the surface of muscovite.
Coupling SPM images with virtual reality (VR) software to aid in the analysis of
environmental surfaces has been recently proposed (Shevchenko et al. 1997). Human perception
imposes certain limitations on how we understand, feel, or perceive graphic images from SPM,
and further development of visualization techniques is required to make outputs from the
instrument more brain-friendly. Versatility is needed whereby an image can be viewed rapidly
from a variety of points and ultimately flight animation over a surface can be generated.
Therefore, hundreds of images are generated in a minute and an immediate understanding of the
nanoscale surface is possible.
In this paper we will: (a) show images of the surface morphology of mineral, humic
substances, and metal reaction products present on environmental surface using SPM, and
determine their morphological characteristics using several different imaging techniques; (b) use
coupled SPM-VR to visualize better and analyze environmental nanoscapes; (c) set forth the
concept of the composite environmental surface as the "real world" surface and how such
surfaces control contaminant speciation, sorption, bioavailability, mobility," transport, and
exposure concentration in aquatic ecosystems.
284
-------
MATERIAL AND METHODS
Starting Materials
Standard Suwanee River fulvic acid (1S101F), Suwanee River humic acid (1S101H), and
peat were obtained from the International Humic Substances Society (IHSS, Golden, Colorado
USA). Indian muscovite and phlogopite were obtained from Ward's Natural Science'
Establishment Inc. (Rochester, New York USA). Transvaal vermiculite and Bancroft biotite
were obtained from Dr. Charles B. Roth, Agronomy Department, Purdue University, West
Lafayette, Indiana USA. Four pure-metal nitrate salts (certified American Chemical Society
grade) - Cr(NO3)3j Fe(NO3)3 La(NO3)3 and Pb(NO3)2 -were obtained from Fisher Scientific Co.
(Pittsburgh, Pennsylvania USA), Na^O, and 1,4-dioxane were obtained from Aldrich Chemical
Company, Inc (Milwaukee, Wisconsin USA) and were used without further purification. They
were used for replacing native exchangeable cations on the mineral surfaces. Stock solutions of
0.5M Cr(N03)3, 0.5M Fe (NO3), 0.5M La(NO3) 3 and 0.1 M Pb(NO3)2 were prepared. A highly
ordered pyrolytic graphite (HOPG, ZYA grade, Union Carbide Coating Service Corp. Cleveland,
Ohio USA) was used as the standard STM substrate. High-grade purity dimethylforamide (DMF)
(99.5% purity, Aldrich Chemical Co.) was used as a solvent. Drierite was obtained from the
W.A. Hammond Drierite Co. (Xenia, Ohio USA).
STM Sample Preparation
A graphite surface was freshly cleaved using a small piece of transparent adhesive tape and
placed in a small glass container (25 ml) with Drierite. Immediately, 5 uL of a sample suspension
(50 mg/L in water, 1,4 dioxane or DMF) was transferred to the substrate surface. After 24 hours
of drying at room conditions in a small container over Drierite, the sample was placed on the
microscope stage and scanned under room temperature and atmospheric conditions.
AFM Sample Preparation
The muscovite, phlogopite, biotite and vermiculite samples were prepared by cutting
several less than 10 x 10 mm pieces from a larger sheet using a pair of scissors or a scalpel. Great
care was taken to avoid disturbing or scratching the center region of the flake where the AFM tip
eventually "touches down." The tops of the phyllosilicate samples were freshly cleaved using a
piece of transparent adhesive tape. Tweezers were used to mount the flakes on the AFM sample
puck, which was precoated with a very thin layer of glue. Light pressure on the edges of the flake
during mounting assured firm contact. i
Reaction products of Pb(II), Cr(III), and Fe(IH) on muscovite were formed in a four-step
process. First, the freshly cleaved, puck-mounted flakes were immersed in prepared 0.1 M
Pb(NO3)2 or 0.5 M Cr(NO3)3 solutions for about 12 hours. During this treatment, K+ ions on the
mica surface were replaced by Pb(II) or Cr(III) ions. Second, the reacted flakes were removed
from the salt solutions and washed three times in deionized water to remove the excess salt.
285
-------
Third, the Pb, Cr, and Fe -saturated flakes were air-dried and water at different hydrogen ion
concentrations was added ( 0.1 M HNO3 and 0.1 M NaOH solutions were used to adjust the pH).
For Pb 2*-saturated muscovite, the pH's were 3.8, 6.1, and 12.4; for Cr3+-saturated muscovite,
the pH's were 3.0, 6.8, and 10.8. and Fe -saturated muscovite was 5.8. Fourth,, after a 1 hour
reaction period, the flakes were washed again and air-dried. The reaction products oathe
muscovite flakes were then ready for AFM imaging. The La3+ -saturated sample was prepared in
the following manner. The cleaved muscovite was placed in 10 ml La(NO3) (8.02 x 10"2 M)
solution. After 4 hours, the flake was taken out and dipped into deionized water three times to
remove the excess salt and each time a freshly prepared 20 ml of deionized water was used. The
La 3*-saturated flake was immediately transferred to a dry tissue with the cleaved surface facing
up. The excess water under the flake was absorbed by the tissue quickly and the flake mounted
on the puck. The puck was placed in a small glass container which had some Drierite to adsorb
the moisture. The sample was analyzed under room conditions.
Five microliters of the fill vie acid solution (50 ml/L in water or in 1,4-dioxane) was
transferred to the metal ion coated mineral surface. After 2 hours of drying over Drierite in a
small desiccator the sample was removed and analyzed on the AFM. This same general
procedure was followed to prepare humic acid, or peat samples for imaging. The untreated
muscovite sample is K4- saturated. The Transvaal vermiculite, having the natural weathering
product, was freshly cleaved and immediately analyzed.
Scanning Tunneling Microscopy
ANanoScope HI scanning tunneling microscope, manufactured by Digital Instruments,
Inc., Santa Barbara, California USA, equipped with a Pt-Ir tip (0.64-cm length, 0.03-cm
diameter) fitted with a 12 urn scan head was used. Prior to instrument engagement, the tip-surface
distance was adjusted to about 0.1 mm by manually lowering the scan head. The initial operating
parameters were: scan size, 10 um; scan rate, 10 Hz; bias voltage, 25 mV; set point, 0.5 nA;
integral gain, 5; and proportional gain, 5. In order to obtain images of the highest possible
resolution, the above parameters were optimized during the real-time scans.
Atomic Force Microscopy
A NanoScope HI scanning probe microscope (Digital Instruments, Inc., Santa Barbara,
California, USA), which contains an AFM, was used. A detailed description of the microscope
and the operating parameters can be found in Gan et al. (1996).
Image Presentation in Virtual Reality
The virtual reality method has been described previously (Schevchenko et al. 1997). The
Vistapro 3.0 software (Virtual Reality Laboratories, Inc., San Luis Obispo, California USA) is
designed for three-dimensional" presentation of the U.S. Geological Survey's Digital Elevation
Mapping (DEM) files.
286
-------
RESULTS AND DISCUSSION
Phyllosilicates
A known background structure is crucial in identifying or distinguishing between the
unknown compound and the background structure. Therefore, micaceous minerals may provide
great opportunities to observe the molecular structure of many nonconductor materials such as
organic and biological materials. Pauling (1930) established the layer-silicate concept of a series
of silicate minerals including mica. Detailed muscovite studies conducted by Jackson and West
(1930, 1933) confirmed Pauling's layer-silicate concept. The hexagonal holes were found to be
ditrigonal in nature due to the strain put on the tetrahedral layer to "fit" the larger octahedral layer
along the b axis. (Radoslovich 1959, 1960, 1961, Radoslovich and Norrish 1962). More general
studies about micaceous minerals and their roles in soil constituents, chemical reactivity, and
weathering process, have been summarized in various publications (Loughnan 1969, Dixon and
Weed 1989, Greenland and Hayes 1978). AFM images of muscovite, biotite, phlogopite, and
vermiculite are shown in Figure 1 and the morphological characteristics are presented in Table 1.
Muscovite. Structurally, muscovite is a dioctahedral mica having no octahedral charge,
zero-Fe in the structure and where 25% of the Si4+ is replaced by Al3+ in the tetrahedral sheet. '
Phlogopite is the trioctahedral end-member where only Mg2+ is in the octahedral sheet while
biotite has both Mg 2+, Fe 2+, and Fe 3+in the octahedral layer. All three micas have K as the
interlayer cation and are 2:1 nonexpansible minerals. Muscovite was a transparent flake and
showed a gentle, wavy surface in the 10 x 10 urn image (Figure 1-1). It was also found that the
vertical distance (or z range) of the surface was a fraction of the size of the view. Based on the
proposed structure of muscovite (Mauguin 1928, Pauling 1930, Brindley and MacEwan 1953),
the basal oxygen groups constitute the cleaved basal surface. The unit cell dimension of muscovite
is a = 5.63 A, b =9.0 A. Yoder and Eugster (1955) and Smith and Yoder (1956) proposed that
the unit cell dimension of muscovite was a = 5.20 A and b = 8.99 A. Therefore, our measurements
were very close to the proposed values. Muscovite has potassium ions in the structure (present
in the ditrigonal hole) as counter ions balancing the net negative charge due to isomorphic
substitution of A13+ for Si4+ (Brindley and MacEwan 1953), and can be replaced on the cleaved
surface by other divalent or trivalent ions via the process of ion exchange. It was found that when
trivalent ions replaced the monovalent ions, the surface property of the muscovite became
hydrophobic. Thus, this trivalent ion-coated muscovite can serve as a good substrate for very soft
biological materials in the method of AFM sample preparation. Also, this should facilitate the
sorption of hydrophobic macromolecules like humic acid or, anthropogenic chemicals on the
muscovite surface.
Biotite. Interesting morphology was observed in the images of Bancroft biotite (Figure 1-
2). This image is after fast Fourier transformations had been done. For the micrometer-scale
image, torn surfaces were observed. When the sample was cleaved, it was possible that the
cleaving force might cause the next layer to become deformed. Some spot-like dots that might be
weathering products, residing between layers, were also noticed. Those spot-like dots lighter
287
-------
'"If"-'- "*fir a»^*!ffl^' ~'.Jr>/-;i»grt'M *• '•'
*Nfe-i ^^'^*^^%;;-'
r
•a.v --^^k^f*^. •'"
f '•* . -ii#?;.-ElS»,'i - • r .'.aafesjf „ >
MUQS'T
MU Ot-'O
288
-------
W5
1
^
^
1
•2
*•— *
H
tf
^
o
x— s.
Q *c3 "%-<
^o "5 t5
PH H B
o"
'S<
g
o
C^ l~~l /fis
oo C^ S
P-l CS ^S
1^
j> 3.
^ 8
Q "3
^O ^"*
PH v_x
g
3 '§
^ s
I-IH "3
of
8
"SnxT
§ S II"
01
60
ra 'S §-
1
i 'g ^^
•S ^ ^H
^ & r3 '
< O v_/
§
1?
INl -S
Substrate
o
»— <
X
en
cs
^
m
O
X
^l-
CS
o
X
T^*
CS
m
m
cs
cs
in
oo
0
o
en
(^
O
0
o
CS
CS
1
o
Indian
muscovite
o
»— <
X
CS
en
en
C1
O
X
*o
1— H
^
+
0
X
*^)
CS
J^_
CS
CS
t^.
o
o
VO
en
0
O
O
O
O
t
O
ll
pg j£
•7
o
*~~<
X
in
en
*?
O
X
*n
cs
en
^
O
X
en
CS
O
OS
cs
OS
r-
CD
OS
vd
cs
en
O
O
8
cs
1
o
Phlogopite
o
N
CS
oo'
;
T
O
X
^j-
c«4
CS ;
<•>
O '
X
OO
'-'
en
00
cs
cs"
OO '
S
o
.
Os"
,
, ,
o
0
o
1
. o
cs
o
1
o :
Transvaal
vermiculite
?
o
X
Os
OS
CS
5
o
X
r-
>n
00
o^
^
^
en
cs
•*
i~~l
Os
CS
o
X
oo
en
cs
00
en
cs
1
0
Kaolinite
CM
o
*— •
X
en
^
eji
o
X
f-
OS
"
n
0
X
i— — i
*si
i
1
OS
VO
cs
o"
CS
en
cs
t—
o
o
o
•H ^s
J B
r*>
•§"
C
•3
n,
CO
1
PH
289
-------
than the background were found to be sharp peaks after the three-dimensional image was
examined. More detailed atomic level images were obtained when the scan parameters such as
scan size (6.0 x 6.0 nm ), z-range (1.8 ), scan rate (61.0 Hz), and set point (0.1) were optimized.
It is clear that some kind of periodicity can be found in the unit cell size, a = 4.97 A, b = 11.75
A, which was measured by using the special measurement capability of the software.
Phlogopite. The phlogopite image at the micrometer-level (not shown) shows that its
surface has wrinkles; if one compares the height of those wrinkles with the size of the view, (10 x
10 nm), then phlogopite still can be categorized as "flat." A more detailed image is showed in
Figure 1-3 where the hexagonal rings of the basal oxygens and the unit cell structure can be
identified. The dimension of the unit cell measured is a = 6.37 A, b = 9.38 A. These data are
consistent with the data that Smith and Yoder (1956) developed using x-ray diffraction
spectroscopy.
Transvaal Vermiculite. The nano-scale image (Figure 1-4) shows the repetition of the unit
cell, which has a dimension of a = 6.64 A, b =11.38 A. It is a trioctahedral analogue that has Fe 2+,
Fe3>* and Mg2* in the octahedral sheet and is a 2:1 expansible mineral with hydrated interlayer
cations balancing the net layer charge due to isomorphous substitution. Atomic (muscovite and
vermiculite) or nearly atomic-scale resolution were obtained for the four phyllosilicate. What was
imaged in all cases was the two-dimensional tetrahedral sheet (octahedral sheet is below the
tetrahedral sheet and cannot be imaged). The tetrahedral sheet consists of three basal oxygens of
each tetrahedron being shared by other tetrahedra, resulting in linked hexagonal or "ditrigonal
rings of tetrahedra. The unshared or apical oxygen of each tetrahedron lies within the 2:1 layer of
which the tetrahedron is a part. From the nano-scale images of these four micaceous minerals, it
is obvious that they have a similar unit cell structure.
The definition of image analysis terminology used ( autocovariance, bearing ratio, fractal
dimension, power spectral density ( PSD) functions and roughness) is found in an article by Gan
et al. (1996). All four mineral surfaces are nearly flat as evidenced by the low roughness (RMS)
values ( 0.028 to 0.085 nm) and very low values for autocovariance. Nearly identical images of
both muscovite and vermiculite were produced after, as well as before, the autocovariance
measurements were made indicating an extremely high replication of the hexagonal configuration
of oxygen atoms along the x-y plane. This was not true for either Bancroft biotite or phlogopite.
The presence of Fe in the octahedral sheet causes an increase in the b dimension which may
reduce the symmetry of the oxygen triads in the tetrahedral sheet. Why this would not affect
Transvaal vermiculite is unknown.
Muscovite and Transvaal vermiculite have a greater fractal dimension than the biotite and
the phlogopite, which would suggest greater disorder or structural complexity. However, the
autocovariance data suggest the exact opposite. The relative surface flatness and high degree of
structural regularity and integrity of mica are the major reason for its being sought and used as a
substrate for mounting samples for AFM. The PSD-frequency values are large in magnitude
while the PSD- isotropic and total power values are very small in magnitude.
290
-------
Kaolinite. The image of kaolinite (1:1 nonexpansible layer-lattice silicate) is of individual
particles not of a cleaved flake. Kaolinite particles were dusted on the surface of a resin, the
resin heated, the kaolinite particles immobilized in the resin, and the sample imaged ( Figure 1-5).
Due to its different nature, it is improper to compare the four 2:1 minerals with kaolinite.
Kaolinite particles were about 200 nm in diameter and were either spherical or oblate in shape.
Attempts to characterize individual kaolinite particles was not successful.
Graphite
The nano-scale image (Figure 1-6) shows hexagonal surface symmetry. In graphite
carbon atoms replace the oxygen atoms found in the micas as members of the hexagonal ring.
The graphite surface is nearly as flat as the micas, as shown by the low RMS value (Table 1). The
structural periodicity of graphite is not quite as regular as the micas, as evidenced by a slightly
larger autocovariance value. Fractal dimension can not be calculated when STM is used to image
a surface.. One would expect the graphite surface to be hydrophobic.
Reaction Products on Phyllosilicate Surfaces
From Figure 2-1 note the lath-shaped particles on the surface the of Transvaal
vermiculite. We believe that they are naturally formed weathering products resulting from either
the reductive dissolution carried out in the interstices of the: vermiculite packet or acid dissolution
of the vermiculite strucuture followed by diagenetic formation of an oxide having a lath-like
mineralogical habit. It may be an silicon or an iron oxide. Presence of separate particles on the
surface resulted in an increase of several orders of magnitude in autocovariance values and a three
or more order of magnitude decrease in PSD-frequency value (Table 2). Similar changes are
noted in the RMS and PSD-2D isotropic values.
Figure 2-2 and Figure 2-4 show the the general morphological images for Fe(III) and
Cr(III) reaction products, respectively. The reactions were!conducted several pH units above
their hydrolysis constants; the Fe (III) hydrolysis constant is 1.8 and that for Cr(III) is 5.1.
Aggregation and precipitation has occurred for Fe(III) as evidenced, by the high roughness values
and autocovariance (Table 2). The solid is not very porous as evidenced by the low fractal
dimension. For Cr(III) a lattice solid is formed but it appears to be a two-dimensional solid.
Fe(II) forms a red colored complex with 1,10-phenanthroline; three molecules of the
complexing agent react with one Fe(II) ion to form a metal-ligand complex with octahedral
coordination. Structurally, 1,10-phenanthroline is a three-ring system, i.e., it has three benzene
rings with a nitrogen atom in the one and ten position in the fused ring system. An examination of
291
-------
2.00
1.00
2.00
-1.00
-3.00
2.00
1.00
3.00
i-2.00
-1,00
1.00
2.00
3.00
Figure 2. AFM images of planar surfaces of freshly cleaved phyllosilicate with
present various types of reaction products. (1) vermiculite with a natural
weathering product on the planar surface (scan size 0-2.411 nm, z-dimension
0-80.0 nm); (2) Fe(III) oxide- muscovite surface( scan size 0-3.885 nm,
z-dimension 0-400 nm); (3) Fe(II) - 1,10-phenanthroline - muscovite surface
complex (scan size 0-2.014 nm, z- dimension 0-1.2 nm); and (4) Cr(III) -
saturated muscovite reacted at pH 10.75 (scan size 0-3.274 nm,
z-dimension 0-1.0 nm).
292
-------
en
1
en
1
J2
0,
a
o
.1
en
ts
3
13
.g
P*
_O
Kj
£
'o
o
•J3
en
1
2
^^-j
O
"3
5b
^5
*o
•p.
1
1
H
_ ^"/-x
C^ ^ "^ *^
oo 'S 6 §
a< H cuo
o"
'a.
p
Q~f
PL, CN O
cr 'g
Q 73
c
-a '§
0 I
2 .§
#
|«f^
O ^§ a
OH S -S
S*
§ '1 ^
P3 2 ^
8
§
i "C ^_,
III
c
o
'to
1
Q •?
N v_/
§
Substrate/ reacti
product
_
O
ON
*/^
C4
+
0
•><
3
o
•x
(N
oo'
.
oo
cs
o\
o
vS
2
o
CO
0)
o
o
00
1
o
oo
a
Vermiculite,
natural weatheri
product
0
•x.
oo
»— •
VO
¥
0
•x.
£
o
>i
0\
ON
oi
oo
i-;
in
Q
o\
8
•
o
Fe3+-saturated
muscovite,
pH8.08
b
T— 1
X
in
^^
T
0
X
p
¥
o
X
8
oi
CS
o
1
1
S
'
^.
o
CD
O
CS
rt
1
o
^.S
Fe2+ -muscovite
1,10-phenathrol:
b
X
r-
CN
C*1
5
O
X
r--
cs
5
o
X
cs
CN
O
cs
CD
CD
CN
CS
O
CD
O
0
p
.— J
1
0
Cr3*- saturated
muscovite,
pH 10.75
If
§
^*
Z2
1
D-,
|
cu
293
-------
Figure 2-3 shows a three-ring system in the lower right corner. Because of the octahedral
configuration it seems logical that the complex could not lie flat on the muscovite surface. The
long thread seen in the center of the image may be the complex, oriented mainly perpendicular to
the muscovite surface. The autocovariance, fractal dimension and the PSD parameters suggest
that the complex covers most of the surface and that there is a reasonably good periodic
orientation of the complex in its two or more possible conformations.
Peat and Humic Acids on Various Substrates
These images can be seen in Figure 3 and the morphological characteristics are shown in
Table 3. Peat imaged on graphite by STM has a low autovariance value, a reasonably high
bearing ratio, low roughness, a moderately high PSD-frequency value and a low PSD-2D
isotropic value. From the image, we also can see the presence of linear chains presumably
aliphatic in character, and in the lower right-hand corner the presence of a benzene ring. Also the
autocovariance image (not shown) suggests a periodic occurrence of linear chains. These linear
chains show periodicity in the x-y plane as evidenced by the very low autocovariance value and
bearing ratio, moderately high PSD-frequency value, and low PSD-2D isotropic value. The axis of
the chain is oriented about 60 degrees, which is the same orientation as the hexagonal oxygen
rings of the tetrahedral sheet of the muscovite substrate. It is not known whether this is due to
the influence of the muscovite surface on the orientation of the chains, or the influence of the
trivalent lanthanum cation, or both.
The nature of the saturating cation influences the morphology of humic acid on muscovite
surfaces. The Na* -saturated muscovite caused the humic acid to be less organized on the surface
than it did for the K*-saturated muscovite. Sodium-saturated montmorillonite is a very dispersed
but stable system. This is shown by a higher autocovariance value and a much lower PDS-
frequency value.
The morphology of humic acid deposited on a vermiculite surface (Mg2+ is the native
saturating cation) is different than that present on muscovite. The z-dimension is greater, the
autocovariance value is over two orders of magnitude greater, the roughness is more than two
orders of magnitude greater, the PSD-frequency is three orders of magnitude greater, and the
PSD-2D isotropic value is seven orders of magnitude greater. The fractal dimension value
suggests that the humic acid does not completely cover the vermiculite surface (Table 3).
Virtual Reality Visualization
Figures 4 shows virtual reality (VR) visualization of various environmental surfaces. The
VR image copies contain the identical information as their SPM originals. Moreover, they
provide a simpler and easier understanding of interrelationships between scanned nano-objects on
the environmental surface and the overall morphology. This results from the VR presentation
294
-------
«u OO'E
295
-------
•a
CO 1
03 1
\^ 1
M
3
•Z3
c
/r?
T3
C
03
0
-r*
I
«
1
i.
&
c
o
•o
O)
00
03
M
c?
w
VM!/
"O
'3
03
3
JS
1
1
(4-1
O
CO
_O
"^^
CO
•g
1
[orphological cha
surfaces.
2
co
_qj
H
,
g
n,^
QS'e
w o S
Pi E-" •&
o"
f
/^\ ^ '"~N
&O ^i §
P-( 01 S
/""N
"t
o* §.
§ e/i
v> P >^
WJ l_ Q
py. FT . \^*
g
*-"' ^
cd c
"cj 52
ca g
tS "§
of
8 '•§ ^
pa 2 tx
^
1
6 3 *£
Ill
Z-Dimension,
(nm)
a
H
OT
•§
C/3
o n ~ ¥
o o o o o
X X X X X
os cs •<* os r~-
oo vo o p oo
pi CN os m' •— '
v>
T-, "o , o
ri i, TH «p -
X X «— ' X O
k> . .*_)
PI •* *• <^
i-J -3- cs r- x
,-J pi •* r-^ CN
¥ ? ?
o o o
XXX
,— oo PI m in
vo cs oo P> cs
VO r— J CS •— ' "— '
t- in -=*• oo m
cs cs ^- PI p
i i .1 i i
o o o o . o
«Jj TJ » 1 W
§s -L| hi< MH
E S
•^ "^ ^
§ ^^&
< -8 -8'IE,
>.^'S^
•° 9 S
1 '•§,.' § § 3
e & J .III
§ § a § s
O *t3 T3 * *^ *
g C O T3 T3 "3
c&ft'll'S
fe •§ .3 .2 o .2
g ^ "S 6 E S
P-I P-I (S K S ffi
— * • • *
w u S 25 ^ ^
PH CU PH 3H PH K
296
-------
» *«"
J *.-
H
£ S
<- ™
d§
f I
&
rn g
X O
m °
.2 B
o ^
13 fa
8 <«
§
I
M
O
OH
-8
J o
•^r ^
^1
C/) ^
^* w
^ a
•o 2
n> 'S
SP ^
g o
.9 "~i
^*—X T—H
8/—-
ro
,ea ^--
S^
11
3 u-i
-^ ^
fe x
II
S3
I -
5 . <^-l
It1
§ |
"s -9
|s
*^"^ .OT
^3-
(L)
297
-------
being fit to subconscious mechanisms of human perception. Coupling SPM and VR may be
especially useful in comparing surface morphologies of different chemical systems. Thus, Figures
4_1 to 4-3 provide a striking illustration of similarities and differences between nanometer-scale
surface arrangements of regular and of irregular structures. The micrometer-scale images of two
chemically related natural samples in Figure 4-4 show how easily this techniques can provide the
understanding of morphological changes of organic matter upon environmental transformations
(Shevchenko and Bailey 1997).
CONCLUSIONS
Scanning probe microscopy (STM/AFM) shows great promise in defining the
morphology (micrometer level and nanoscale levels of atomic resolution) of environmental
surfaces, i.e., composite environmental surfaces. STM images of humic substances on graphite
have actually proved that the tunneling current can flow on the top of the fulvic acid structure,
i.e., even for nonconductor organic molecules, the small tunneling current still can probe the
surface morphology of humic substances. Sample preparation plays a critical role in the overall
observation because the pH, drying method, concentration, substrate, and ionic strength greatly
affect the morphology of the environmental surface. Therefore, these factors also affect the
interactions or reactivity changes between pollutants and the environmental surface.
Understanding the morphology, structure, electronic and surface properties, and chemical
reactivity of the environmental surface, as well its provenance and the transport processes
responsible for its entry into surface and groundwater, enhances our capability of predicting
pollutant fate and environmental exposure, and of developing of risk assessment models.
Surface morphology analysis descriptors (including autocorrelation, RMS, fractal
dimension, PDS, section analysis, and bearing ratio) when used singularly and in combination,
were found to be highly useful to characterize: (a) the roughness of the surface, (b) the periodicity
of molecular arrangement in the x,y, z dimensions, and (c) the extensiveness of coverage of the
surface in the x-y plane. These descriptors demonstrated comparable ways of examining surfaces
of imaged compounds. SPM sample preparations allowed us to image compounds in a normal
atmosphere and at room temperature without altering the characteristics of samples. SPM is a
valuable approach to obtain structural information at the micrometer level of resolution and at the
nanoscale or atomic level of resolution, both on well-ordered surfaces and those of low molecular
periodicity, high or low amorphous or crystalline character, variable surface coverage, and at
room conditions. Similarly, with the use of the fluid cell, precipitation and dissolution studies can
be carried out in real time and the kinetics and change in morphology followed as the chemical
reaction on the surface progresses. The advent of the tapping mode head makes it possible to
image organic substrates with a lessen probability of fragile surfaces being damaged. We believe
that VR software is a useful supplement to the standard SPM off-line analysis and that it will
become a routine way of viewing and presenting images.
298
-------
ACKNOWLEDGMENT
This research was conducted while Humin Gan and Sergey M. Shevchenko were research
associates of the National Research Council of the National Academy of Science at the U.S.
Environmental Protection Agency's Ecosystems Research Division, National Exposure Research
Laboratory in Athens, Georgia. This research was supported in part through U. S.
Environmental Protection Agency, Contract No. 68-C1-0024 to DynCorp-TAI Inc. Mention of
trade names or commercial products does not constitute endorsement or recommendation by the
U. S. Environmental Protection Agency. The authors gratefully acknowledge the technical
assistance of Leonid G. Akim and Z. Z. Zhang and the editorial review of Rob Ryans.
REFERENCES
Albrecht, T.R., and C.F. Quate. 1987. Atomic resolution imaging of nonconductors by
atomic force microscopy. Journal of Applied Physics 62:2599-2602.
Bailey, G.W., S.M. Shevchenko, Y.S. Yu, and H. Kamerinans. 1997. Combining
scanning tunneling microscopy and computer simulation of humic substances:Citric
acid, a model. Soil Science Society of America Journal 61:92-101.
Beveridge, T.J., and S.V. Koval. 1981. Binding of metal to cell envelopes ofEscherichia coli
K-12. Applied Environmental Microbiology 42:325-335.
Beveridge, T.J., C.W. Forsberg, and RJ. Doyle. 1982. Major sites of metal binding in
Bacillus licheniforms walls. Journal of Bacteriology 150:1438-1448.
Beveridge, T.J., J.D. Meloche, W.S. Fyfe, and R.G.E. Murray. 1983. Diagenesis of metals
chemically complexed to bacteria: Laborataory formation of metal phosphate, sulfides and
organic condensates in artificial sediments, Applied Environmental Microbiology
45:1094-1108.
Beveridge, T.J,. and D.J. Doyle (Eds.). 1989. Metal Ions and Bacteria. Wiley & Sons New
York.
Binnig, G., C.F. Quate, and C. Gerber. 1986. Atomic force microscope. Physical Review
Letters 56:930-933.
Binnig, G., and H. Rohrer. 1982. Scanning tunneling microscopy. Helvetica Physica Acta
55:726-735.
299
-------
Brindley, G.W., and D.M.C. MacEwan. 1953. Structural mineralogy of clays, ceramics - a
symposium, ppl5-59, British Ceramic Society.
Burkert, U., and N.L. Allinger. 1982. Molecular mechanics, ACS Monograph 177,
American Chemcial Society, Washington, D.C.
Dixon, J.B., and S.B. Weed (Eds.). 1989. Minerals in the Soil Environment. 2nd ed. Soil
Science Society America, Madison, Wisconsin.
Drake, B., C.B. Prater, A.L. Weisenhorn, S.A.C. Gould, T.R. Albrecht, C.F. Quate, D.S.
Cannell, H.G. Hansma, and P.K. Hansma. 1989. Imaging crystals, polymers, and
" processes in water with the atomic force microscope. Science 243:1586-1589.
Can, H., G.W. Bailey, and Y. S. Yu. 1996. Morphology of lead(II) and chromium (III)
reaction products on phyllosilicate surfaces as determined by atomic force microscopy.
Clays and Clay Minerals 44:734-743.
Gjessing, E.T. 1975. Physical and Chemical Characteristics of Aquatic Humus. Ann Arbor
Science Publishers, Inc., Ann Arbor, Michigan.
Golovchenko, J.A. 1986. Tunneling microscopy: A new look at the atomic world. Science
232:48-53.
Greenland, D.J., and M.H.B. Hayes (Eds). 1978. The Chemistry of Soil Constituents. John
Wiley and Sons, New York.
Hansma, P.K,. V.B.Elings, O. Marti, and C.E. Bracker. 1988. Scanning tunneling
microscopy and atomic force microscopy: Application to biology and technology.
Science 242:209-216.
Hansma, PK, and J. Tersoff. 1987. Scanning tunneling microscopy. Journal of Applied Physics
61:R1-R23.
Hartman, H., G. Sposito, G. Yang, S. Manne, S.A.C. Gould, and P.K. Hansma. 1990.
Molecular-scale imaging of clay mineral surfaces with the atomic force microscope. Clays
and Clay Minerals 38:337-342.
Hayes, M.H.B., P. MacCarthy, R.L. Malcolm, and R.S. Swift. 1989. Structure of humic
substances: the emergence of "forms". In: Humic Substances II: In Search of Structure.
Hayes, M.H.B., P. MacCarthy, R.L. Malcolm, and R.S. Swift (Eds.). John Wiley and
Sons, New York. 690-733.
300
-------
Hochella, M.F. Jr., C.M Eggleston, V.B. Elings, G. A. Parks, G. Brown, C.M. We, and
K. Kjoller. 1989. Mineralogy in two dimensions: Scanning tunneling microscopy of
semiconducting minerals with implications on geochemieal reactivity. American
Mineralogist 1233-1246.
Hoyle, B.D., and T. J. Beveridge. 1983. Binding of metallic ions to the outer membrane of
Escherichia coli. Applied Environmental Microbiology 46:749-752.
Jackson, W.W., and J. West. 1930. The structure of muscovite, Part I. Zeitschrift
Kristallographica 76:211-227.
Jackson, W.W., and J. West. 1933. The structure of muscovite, Part II. Zeitschrift
Kristallographica 85:160-164.
Loughnan, F.C. 1969. Chemical Weathering of the Silicate Minerals. American Elsevier
Publishing Company, Inc. New York.
Mauguin, C. 1928. Study of micas by X-rays, C.R. Academy of Science, Paris, 186:879-
881.
Pauling, L. 1930. Structure of micas and related minerals. Proceedings of the National
Academy of Science, Washington, 16:123-129.
Radoslovich, E.W. 1959. Structural control of polymorphism in micas. Nature 183:253.
Radoslovich, E.W. 1960. The structure of muscovite, KAl2(Si3Al)O10(OH)2 Acta
Crystalographica 13:919-932. '
Radoslovich, E.W. 1961. Surface symmetry and cell dimensions of layer lattice silicates.
Nature 191:67-68.
Radoslovich, E.W., and K. Norrish. 1962. The cell dimensions and symmetry of layer lattice
silicates I. Some structural considerations. American Mineralogist 47:599-616.
Schnitzer, M., and S.U. Khan, 1972. Humic Substances in the Environment. Marcel
Dekker, Inc., New York.
Shevchenko, S.M., and G.W. Bailey. 1996. Life after death: Lignin-humic relationships
reexamined. Critical Reviews in Environmental Science and Technology 26: 95-153.
Shevchenko, S.M., G.W. Bailey, and Y. S. Yu, 1997. Nanoscapes: Coupled SPM/VR
analysis of environmental surfaces, American Laboratory News, March, pi 8.
301
-------
Smith, J. V., and H .S. Yoder. 1956. Studies of mica polymorphs. Mineralogical Magazine
31:209-235.
Stevenson, F. J. 1994. Humus Chemistry. Genesis, Composition, Reaction. 2nd Ed. Wiley,
New York.
SufFet, I. H., and P. MacCarthy. 1989. Aquatic Humic Substances: Influence on Fate and
Treatment of Pollutants. American Chemical Society, Washington B.C.
Tersoffj J,. and D. R. Hamann. 1983. Theory and application for the scanning tunneling
microscope. Physical Review Letters 50:1998-2001
Thurman, E.M., and R. L. Malcolm. 1983 . Structural studies of humic substances: New
approaches and methods. In: Aquatic and Terrestrial Humic Materials. Christman, R.F.,
and E.T. Gjessing (Eds.). Ann Arbor Science Publishers, Ann Arbor, Michigan.
% *
van Olphen, H. 1963. An Introduction to Clay Colloid Chemistry. Interscience Publishers,
New York.
Wipff, U. (Ed.). 1994. Computational Approaches in Supramolecular chemistry. Kluwer
Academic Publishers, London.
Yoder, H.S., and H. P. Eugster. 1955. Synthetic and natural muscovite. Geochimica et
Cosmochica Acta 8:225-280.
302
-------
HIGHLIGHTS OF WATER POLICY IN FINLAND
Hannele Nyroos1
ABSTRACT
During recent decades, ambitious pollution reduction programmes have been passed and
comprehensive regulations applied to different environmental sectors to control environmental
pollution in Finland. Pollution loading from point sources has decreased significantly during the
1970s and 1980s. However, though water pollution has decreased in many areas of Finland, the
condition of the Finnish water bodies does not meet all requirements of usability and of
conservation. Nonetheless, the general condition of the Finnish watercourses is quite good. The
general trend in Finland is that polluted water areas have decreased, but the areas which are in a
"natural state" have similarly decreased. The main problems are eutrophication, acidification,
deterioration of groundwaters, unsustainable land use patterns, regulation of water flows and lake
levels, and increase of impacts caused by different risk factors. There are two main reasons why
.it has been possible to enforce quite coherent water policy in Finland. The Water Act issued in
1962 has given the juridical basis to regulate activities which may cause adverse effects in water
bodies, and an integrated water management administration was established in 1970. Water
management and water protection planning has played a crucial role in developing the Finnish
water policy by defining national goals, strategies and measures for implementation. Third
national water protection plan will be compiled this year. In Finland the permit conditions for
wastewater discharges are defined on a case-by-case-basis as opposed to the composition of
general norms. The Finnish water legislation was geared to meet the European Community
statutes when the European Economic Area agreement entered into force in 1994. The Baltic
Sea is badly eutrophied as a result of heavy nutrient load. Repeatedly, there is a serious oxygen
depletion in some of the deeper parts of the sea. Consequently, the Baltic Sea Joint
Comprehensive Environmental Action Programme was drawn up. The programme identified
132 "hot spots" which are significant point sources of pollution, 47 of which were assigned top
priority. The implementation of the entire programme is estimated to cost at least 18 billion
ECU (24 billion U.S. dollars) over a 20-year period.
'Ministry of the Environment, Helsinki, Finland.
303
-------
INTRODUCTION
Issues related to waterbodies are of special importance in the Finnish society. Finns have
always lived in near contact with nature and it is an essential part of the Finnish way of life.
There are approximately 188,000 lakes in Finland, which cover about 10 percent of the country's
surface area. The volume of water hi lakes is not large, however, only 230 km3, as the lakes are
quite shallow, the mean depth being only 7 meters. This is one reason why the Finnish water
bodies are quite sensitive to pollution. The considerable humus loading coming from peatlands
and bogs consumes oxygen storage of water bodies. In the winter ice cover prevents the
filtration of oxygen from air to water. Because of these natural characteristics we need a very
high standard of water protection.
Unlike many other countries, in Finland water sufficiency is nota problem. The average
discharge to the sea is about 3,100 mVs. For every Finn there are about 59,000 liters of water
available per day. Today half of the drinking water comes from groundwaters. The groundwater
resources are about 6-8 million km3 a day, and at present only about 0.6 million km3 a day is
used. Naturally, however, there are some local problems hi groundwater availability. We aim at
increasing the share of groundwaters as drinking water to about 70-75 percent by the year 2010.
This is because the groundwaters are sheltered better than surface waters against pollution.
Compared to most other countries, quite a small area, just a couple of per cent of agriculture
land, is irrigated.
CURRENT WATER MANAGEMENT PROBLEMS
Pollution loading from point sources decreased significantly hi the 1970s and 1980s.
Even though water pollution has decreased in many areas, the condition of Finnish water bodies
does not meet the requirements of usability and of conservation. The general condition of
Finnish watercourses is quite good. According to the classification developed by the National
Board of Waters and the Environment, water quality is excellent or good hi about 80 percent of
the Finnish lakes. The share of areas which are hi bad or very bad condition is only 3.5 percent.
The situation is, however, worse in rivers. In one third of the length of the rivers the water
quality is bad. The general trend hi Finland is that polluted water areas have decreased, but the
areas which are in a "natural state" have similarly decreased. The principal causes of this
decrease of water hi natural state are eutrophication, acidification, improper land use, and
artificial regulation of water flows and lake levels.
Eutrophication
Eutrophication results from excessive nutrient pollution, the main sources of this being
agriculture and forestry. Also pulp and paper industry, municipalities, rural settlement, and fish
cultivation are significant polluters. In the inland waters, eutrophication is caused mainly by
phosphorus. In coastal waters nitrogen is mostly a minimum nutrient. Eutrophication:might also
be caused by nutrient release from sediments. After wastewater pollution has ceased, the old sins
in the form of nutrient rich sediments cause excessive phytoplankton growth in many areas.
304
-------
At present, quite effective instruments are available to regulate pollution from point sources.
In order to decrease pollution caused by agriculture, environmental programmes for each
individual farm will be drawn up in the near future. The funding from the European Union for
environmental projects of agriculture will play an important role. Nitrogen removal is necessary
in many treatment plants in municipalities.
Acidification :
The critical loads of acid depositions are exceeded in almost every part of Finland. Most
of this, about two thirds, comes from other countries. Finland has decreased the sulphur
emissions by almost 80 percent. Nationally the main problem is nitrogen emissions from traffic,
energy production, and agriculture.
Deterioration of Groundwaters
The quality of Finnish groundwaters is generally good. They can often be used as
drinking water without some specific treatment. Slow deterioration can be seen in many areas in
the form of increasing nitrate concentrations and increase in acidification. Risks to groundwaters
are caused by intensive agriculture, landfill sites, industrial and traffic accidents, gravel
extraction, and transboundary depositions.
Improper Land Use Patterns ;
Deterioration of the condition of water bodies is nowadays often caused by actions in the
catchment area. In recent decades agriculture has become more intensive and it is often based on
monoculture. The use of pesticides and fertilizers has increased even if the use of pesticides is
significantly lower in Finland compared with the average use in OECD countries. As a result of
modern cultivation methods, many animal species living in contact with agricultural areas have
become endangered. Drainage of farm lands and forests, as well as flood prevention actions, have
destroyed the habitats of many species.
Regulation of Water Flows and Lake Levels
In Finland about one third of the lake area is regulated for producing hydropower, flood
prevention purposes, or for water supply. The passage of migratory fishes is blocked in most
major rivers because of dams. We have two major artificial lakes which.lie in northern Finland.
Operation and maintenance of regulations have been developed recently in many areas designed
to decrease the harmful effects to animal species of water uses.
EVOLUTION OF FINNISH WATER POLICY
There are two main reasons why it has been possible to enforce quite coherent water
policy in Finland. First, a new (revised) Water Act was issued in 1962, and it has provided the
juridical basis to regulate activities which may cause adverse effects in water bodies. Second,
305
-------
an integrated water management administration was established in 1970. The National Board of
Waters was established and it had broad authorities to enforce water policy at the national level
by water management planning, supervision, and research. The district offices have been
responsible for the regional implementation of national policies. It has also had an important
role in monitoring water quality as well as being the supervising authority. Uniting the water
issues under the same authority has made it possible to integrate water issues in an effective way.
This has also caused some conflicts, as water authorities promoted water development projects
which had harmful effects from the point of view of water protection. Another deficit was that
water issues were separated from other environmental issues. Later the organization was
changed so that activities related to water development projects decreased and the scope of other
environmental issues was broadened. Nowadays the environmental administration deals also.
with waste management, nature protection, environmental impact assessment, and dissemination
of environmental data.
The Finnish Ministry of Environment was established in 1983. This meant that the role
of environmental issues increased and the scope of environmental issues was also broadened. In
1987 a major revision was enforced hi the Water Act More emphasis was given to nature
protection and recreational use of water bodies. Some administrational changes related to permit
procedures were passed. In 1994 the EEA (European Economic Area) agreement was signed and
in 1995 Finland became a member of the European Union. Most recently, Finland's
environmental administration was reorganized in March 1995. Thirteen regional environmental
centers incorporating the former water and environment districts and the environmental units of
Provincial Offices were established as a part of the regional administration. The former National
Board of Waters and the Environment, now known as the Finnish Environment Agency, has been
turned into a center for environmental research and development. The regional Environmental
Centers and the Finnish Environment Agency, all subordinate to the Ministry of the
Environment, are responsible for the use and management of water resources falling within the
auspices of the Ministry of Agriculture and Forestry. The reorganized environmental
administration will be better equipped to coordinate the handling of environmental matters.
Today, the Finnish environmental administration employs a total of 2,200 people, of whom
310 work for the Ministry of the Environment, 430 for the Finnish Environment Agency, and
1,460 for Regional Environment Centers.
CURRENT WATER MANAGEMENT POLICY
*
Water management and water protection planning enforced by water and environmental
administration has played a crucial role in developing the Finnish water policy by defining
national goals, strategies and measures for implementation. The first national environmental
policy plan was compiled in March 1995. It represents the view of the Ministry of the
Environment on how society should fulfill the requirements of sustainable development. It also
discusses how to integrate water management in the different sectors of the society and how to
integrate water issues with other environmental sectors. The third National Water Protection
Programme is scheduled to be completed hi the next few months. The programme has not any
306
-------
mandatory status; instead it provides the framework for developing national wide water pollution
measures. The earlier programmes have had a great influence on promoting water protection.
Concrete targets have been set for the most prominent pollution sources. Discharges from the
most significant source of pollution, the pulp and paper industry, as well as that from
municipalities, have decreased significantly. Regional water management and water protection
plans have been drafted by the environmental administration for areas with water protection
problems or for areas where there are conflicts between different water uses and water protection
interests.
Defining Water Quality Requirements for Polluters
Juridical bases are defined in the Water Act. It includes three bans which may not be
violated without a permit issued by the Water Courts. These bans concern impending passage,
altering flows and water levels as well as polluting water bodies. One special feature of the
Finnish practice is that the permits are issued by special Courts and not by administrative
authorities. Applications for permits, addressed to the Water Court, are handled either by public
announcement or by applying the inspection procedure. In the inspection procedure the benefits
and adverse effects of project are assessed. Conditions in different forms of obligations are set in
the permits, e.g. deadlines, monitoring of impacts in the aquatic environment, and levels of
effluent treatment. Another special feature of Finnish water management is the fact that water
bodies are privately owned. Therefore one task of the inspection procedure is to define the
compensations which shall be paid to owners of water bodies as a result of deteriorated water
areas.
In water management there are two different kinds of approaches of how to define
permit conditions. These are uniform discharge controls, and water quality targets or objectives
defined for waters receiving discharges leading to local discharge standards. In Finland the
permit conditions are defined on a case-by-case-basis according to the last approach as opposite
to the composition of general norms, and the following aspects are taken into consideration:
(1) Characteristics, protection and use of the watercourses in question; (2) Extent of
environmental impacts; (3) Existing treatment installations and water protection standards; and
(4) Available treatment methods.
The degree of wastewater treatment required has gradually increased as a result of the
development of environmental technology. Permit conditions are usually fairly uniform across
the country, but local conditions are taken into consideration.
European Community Directives
The European Community has adopted a series of water quality directives specifying or
requiring numerical standards for a broad range of water quality variables. The directives specify
binding policy objectives, but they leave the means for achieving them to each member state.
Legislation on discharge levels takes the form of limiting values of each discharged substance
and form of industry. These directives are: (1) User based (e.g. bathing water); (2) EQO based
307
-------
List II; (3) EQO or discharge based List I; and (4) BATNEEC type (Urban wastewater
treatment). The Finnish water legislation was geared to meet the European Community statutes
when the European Economic Area agreement entered into force in 1994. Water protection will
continue to be implemented by means of the Water Court permit procedure set in the Water Act,
whereby permits are issued on an individual basis after case-by-case considerations. In most
cases the changes in the Water Act have mainly formal importance, because the Directive
requirements are normally included in the permit practices of Water Courts. The standards set in
Directives serve only as the absolute minimum standards. When adapting Finnish legislation to
that of the European Community it is our intention to preserve the high standard of Finnish water
protection.
THE BALTIC SEA
At the end of the 1960s there were already clear signs of an imbalance in the Baltic Sea
ecosystem. One serious indication was that seal populations declined, the grey seal being then
near extinction. The states around the Baltic Sea became concerned and started progressive
cooperation. As the first remarkable result of this cooperation, the Convention on the Protection
of the Marine Environment of the Baltic Sea Area was signed in 1974 and came into force in
1980. A special international commission, the Helsinki Commission (RELCOM), was
established to issue recommendations for improving the state of the Baltic Sea. This
Commission has worked very actively by monitoring the state of the sea and issuing
recommendations on reduction of land-based pollution and pollution from ships. It also forbade
use of the most toxic substances (e.g., DDT and PCB) and the dumping of wastes, and created
cooperation in combating dumping of oil and other harmful substances.
Although tangible progress in environmental conditions can be seen as a result of work
under the Convention, the present state of the Baltic Sea remains a cause of serious concern. The
Baltic Sea is badly eutrophied as a result of a heavy nutrient load. Repeatedly, there is a serious
oxygen depletion in some of the deeper parts of the sea. In 1992 the "dead bottoms" area was
approximately one third of the seabed; when it was at its worst at the end of the 1980s, its share
was two thirds. The effects of harmful organic compounds is a third prominent problem.
Apparently, these substances have contributed to reproduction failures of seals in the Baltic Sea.
Their populations are still so small that all seal species in the Baltic Sea are endangered despite a
slow recovery of the population.
Because of the alarming polluted state of the Baltic Sea, a diplomatic conference was
held at Ronneby, hi Sweden in 1992. The conference launched "a concrete effort to assure the
ecological restoration of the Baltic Sea, ensuring the possibility of self-restoration of the marine
environment and preservation of its ecological balance." Consequently, the Baltic Sea Joint
Comprehensive Environmental Action programme was drawn up. The Programme identified
132 "hot spots", 47 of which were assigned top priority. Most of these spots are situated in the
former Soviet Union and in other formerly centrally-planned economies. For implementation of
the action plan in these countries, the establishment of a basic infrastructure is a prerequisite.
This means organizational and human capacity, legislation, and education.
308 .
-------
In the first years, emphasis has been placed on policy reform, limited public investment
in the highest priority projects, and promotion of private investments. The long-term Baltic Sea
programme for specific action includes the following components: (1) Policy, legal and
regulatory measures; (2) Institutional strengthening; (3) Investment in point and non-point source
control; (4) Management programmes in control lagoons and wetland; and (5)Public awareness
and public education.
Specific activities to implement the Baltic Sea programme are: (1) Emergency support
and systems; (2) Municipal wastewater treatment; (3) Combined municipal and industrial
wastewater treatment; (4) Pulp and paper industry environmental control; (5) Environmental
control of other industries; (6) Solid and hazardous waste management; and (7) Air quality
management
The costs to implement such an action programme are enormous. The implementation of
the entire long-term programme is estimated to cost at least 18 billion ECU over a 20-year
period. The first phase should be implemented during 1993-1997 at a cost of 5 billion ECU; the
costs of the projects to Denmark, Finland, Norway and Sweden together would be 1.5 billion
ECU. The second phase, 1998-2012, is estimated to require an additional 13 billion ECU. In
1991 Finland launched an Action Programme to promote and support environmental protection
projects in the neighboring areas.
The formerly centrally-planned economies are going through dramatic economic
restructuring. The short-term prospects for major economic improvements are limited.
Therefore foreign funding will be an important source of financing of the Joint Comprehensive
Programme in those countries. Foreign financing should come from bilateral funding, the
European Union and international financing institutions, e.g., the European Bank for
Reconstruction and Development, the European Investment Bank, the Nordic Investment Bank,
the Nordic Environmental Finance Corporation and the World Bank. Joint financing is often the
method of choice. Overall, however, cooperation amoung the countries around the Baltic Sea
has been favorable, and even if the steps for the protection of the environment have not
proceeded as fast as hoped, we have already achieved appreciable results. One can ask what the
state of the Baltic Sea would be without our joint efforts so far.
CHALLENGES FOR THE FUTURE
In summary, our challenges for the future to preserve the high standard of Finnish water
may be summarized as follows: (1) Better understanding of environmental cause-effect
relationships; (2) Integration of environmental legislation and procedures; (3) BAT, life-cycle
approach; (4) Better integration of land use and water management (Agriculture and forestry);
(5) Broader use of environmental instruments (Economic instruments, Ecoauditing, EIA,
environmental awareness); and (6) International cooperation to solve the environmental problems
hi the neighboring areas.
309
-------
SUPPORTING DOCUMENTS
Helsinki Commission. 1993. The Baltic Sea Joint Comprehensive Environmental Action
Programme. Baltic Sea Environmental Proceedings, No. 48. Helsinki, Finland.
Matthews, PJ. 1995. Water quality objectives: a tool to ensure environment protection and
wise expenditure. In: River Basin Management for Sustainable Development. International
Specialized Conference, Kryger National Park, South Africa, May 15-17,1995. Volume I,
Paper 2,14pp.
Ministry of the Environment. 1989. Water Protection Programme to 1995: Decision-in-
Principle by the Finnish Council of State. Helsinki, Finland. Environmental Protection
Department, Series B 14,1989. 24 pp.
Ministry of the Environment. 1996. National Environmental Policy Programme. Helsinki,
Finland. 151 pp.
National Board of Waters. 1974. The Principles of Water Pollution Control up to 1985.
Helsinki, Finland. 40 p., Appendices: 3 maps. Publications of the National Board of
Waters No. 12.
Nyroos, H. 1984. The significance of water quality parameters for decision-making in water
pollution control - a case study: Lake Paijanne in Finland. Water Science and Technology
16:407-418.
Nyroos, H. 1989. Assessment of water quality in the planning of water protection measures in
Finland. In: Laikari, H. (Ed.). River Basin Management - V: Proceedings of an IAWPRC
Conference Rovaniemi, Finland. My 31 - August 4,1989. Oxford, Pergamon Press.
pp.413-418.
Nyroos, H. 1994. Water Quality Assessment hi Water Protection Planning. Helsinki, Finland.
Publications of the Water and Environment Research Institute No. 14. National Board of
Waters and the Environment. 85pp.
Nyroos, H. 1995. Water quality assessment hi the strategic'water protection planning. In:
River Basin Management for Sustainable Development. International Specialized
Conference, Kryger National Park, South Africa, May 15-17,1995. Vol. II, Paper 44. 10 pp.
• 310
-------
WATER QUALITY MONITORING IN RUSSIA
Vladimir V. Tsirkunov1
ABSTRACT
Russia has a long history of the development of geophysical monitoring systems,
including Water Quality Monitoring System (WQMS). Unfortunately, positive potential
accumulated in Russia and the former Soviet Union during many decades has been seriously
threatened in recent years as a result of ecomomic decline and state disorder. The objective of
this presentation is to describe briefly the history of water quality monitoring system, to outline
the issues the system is currently facing, and to discuss the perspectives of WQMS development
in Russia.
HISTORY OF DEVELOPMENT AND CURRENT STATUS
Initial studies of the river water quality were made in Tsarist Russia at the end of the last
century. At the same time, hydrological observations of surface and ground water bodies were
initiated. For example, in 1993 it was one the hundredth anniversary of the observations of a
ground water regime at site Nl in the Kamennaya Steppe in Boronezh Oblast. Before the
Second World War, concentations of major ions, pH, temperature, color, transparency, and some
other basic parameters had been regularly measured at a few hundred river sites. Regular river
water quality data started to be published hi 1936 in the Hydrological Annuals.
After the War, the monitoring network was expanded to 1500-2000 sites located on major
rivers, lakes, and reservoirs. At that time measurements of nutrients, Fe, Si, O2, and COD
became a part of the routine monitoring program. Regular monitoring of such pollutants as oil
and grease, surfactants, phenols, and heavy metals have been underway since 1968, and
chlororganic pesticides have been monitored in surface water bodies for about 20 years.
Biological monitoring of water ecosystems was introduced in late 1970s and early 1980s. The
observations of ground waters were initially concentrated on the aquifers closest to the surface,
and in 1950s-60s the network was expanded to include deeper confined aquifers as well. Since
that tune the network has been expanded continuously, and hi 1970s there were about 22,000
sites on the territory of the former Soviet Union (FSU). The initial objective of the network was
World Bank, Moscow Office, Moscow, Russia.
311
-------
to study the regularities of variations in the natural ground water regime. However, due to
growing anthropogenic impact, more and more observations in the areas of water intakes,
irrigated lands, industrial and mining enterprises have been carried out.
The present water quality monitoring system in Russia has been inherited from the FSU.
It was designed in the mid 1970s and as the result the National System of Observation and
Control of Environmental Pollution was established (Izrael et al. 1978). The System has been
formed primarily on the basis of operational and scientific institutions of the USSR
Hydrometeorological Service (surface waters), Ministry of Geology (ground waters), and
Ministry of Reclamation (water use and impact monitoring). At the end of the 1970s it was one
of the largest national water quality monitoring systems on the globe. Compared to similar
systems in the USA in the 1970s, it was certainly less advanced in terms of laboratory
capabilities, sampling equipment, and logistics support, but it had some other advantages, for
example, almost all monitoring sites were provided with measured or calculated hydrological
parameters, unified analytical techniques were used hi all laboratories, routine aquatic biology
observations were initiated and national reviews on water quality were published. However,
during the most recent 20 years, the water quality monitoring system was not significantly
modified and thus, it is now outdated and does not meet actual demands.
Functioning of WQMS in Russia is regulated by a number of governmental decrees, but
primarily by guidelines and regulations of different federal agencies. The "Fundamentals of
Water Legislation of Russian Federation" determines federal monitoring of water bodies as a
"purposeful system of regular observations of condition of water bodies and changes occurring
within them under the impact of natural and anthropogenic factors to obtain information, assess
and predict the condition of water bodies and their relationship with environment." There is also
'a clause stating that monitoring is funded from the federal budget and budgets of the Subjects of
Federation (Autonomous Republics, Oblasts, Krais, etc). Today, water quality monitoring is still
almost fully supported by federal agencies, and most of the work on water quality monitoring is
being carried out by the following agencies:
Russian Ministry of Environmental Protection and Natural Resources (Minpriroda).
Coordination of monitoring activities of different agencies, organization of monitoring of sources
of anthropogenic impact and zones of their direct influence, organization of the services of
analytical control inspections;
*
Federal Service of Russia for Hydrometeorology and Environmental Monitoring
(Roshydromet). Ambient water quality and quantity monitoring of water bodies, baseline
monitoring, development and approval of the techniques of chemical analyses, preparation of the
Surface Water section in the Water Cadaster;
Russian Federation Committee for Water Management (Roscomvod). Monitoring of
man-made water systems and hydroconstructions in places of water intakes and waste water
discharges, preparation of the section "Water use" in the Water Cadaster;
312
-------
Russian Federation Committee for Geology and Use of Entrails of the Earth
(Roscomnedra). Monitoring of ground waters, preparation of the Ground Water section in the
Water Cadaster;
Russian Federation Committee for Sanitary - Epidemiological Inspection
(Roscomsanepidnadzor). Monitoring of the sources of drinking water supply, monitoring of
drinking water quality;
Russian Federation Fishery Committee (Roscomrybolovstvo). Monitoring of major water
bodies used for industrial fishing as well as monitoring offish and other aquatic biota,
development of maximum allowable concentrations (MACs) for pollutants in aquatic
ecosystems.
All monitoring systems of federal agencies are based on a similar principles and have
hierarchical structure of several levels. In general, they include: (1) a network of fixed
observation sites; and (2) subdivisions in Subjects of Federation (Minprirody,
Roscomsanepidnadzor), and regions (Roshydromet, Roscomnedra) or basins of major rivers
(Roscomvod, Roscomrybolovstvo). The federal level is represented by an agency's headquarters
as well as by leading agency research institutes providing scientific, methodological and
technological guidance. Coordination of interagency activities in water quality monitoring
should be conducted by Minpriroda, however its capabilities in this respect are quite limited and
actually, every agency pursues its own policy.
One of the weakest parts in Russian monitoring system is the monitoring of pollution
sources. It is based primarily on wastewater quality and quantity data provided by major water
users. A number of reasons make the data very unreliable and their value may deviate by order
of magnitude from the true figure. Though the Roscomvod system of water use accounting
covers more than 46,000 water users, some point sources of pollution are not included in this
system. There is a lack data on discharges of storm an drainage waters, wastewaters from large
feeding lots, poultry operations and other agricultural activities, facilities of mining industries,
etc. Only 10-15 general water quality parameters are usually measured in wastewaters.
Measurements of toxicity of wastewaters are very limited and not used for regulatory purposes.
Minpriroda has a service of special inspections of analytical control consisting of 189
subdivisions in Subjects of Federation. The basic functions of this service are to control self-
monitoring data of water users, to obtain the data on composition and properties of discharges,
and data on the efficiency of water protection measures. This analytical inspection service
controls more than 20,000 point sources of pollution and 10,000 industrial laboratories. The
service may control point sources once a year at best. Analytical methods used for this control
are labor consuming and the instruments are outdated and inefficient.
313
-------
SYSTEM OF GROUND WATER MONITORING
The system is based on the regional regime observation network and the reference
network. At the observation network the wells are grouped in 5-10 cm cross-sections to monitor
ground water flows from surface waters or in clusters or group piesometers to observe flows
between aquifers. This regional network is intended to study the natural (background) regime
and the quality of ground waters, to assess and study formation of ground water resources, and to
predict changes. The network covers the upper hydrodynamic zone of all major artesian and
river basins. The so-called reference system has been introduced at local sites typical in terms of
hydrogeological conditions or sites important for economic consideration.
In 1994 the monitoring network in the territory of Russia consisted of 17,927 wells:
6,686 of these were intended to study ground water regime in natural conditions, and 11,241 in
disturbed conditions. In particular, 6,925 wells were located at water intakes, 859 in the areas
influenced by reservoirs, 1,462 in the areas of drainage and irrigation of lands, and 1,754 in the
urban territories. The period of observation at most of the sites is longer than 30 years.
Monitoring of ground water table regimes is carried out from three to ten times a month, and
water quality parameters are determined four times a year. Concentration of major ions and
water salinity are measured regularly while a wider list of parameters is measured randomly
(there is no recommended frequency of measurement). In case considerable variations in
concentration of some components or pollutants are detected, these components are included hi
the monitoring program and sampling frequency may be increased.
Informational structure of ground water monitoring consists of three levels: local (site-
specific), regional (territorial), and federal. Data generalization and presentation to the
authorities is conducted at each of the levels. At the federal level the data on ground water
regimes are published annually as an informational bulletin and included to the Federal Report
on State of the Natural Environment.
SYSTEM OF SURFACE WATER QUALITY MONITORING
The goal of the system is "systematic data acquisition on water quality of water bodies
and streams and provision to central administrative and economy authorities as well as to all
organizations concerned systematic information and predictions of the level of pollution of water
bodies and emergency information on sharp variations in the level of water pollution"
(Roshydromet 1992). Organization and conductance of observations are determined by the
following basic principles: comprehensiveness and regularity of observations, coordination of the
time of their conductance with typical hydrological phases, and measurement of parameters
using unified methods. The system is based on the network of fixed stations of so called "regime
observations" located at water bodies both in the areas of considerable anthropogenic impact and
unpolluted locations. The sites are selected taking into consideration the present use of a water
body for the needs of the economy and future development plans. In accordance with existing
regulations and guidelines (Izrael et al. 1978, Roshydromet 1992), the monitoring system should
314
-------
also include field surveys for more accurate location of fixed sites and determination of
monitoring programs. However, this type of monitoring activities is very limited primarily due
to poor financial support.
In 1994, the network of Roshydromet (the agency collecting most of the data on the
quality of water bodies) consisted of 1,892 sites including 2,604 cross-sections located at 1,326
water bodies (1,175 streams, 83 lakes and 68 reservoirs). These monitoring sites are subdivided
into four categories determined by economic significance of the water body, water quality, size,
volume, and water flow as well as other factors. By 1994 the network included 12 sites in
Category 1,40 sites in Category II, 709 sites in Category III, and 131 sites in Category IV.
Observations of inland surface water pollution are based on physical, chemical, and
hydrobiological parameters with simultaneous measurement of hydrological parameters.
Measurement of physical and chemical parameters is conducted at all the sites and
hydrobiological parameters are measured at 10% of the sites. More than 90% of monitoring sites
are provided with measured or calculated hydrological parameters. Observations at the sites are
conducted according to certain programs depending on the site category. Sampling based on a
obligatory standard program, including the measurement of 35-40 physical and chemical
parameters, is usually carried out at all the sites in major hydrological phases four times a year
for lakes and reservoirs and six or seven times a year for streams and rivers. Daily, monthly, or
10-year sampling (according to a reduced list of parameters) is added correspondingly to the sites
of Categories I-III (Table 1).
Table 1. Program and frequency of water quality observations for hydrochemical parameters
(from Roshydromet 1992).
Frequency of
Observations
Every day
Monitoring program for the sites of different categories
Category
I
Contracted program 1 (CP1) = water discharge, m3/s
(for streams) or water level, m (for lakes and reservoirs);
visual observations; temperature, °C; specific
conductance, Cm/m; dissolved oxygen, mg/L
Category Category
II III
Visual
observations
Category
IV
Every decade Contracted program 2 (CP2) = (CP1) + pH, suspended Contracted
substances, mg/L; biochemical oxygen demand (5 days), program 1
mg/L; concentration of 2-3 pollutants, typical for the
particular site
Every month Contracted program 3 = (CP2) + stream velocity, m/s (for reference discharge
measurements); concentration of all pollutants for the particular site
During major Obligatory program = (CP2) + stream velocity, m/s (for reference discharge measurements);
hydrological color, degrees; transparency, sm; odor, ranks; reduction-oxydation potential (Eh), mV; dissolved
events gases: oxygen, carbon dioxide, mg/L; major ions: Cl% SO42% HCCV, Ca2+, Mg2*, Na+, K+, IDS,
mg/L; nutrients: NH4+, NO2% NO3VPO43-, Feto
grease, volatile phenols, heavy metals, mg/L.
, Si mg/L; widely distributed pollutants: oil and
315
-------
Measurement of aquatic biology parameters at the least polluted sites is conducted every
month during the vegetation period (Categories I-III) or quarterly (Category IV). At the sites
with a higher pollution level some parameters subjected to the most serious variations when
exposed to anthropogenic impact, may be measured every 2 to 3 years.
Collection and shipment of samples is one of the weakest parts in the system of
observation of inland surface water quality. About 75% of sampling is carried out by the
Hydrological Service of Roshydromet. The samples are sent to laboratories by regular mail, so
for 30-35% of the sites shipment of the samples may take more than 10 days. The samples are
sent to 80 analytical laboratories belonging to 22 Roshydromet Regional Departments. Major
laboratories conduct analyses for 50-60 parameters. The laboratories are equipped primarily by
home made devices, optical and electrochemical instruments, gas chromotographs, etc. Most of
the instruments are outdated and there is a shortage of spare parts, supplementary equipment,
reagents, materials, glassware, sampling equipment, etc. There is, however, a system of
chemical analysis quality'control. Every year the laboratories analyze standard samples of a few
individual substances distributed among the laboratories. However, the volume of this work is
insufficient and quality control is not conducted at other stages of acquisition and processing of
data.
Data obtained in the analytical laboratories goes to the regional computer centers for
recording and processing. Roshydromet Regional Departments prepare, publish, and disseminate
to the concerned organizations water quality information on the territories they monitor. Such
information for the whole territory of Russia is prepared by the Hydrochemical Institute
(physical and chemical data on surface water quality) and the Institute of Global Climate and
Ecology (aquatic biology data). In addition to routine "regime" information, the existing surface
water quality monitoring system should provide short-term (operational) information primarily
in the form of cables and reports on both high and extremely high pollution events of the water
bodies monitored.
A publication, "The Annual on Surface Water Quality in the Russian Federation," is the
basic information document used in preparation of all major water management projects. This
document includes water quality information for hydrographic areas (sea basins) of Russia and
major river basins. Water quality is estimated by means of analysis of the frequency and value
of maximimum allowable concentration (MAC) exceedances for individual substances as well as
the information in the cases of high and extremely high pollution. The attempts to use various
modifications of integrated indices have been made. Conclusions on improvement or
deterioration of water quality are made based on comparison with previous years. The Annual
presents available information on the cases and sources of unfavorable status of water quality and
its variations as well as lists of the most heavily polluted water bodies. In Russia hi 1992, at 861
observation sites at 349 water bodies, concentrations of one or several parameters exceeded
MAC 10 or more times, and 30-fold or higher exceedance of MAC was registered at 82
observation sites located at 66 water bodies (Roshydromet 1993). Some characteristics of
pollutant concentrations in ambient water bodies are presented in Table 2.
316
-------
Table 2. Number of measurements made for specific components and percentage of MAC exceedance
in the surface waters of the Russian Federation (1994) (from Roshydromet 1992).
Parameters
02
BOD5
so42-
ci-
TDS
NH/-N
N03-N
NO2-N
Fe
Cu
Zn
Ni
Cr6+
Pb
Hg
Cd
As
Phenols
Oil and grease
Surfactants
DDT
Hexachloro-
cyclohexane
Number of
measurements
23,982
18,756
12,274
13,174
12,917
17,947
15,169
17,075
14,833
16,599
14,866
6,042
4,363
3,896
2,776
1,317
523
15,924
18,189
14,375
4,985
4,988
Maximum allowable
concentration (MAC), mg/L
Not less than 4.0 in winter,
and 6.0 in summer
3.0
100
300
1,000
0.39
9.0
0.02
0.1
0.001
0.01
0.01
0.001
0.03
0.00001
0.001
0.05
0.001
0.05
0.1
absence
absence
Median,
mg/L
9.83
1
2.13
28.5
11.5
: 211.2
0.21
0.27
0.01
0.25
0.004
0.009
0.002
: o.ooo
0.001
0.000
0.000
0.008
0.001
0.117
0.002
0.000
0.000
Percentage of measurements with
concentrations exceeding:
1MAC
4.04
41.1
11.4
1.56
3.17
30.75
0.09
24.84
68.05
77.77
41.67
15.60
7.58
12.13
19.2
1.50
0.28
41.22
60.10
5.34
14.7
34.8
10 MAC
_
0.14
0.70
0.28
0.23
1.06
_
1.93
9.04 .
16.77
1.60
1.85
1.15
0.02
_
_
_
6.37
9.80
_
14.7
34.8
50 MAC
.
0.005
0.10
0.022
0.03
0.005
_
0.146
0.19
1.17
0.24
0.63
0.48
„
_
_
_
0.20
0.47
_
14.7
34.8
Water quality information collected and presented by different agencies is generalized by
Minpriroda in the section "Water Resources, Their Status, Protection, and Use" of the Federal
Report on the State of Natural Environment in the Russian Federation. Another important state
document is Water Cadaster where three sets of data on surface waters (presented by Hydromet),
ground waters (Geolcom), and water use (Roscomvod) are summarized in jointly agreed formats'.
: 317
-------
Regularity and long periods of routine observations enabling the accumulation of long
series of observations at the monitoring sites and, making it possible to detect long-term trends in
parameters, may be attributed to advantages of the existing monitoring system in Russia.
Unique data on long-term variations of chemical composition of river and ground waters in
different physico-geographical zones exposed to different rates of anthropogenic impact have
been accumulated for the FSU. An attempt to summarize and present part of this information
will be made in the monograph "Assessment of Water Quality in the Territory of the Former
Soviet Union." Preparation of this monograph is supported by the World Health Organization
(WHO) and the United Nations Environmental Program (UNEP) in the frameworks of the Global
Environmental Monitoring System (GEMS) Water Program.
DISADVANTAGES AND ISSUES OF RUSSIAN WQMS
In addition to disadvantages mentioned above there are some other disadvantages and
issues, the most important of which are the following: (1) Lack of clearly formulated goals and
objectives of monitoring system reflecting its close relationship with environmental management
system; (2) Unclear distribution of responsibilities between the agencies at the federal level and
frequent organizational reconstructions, poor coordination at interagency level, resulting in
duplication of some functions and loose of others; (3) Outdated principles of monitoring based
primarily on fixed station approach and, thus, quite limited possibilities for estimation of spatial
parameters of pollution and detection of emerging issues; attempts to answer all the questions
within the limits of single system managed from the center; (4) Lack or complete absence of
specific monitoring programs for the basins of major rivers and aquifers as well as the
monitoring programs for assessment of serious pollution problems (eutrophication, salinization,
etc.); (5) Insufficient number of observation of bottom sediments and suspended solids, physical
parameters of habitats, concentrations of toxicants in organs and tissue of biota, ecotoxicological
parameters, etc.; (6) Outdated instruments and equipment and, as the result, quite limited
possibilities to detect toxicants, pesticides, and heavy metals; (7) Inefficient quality
assurance/quality control systems of monitoring activities and, hence, low reliability of
monitoring results; (8) Water quality assessment based on outdated and inflexible MAC system-
-lack of modern information technologies and limited possibilities for dissemination, processing
and presentation of information; (9) Poor accessibility, low reliability and sometimes absence of
information on land and water use, pollution sources, data on population and economic activities;
and (10) Inadequate financial and logistics support of the monitoring activities.
The above disadvantages indicate that water quality monitoring system in Russia is
outdated and does not meet the objectives of modern monitoring systems (U.S. EPA 1987, U.S.
ITF 1992), i.e., the existing Russian WQMS is not able to provide adequate information to assess
water quality status and trends, identify existing and emerging problems, provide information to
support development of priorities, plans, and programs of water quality management, and
evaluate effectiveness of programs.
• 318
-------
A new economic and political situation in Russia has created new needs and problems in
water quality management and monitoring. One such problem is related to the tendency for
decentralization and transfer of larger parts of authorities to the level of the Subjects of
Federation (Republics, Oblasts, Krais, etc.). Thus, the rights and responsibilities of the territories
in environmental quality management have been increased. However, water quality monitoring
is still supported and carried out by the federal agencies. The process of signing the River Basin
Agreements on the use and protection of water resources by the Subjects of Federation sharing
major river basins is under way. Development of basin monitoring systems should be one of the
components of the agreements. '
Another urgent problem is how to minimize negative consequencies of the dramatic
decrease of financial and technical support for monitoring system operations. Such problems are
particularly urgent for the vast northern and eastern territories of the country where monitoring
was based almost exclusively on the activities of observers-hydrologists who lived in those
remote areas near monitoring sites. Under existing constraints the observers cannot survive,
thereby abandoning monitoring sites and breaking long-term observation records. Wider
application of seasonal surveys, remote sensing and unattended automatic water quality
monitoring equipment may be considered as possible alternatives. Due to poor condition of most
industrial and transport facilities, deterioration of working discipline, outdated treatment
facilities and other factors, the risk of major accidents with dangerous environmental
consequences has been considerably increased. This requires development of early warning
monitoring systems in the most valuable and vulnerable areas with potentially high accident
risks. These systems should be designed for the earliest detection of pollution events, protection
of major water users, and decrease of level of damage.
An issue which has appeared recently after disintegration of the Soviet Union and
formation of the Russian Federation as an independent state is monitoring of transboundary
water bodies, especially on the boundaries with Ukraine, the Baltic States, and Kazakhstan.
When pollution becomes multimedia, and infiltrates several compartments within the
environment, it is particularly important to turn down the traditional institutional boundaries and
improve interagency coordination for development of integrated monitoring systems. This will
require cooperation and data exchange with existing air and soils monitoring systems, closer
relationship between ground and surface water monitoring, monitoring of pollution sources and
monitoring of bottom sediments and biota.
Increased international support and more intensive international cooperation are also
urgently needed now. It seems to me that the USA and Canadian experience-will be especially
useful for improvement of the Russian Federal WQMS because these countries are comparable
with Russia in terms of physiographic conditions, dimensions, and diversity and complexity of
water quality problems. At the same time, the experiences of France, Netherlands, Germany, UK
and other nations can be very appropriate for designing regional and river basin monitoring
networks in Russia. Cooperation with such international organizations as WHO, UNEP, World
Meteorological Organization (WMO), United Nations Development Program (UNDP),
International Association of Hydrolqgical Sciences (IAHS), and others, is also very important.
319
-------
PERSPECTIVES FOR DEVELOPMENT
Summarizing the foregoing one may say that water quality information needs have never
been so great in Russia, and the possibilities for obtaining it so limited. To resolve this
contradiction and transfer the present WQMS into a modern integrated system satisfying the
basic needs of water resources management is an extremely complicated problem. However, it
seems quite certain that full-scale national reconstruction of the whole monitoring system is both
impossible and unfeasible under the present economic situation. In this relation, plans of the
Ministry of Environment to develop an ambitious Integrated Federal System of Ecological
Monitoring which has to include "systems of monitoring of natural'environment, natural
resources, natural engineering systems, natural complexes, ecosystems and sources of
anthropogenic impact" should be reconsidered. Reconstruction of the WQMS should be made
with great care, to prevent desintegration and destruction of the existing network which is the
only source of environmental quality information. Gradually well designed and tested
modifications can be introduced in the system. The reconstruction could possibly be initiated
simultaneously at the federal and regional (basin) levels.
At the federal level such activities should be implemented that do not need considerable
expenditures but have significant effects. Such measures comprise development of necessary
legal and organizational prerequisites for improvement of monitoring systems including
development and coordination of the goals and objectives of monitoring systems based on the
management needs, interagency coordination and information exchange, development and
introduction of new systems of water quality evaluation including modification of the present
system of MAC and others. Recommendations on priority and order of improvements should be
developed by a specially created Interagency Task Force including, in addition to representatives
from the agencies responsible for WQMS, scientists, managers, representatives of major water
users, and ecological organizations.
Further improvement of the programs of monitoring including cessation of observations
of poorly informative parameters, reduction of their frequency and increase of the number of
observations of bottom sediments, biota, toxicological parameters, etc. is necessary. Special
attention should be paid to gradual introduction of the so called synoptic surveys which allow
identification of emerging water quality problems and document spatial distribution of
concentrations and loads of variables (Hirsch et al. 1988). It seems that conclusion made by
Rickert (1991), that for the developing countries, a synoptic survey approach is superior to the
fixed station approach, is applicable to some extent to Russia in the present situation.
*"""
In case of adequate financial support, then improvements should be made in sampling
systems and analysis, introduction of modern informational technologies, elaboration of the
system of quality control/assurance, and unification of observations between different monitoring
subsystems (surface, ground waters, pollution sources) to get comparable results.
At the regional and basin levels it is necessary to start designing one or several integrated
water quality monitoring systems in the regions or river basins with the most serious water
320 .
-------
resources and ecological problems where developed institutional and scientific infrastructure as
well as strong support from regional and local authorities are available. In the process of
designing and developing such monitoring systems it is possible to gain certain experience,
elaborate interagency cooperation and train personal. It should be noted that some of the
regional and municipal governments where enviornmental and water quality concerns are
especially strong started to fund monitoring activities to study possible negative impacts on
public health and ecosystem integrity.
The pilot project "Prototype Integrated Water Resources Monitoring System" prepared in
the frameworks of the Russian Environmental Management Project funded by the World Bank
Loan, may be used as a basis for development of such system. The scope of the project was
prepared jointly by experts of the U.S. Geological Survey, U.S. Environmental Protection
Agency, World Bank, and from Russia. The project will be implemented hi the Lower Don
River Basin for the period of 3.5-4 years. The following steps are supposed to be taken
sequentially: development of a data base for water resources management; improvement of Ihe
present system of water resources data collection; identification of the most serious data gaps and
collection of selected data; assessment of current water quality conditions; identification of
management needs; and designing of a new monitoring program. Experience gained in the
process of the pilot project implementation will be disseminated to the other regions and river
basins of Russia. There are number of other bilaterally (e.g., U.S. Agency for International
Development) and internationally funded projects (e.g., WHO, and the Technical Assistance
Program of the European Union for New Independent States and Mongolia (TACIS)) fully or
partially aimed at WQMS improvement. Such projects can provide the seeds of new elements of
future water quality monitoring systems. WQMS in the future should be based on a modified
federal system which will use the limited available resources in the most efficient way and
generate data critically needed for federal decision making. The system should, at the same tune,
take advantage of all innovations and positive experience gained at regional and local level as the
result of actitities of regional and municipal governments, private sector as well as externally
funded projects.
Today, Russia is facing extremely complicated and intertwined social, economic, and
ecological issues. The problem discussed is only a small fragment in this complex spectrum of
major problems. It is difficult to predict perspectives for WQMS development under such
circumstances. What is clear, though, is that the federal government will not be able to afford (as
it once did in the USSR) to finance massive monitoring systems comprising all monitoring
activities. Such systems in the future should meet the needs of various users (basin and
municipal authorities, private sector, etc.) and rely upon various financial sources with strong
technical guidance from the federal government.
DISCLAIMER
This article is a product of the author. Judgements contained herein do not necessarily
reflect the views of the World Bank.
321
-------
REFERENCES
Hirsch, R.M., W.M. Alley, and W.G. Wilder. 1988. Concepts for a National Water Quality
Assessment Program. U.S. Geological Survey. Circular N 1021. 42pp.
Izrael, Yu. A., N. K. Gasilina, Ph. Ya. Rovinsky, and L. M. Philippova. 1978. Functioning of
USSR Monitoring System of Environmental Pollution. Leningrad, Gidrometeoizdat. 117
pp. (In Russian).
RickertjD.A. 1991. Water quality assessment. A new concept for measuring the quality of
surface and ground water. In: Information Needs for Water Quality Assessment and
Management. WMO, Technical Reports hi Hydrology and Water Resources, Number 34. p.
39-79.
Roshydromet.. 1993. Annual on Surface Water Quality of the Russian Federation hi 1992.
Hydrochemical Institute. Obninsk, VNIIGMI-MTsD. 387pp. (In Russian).
Roshydromet. 1992. Leading Document. Instructions on methodic. Environmental Protection,
Hydrosphere, Organization and Implementation of Routine Observations of Inland Waters
Pollution on Roshydromet Network. L.D. 52,24,309-92. St. Petersburg. 67 pp. (In
Russian).
U.S. EPA (U.S. Environmental Protection Agency). 1987. Surface Water Monitoring: A
Framework for Change. Washington, D.C., September, 1987. 41 p.
U.S. ITF (U.S. Intergovernmental Task Force on Monitoring Water Quality). 1992. Ambient
Water Quality Monitoring in the United States. First Year Review, Evaluation, and
Recommendations. Washington, D.C., December, 1992. 26 pp.
322
-------
WATER QUALITY REGULATION AND RESEARCH IN THE UNITED STATES
ENVIRONMENTAL PROTECTION AGENCY
Rosemarie C. Russo1
ABSTRACT
The mission of the U.S. Environmental Protection Agency (EPA), and its unique role
within the U.S. federal government, is the protection of human health and the environment.
The function of the Office of Research and Development within EPA is to develop, direct, and
conduct a national research, development and demonstration program in pollution sources, fate,
and health and welfare effects; pollution prevention and control and waste management and
utilization technology; environmental sciences and monitoring systems. Scientists conduct
research to advance the state of the art in scientific and technical knowledge related to human
health and ecosystems; this includes analytical methods, monitoring techniques, and modeling
methodologies. The research conducted is in support of the reegulatory offices of EPA, to ensure
that EPA regulations are based on sound science. This paper presents background and current
information on regulations in the area of water quality.
During 1994 the Office of Research and Development undertook a comprehensive
analysis of its organization and structure, including both research laboratory and headquarters
components. This assessment led to a total reorganization that realigned twelve research
laboratories into a structure that is focused under a risk assessment/risk management model.
In 1995 the restructured organization established three new National Research Laboratories,
two new National Centers, and headquarters offices.
REGULATION AND RESEARCH
Historically, water quality requirements in the United States have been under
development since the mid-1950s, with acceptable concentrations of pollutant chemicals in
surface water bodies being suggested by a number of authors (Alabaster and Lloyd 1980, McKee
and Wolf 1963, NAS et al. 1973, U.S. EPA 1976,1987). In 1970 the national United States
Environmental Protection Agency (EPA) was created, and it was charged with administering the
Glean Water Act and other national environmental laws.
National Exposure Research Laboratory, U.S. Environmental Protection Agency, Athens, Georgia, USA.
323
-------
The mission of EPA, and its unique role within the United States federal government, is
the protection of human health and the environment. The function of the Office of Research and
Development within EPA is to develop, direct, and conduct a national research, development
and demonstration program hi pollution sources, fate, and health and welfare effects; pollution
prevention and control and waste management and utilization technology; environmental
sciences and monitoring systems. Scientists conduct research to advance the state of the art in
scientific and technical knowledge related to human health and ecosystems; this includes
analytical methods, monitoring techniques, and modeling methodologies. The research
conducted and results obtained are used in support of the regulatory offices of EPA, to ensure
that EPA regulations are based on sound science.
The first federal law in environmental protection hi the United States was the Rivers and
Harbors Act of 1899. The Water Pollution Control Act of 1948 was the first national law
enacted specifically to address water pollution problems. The Federal Water Pollution Control
Act of 1956 followed, with the Water Quality Act of 1965 being the third law for protection of
water quality. These laws established principles of federal and state cooperation in this area.
In the 1972 Amendments to the Federal Water Pollution Control Act, a discharge permit system
was established and federal grants were provided for hi order to assist local governments in
financing sewage treatment systems. The Federal Water Pollution Control Act was amended in
1977,1981, and 1987; the Act and its major amendments are generally known as the Clean
Water Act. The Safe Drinking Water Act was passed in 1974, and Safe Drinking Water
Amendments were signed hi August 1996; this is the primary law for protection of drinking
water sources in the United States Legislative mandates to address nonpoint source pollution are
primarily the Clean Water Act of 1972, as amended in 1987, and the Coastal Zone Management
Act (Reauthorization Amendments of 1990).
The administration of eleven principal United States environmental statutes is the
responsibility of EPA. Some of these environmental laws, including the water protection laws,
were established since EPA was created; others are amended versions of legislation first enacted
as far back as 1899. These eleven principal statutes are:
1. Clean Air Act (CAA)
2. Asbestos School Hazard Abatement Act of 1984 (AHSAA),
Asbestos Hazard Emergency Response Act of 1986
3. Research, Development and Demonstration Act (R&D)
4. Federal Insecticide, Fungicide and Rodenticide Act (FIFRA)
5. Marine Protection, Research, and Sanctuaries Act, including
Ocean Dumping and Ocean Incineration (MPRS A)
f
324
-------
6. National Environmental Policy Act, including Wetlands and Estuaries Issues (NEPA)
7. Solid Waste Disposal Act (SWDA) and Resource Conservation and Recovery Act
(RCRA)
8. Comprehensive Emergency Response, Compensation and Liability Act
(CERCLA/Superfund)/ Superfund Amendments arid Reauthorization Act of 1986
(SARA)
9. Safe Drinking Water Act (SDWA)
10. Toxic Substances Control Act (TSCA)
11. Clean Water Act (CWA)
A number of related laws also are in place to ensure environmental protection for both
human health and ecological systems. These include: Uranium Mill Tailings Radiation Control
Act; Federal Food, Drug, and Cosmetics Act; Nuclear Waste Policy Act; Public Health Service
Act; Acid.Precipitation Act; Lead Contamination Control Act; Emergency Wetlands Resource
Act; Food Security Act; Oil Pollution Act.
The regulatory program offices (Figure 1), located in Washington, D.C., are charged
with administering specific environmental laws; these offices are the Office of Water; Office of
Air and Radiation; Office of Prevention, Pesticides, and Toxic Substances; and Office of Solid
Waste and Emergency Response. The Office of Water is the office that issues water quality
regulations and administers the water quality laws; a summary of their information resources is
available (U.S. EPA 1990). There are ten Regional Offices (Figure 2), located across the
country, that implement the regulations and assist the states in developing water quality
standards for intrastate and interstate waters.
The Office of Research and Development (Figure 3), headquartered in Washington, D.C.,
with geographically dispersed research laboratories, conducts and supports research to provide
the scientific knowledge to ensure that regulations are based on sound science. The research
office also supports EPA's Center for Exposure Assessment Modeling, which distributes and
supports water quality and other predictive models, and provides technical assistance and
training in the use of these models. Figure 4 indicates how scientific and engineering research is
integrated into and provides support for the overall cycle of regulation development and
implementation. A wide variety of technical data and modeling tools are currently used as
underpinnings of environmental regulations, and improvements and updates are continually in
progress as additional information becomes available through research.
325
-------
immunicauons
Atfairs
ilrator lor Ct
, and Public
•£•2
"O O
f
<
|
~
1
•I
3
.*
9
»
Adffiinistrativ
1.
7
V
i:
<
§
§;
A
1
S
H
1
O
§
a Administra
-§
Q
i
i
1
E
i
c
*
c
f
"
c
1
1
c
I
a
S
8
o
15
|
J and Legisl
c
J5
0
01
a
•
i
^
E
<
1
>.
i
C
3
r
9
i
i
i
3
L
C
O
1
i
j
^
i)
i
s
>s
,§
.1
$
3
1
-a T
S o
IEnvironmgr
Cn
sr Regional
ministrator f
3
S
'a
1
I
£
a
Small and 0
i
i
to
C
s
1
UJ
ra
«
I
O
c
1
S
o
i
^w
—
•••I
Administrator!
donal Activities
Assistant
tor Interna
0>
c
5
o
D
O
|
§
a
13
ID
1
111
is 3
•§ M
*t o S
lit
<2 "*~ O
2 =
g|
o.Q
.E 13
11
IS
S £
ninistrator 1
soon and 1
nagament
•S E «
Assistant A
for Adminis
Resources h
—
—
^H
••M
1
VI
1 s
<|:
" O
I
|-D S
1*1
|l g.
Ill
^ ul
2
!E « °
•o S §•
Is^
l~
~ w" g
ill
=21
i 13
.a I?
ad: §
•S
S.fc
.- jo n
1 -1
111
J'"2
_
—
•MM
^Ml
•^•1
1J
OCQ
II
oco
c S
fS
^
(0
CT 'Jc
c 5-
O ta
ll
E
II
^ C
•2 yj
oc
'-
__
•"
aum
mamm
0
0 «
**
0>§
H
ll
I6
if
(D
326
-------
Rgure 2. EPA Regional Offices
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
AL
AK
AZ
AR
CA
CO
CT
DE
DC
FL
GA
HI
ID
IL
IN
IA
KS
KY
LA
4 Maine
10 Maryland
9. Massachusetts
6 Michigan
9 Minnesota
8 Mississippi
1 Missouri
3 Montana
3 Nebraska
4 Nevada
4 New Hampshire
9 New Jersey
10 New Mexico
5 New York
5 North Carolina
7 North Dakota
7 Ohio
4 Oklahoma
6 Oregon
ME
MD
MA
Ml
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
-NC
ND
OH
OK
OR
1 Pennsylvania
3 Rhode Island
1 South Carolina
5 South Dakota
5 Tennessee
4 Texas
7 Utah
8 Vermont
7 Virginia
9 Washington
1 West Virginia
2 Wisconsin
6 Wyoming
2 American Samoa
4 Canal Zone
8 Guam
5 Puerto Rico
6 Trust Territories
10 Virgin Islands
PA
Rl
SC
SD
TN
TX
LTT
VT
VA
WA 10
WV 3
W!
WV
AS
CZ
GU
PR
TT
VI
Region 1
Region 2
CT Regions
ME
MA
NH
Rl
VT
NJ
NY
PR
VI
DE Region 4
DC
MD
PA
.VA
WV
AL Regions
FL
GA
KY
MS
NC
SC
TN Regions
'
IL Region 7
IN
Ml
MN •
OH
Wl Regions
AR
LA
NM
OK
TX
5
8
9
9
2
IA Regions AZ
KS CA
MO HI
NE NV
AS
CO GU
MT
ND Region 10 AK
SD ID
UT , OR
WY WA
Figure 2. Regional offices of the U.S. Environmental Protection Agency.
327
-------
ffii
8.SS
Snn
B S
m
i.S'f
id!
a
ill
•afi
E
§•
(U
Q
ton
8
II
fin
I
Ji
-------
KM-saSMffl. ,. .. i.-xiij i&g:i
•jff.j •»S:-;/S'tiv<.^\i~l.::J»fcl^'A-i,ffi.<,i...-..;iXl.-v.>..:U.Wt'au«#3^fc;ij,1o-Ail
I
1:3
•§
•
03 0)
*3 §
tan 3
O
.a P^-
«S 9
^3 °
o ^
• i—(
!
I
too
329
-------
During 1994 the Office of Research and Development undertook a comprehensive
analysis of its organization and structure, including both research laboratory and headquarters
components. This assessment led to a total reorganization that realigned twelve national research
laboratories into a structure that is focused under a risk assessment/risk management model. In
1995 the restructured organization established three new National Research Laboratories, two
new National Centers, and headquarters Offices.
ORD also issued a Strategic Plan (U.S. EPA 1996) that described its framework for
research priorities. Nine strategic operating principles were set forth: (1) focus research and
development on the greatest risks to people and the environment; (2) focus research on reducing
uncertainty hi risk assessment; (3) balance human health and ecological research; (4) infuse
ORD's work with a customer/client ethic; (5) give priority to maintaining strong and viable core
capabilities; (6) nurture and support the development of outstanding scientists, engineers, and
other environmental professionals; (7) increase competitively awarded research grants;
(8) require the highest level of independent peer review and quality assurance; (9) provide the
infrastructure to achieve and maintain an outstanding research and development program. The
risk paradigm is the basic framework used to set the overall structure and research agenda of the
Office. The Strategic Plan also established the Office's research priorities for the next few years.
The six high priority areas are: (1) drinking water disinfection; (2) particulate matter;
(3) endocrine disrupters; (4) research to improve ecosystem risk assessment; (5) research to
improve health risk assessment; (6) pollution prevention and new technologies.
The mission (U.S. EPA 1996) of the Office of Research and Development, after its
reorganization based on the risk assessment paradigm, is to: perform research and development
to identify, understand, and solve current and future environmental problems; provide responsive
technical support to EPA's mission; integrate the work of ORD's scientific partners (other
agencies, nations, private sector organizations, and academia); provide leadership in addressing
emerging environmental issues and in advancing the science and technology of risk assessment
and risk management.
EPA issued the first water quality regulation in 1975. These were revised in 1983, and
the 1983 regulation is currently hi use. This regulation and supporting documents provide
guidance to controlling pollutants by use of numerical criteria for specific pollutants and also by
biotesting of effluents and receiving waters.
•
The first national water quality criteria were published in 1976 by EPA (U.S. EPA 1976)
and reviewed by the American Fisheries Society (Thurston et al. 1979). Subsequent to the 1976
criteria, the United States system evolved to provide guidelines for development of individual
national criteria, and specific criteria documents were prepared for a series of important chemical
pollutants. The states were charged with developing water quality standards and monitoring
compliance with these standards. Recognizing that local conditions might require less or more
stringent water quality requirements in order to provide water quality suitable for the water body
in question, EPA also developed a system whereby site-specific water quality standards could be
330 ' .
-------
established and adopted by a state or states. Point source pollutant generators, such as
municipalities and industries, were required to obtain National Pollutant Discharge Elimination
System permits for discharge of wastes into receiving water bodies. Guidelines were developed
for "whole-effluent" discharges as well as for individual chemical pollutants.
Protection of water quality attempts to achieve, through criteria and standards and their
implementation, the water quality goals of a specific water body for its designated use or uses.
The types of uses are described in the Clean Water Act: public water supplies; protection and
propagation offish, shellfish, and wildlife; recreation; agriculture; industry; navigation; coral
reef preservation; marinas; groundwater recharge; aquifer protection; and hydroelectric'power.
Currently, an integrated approach is used for water quality requirements, including both
whole effluent and specific chemical considerations, and is intended to protect both aquatic life
and human health (U.S. EPA 1991a). More recently, EPA has worked on developing sediment
quality criteria (U.S. EPA 1993, Ankley et al. 1996). As the water quality requirement system
continues to evolve, increasing emphasis is being placed on consideration of multiple stressors
(e.g., pollutants, land use, habitat changes, etc.) upon aquatic organisms, populations,
communities, and ecosystems. Increasing attention is also being given to nonpoint source
pollution. The research office of EPA is currently focusing on a risk assessment - risk
management approach to the study of exposure and effects of stressors on human health and the
environment.
Specific chemicals have discharge limits that are based on national water quality criteria
that are incorporated within a state's water quality standards. Water quality criteria are
developed using the method provided in "Quality Criteria for Water" (U.S. EPA 1987) and
"Guidelines for Deriving Numerical National Water Quality Criteria for the Protection of
Aquatic Organisms and Their Uses" (Stephan et al. 1985). Based on laboratory studies of acute
and chronic toxicity and bioaccumulation of specific chemicals to a variety of aquatic organisms,
a numeric water quality criterion is established for a specific chemical.
A number of compilations of aquatic toxicity data are available (Russo and Pilli 1984,
Mayer and Ellersieck 1986, Hollis and Lennon 1987, Wood 1987, MacPhee and Cheng 1989).
The United States EPA has what is probably the largest compilation of aquatic toxicity data, in
terms of both chemicals and species. This is the AQUIRE (Aquatic Information Retrieval)
toxicity data base (Russo and Pilli 1984), which is a computerized compilation of literature
toxicity data from acute and chronic tests for freshwater and marine organisms. AQUIRE
contains information from over 100,000 toxicity tests for 5000 chemicals on 2300 aquatic
species. Data included in this data base are rated for quality, based on the toxicity test
procedures used.
The toxicity of whole effluents is tested by acute and chronic toxicity tests using
wastewaters. These tests are intended to assess the overall effects of a multitude of pollutants
contained in waste effluents. Specific methodologies for conducting these toxicity tests are
331
-------
available for both freshwater and saltwater species (U.S. EPA 1991b, 1991c, 1991d). Analytical
methods are also available for analysis of chemical concentrations (U.S. EPA 1979,
APEAetal. 1985).
Figure 5 provides an overview of the process followed in the United States in going from
criteria to discharge permits. In the United States there are over 61,000 municipal and industrial
direct discharges of pollutants from point sources to waters. Under the National Pollutant
Discharge Elimination System, over 7,000 major permits and over 50,000 minor discharge
permits have been issued. There are approximately 15,000 publicly owned water treatment
works nationally, and about ten percent of these have pretreatment programs.
Water quality criteria for the protection of human health are established based on
carcinogenic, toxic, or taste and odor characteristics of specific pollutants. Aquatic exposure to
humans of pollutants may occur by a number of pathways, including direct ingestionfrom water,
consumption offish or other animals in water, inhaling the pollutant from ambient air, and
contact with skin. Most human health criteria are based on exposure data from water
consumption and aquatic organism consumption. Criteria for carcinogens are provided as a
range of water concentrations that are associated with a range of incremental risk of cancer.
There is no scientific basis for "safe" concentrations for carcinogens, therefore for maximum
protection of human health the exposure concentration would have to be zero. However, EPA
has provided (U.S. EPA 1987,1991a) arange of concentrations corresponding to incremental
cancer risks of 10'7 to 1O'5; i.e., one additional incidence of cancer in populations ranging from
10 million to 100,000. Criteria for non-carcinogens are provided as concentration levels at which
specific chemicals are not expected to produce adverse effects in humans. EPA supports a data
base (IRIS, Integrated Risk Information System) containing information on the potential adverse
human health effects of about 500 specific substances (U.S. EPA 1994b).
Water quality criteria alone are not sufficient to ensure protection of aquatic life and
human health. Lakes, rivers, wetlands, and coastal waters can also be subject to degradation due
to contamination of bottom sediments. The bioavailability of pollutants in sediments is the key
factor in determining quality criteria for chemicals in sediments. Toxicity tests conducted for
some chemicals in various sediments have shown that a range of toxicity exists in the different
sediments; consequently, toxicity data are not generalizable across different sediments.
Therefore, for national sediment quality criteria for nonionic organic chemicals, EPA has
used the equilibrium partitioning method (U.S. EPA 1993). That is, it is assumed that the pore
water and sediment carbon are hi equilibrium and the concentrations are related by a partition
coefficient. The sediment-pore water equilibrium system provides the same exposure to benthic
organisms as a water-only exposure; the use of pore water toxic units removes sediment-to-
sediment differences. To calculate a sediment quality criterion for a given.chemical (ibid.), the
chronic water quality criterion (Stephan et al. 1985) is multiplied by the partition coefficient
(ratio of chemical concentration hi the sediment and in the pore water at equilibrium).
*
332
-------
Define water quality
objectives, criteria, and standards
Establish priority
waterbodies
Chemical-specific Effluent
Characterization
- Evaluate for excursions above standards
I
Determine 'reasonable potential'
I
Generate effluent data
Whole Effluent Toxfcfty Effluent
Characterization
I
Evaluate for excursions above standards •
I
Determine 'reasonable potential"
I
Generate data
Evaluate exposure
(critical flow, fate modeling,
and mixing) and calculate
wasteload allocation
Define required discharge
characteristics by the
wasteload allocation
Derive permit
requirements
Evaluate toxfclry reduction
i
Investigate Indicator
parameters
Final permit with
monitoring requirements
J
Figure 5. Overview of the water quality-based "Standards to Permits" process
for toxics control. (From U.S. EPA 1991a)
333
-------
This number is then multiplied by 0.001 kilograms/gram to give the sediment criterion in
micrograms/gram of sediment. The partition coefficient is calculated from the octanol-water
partition coefficient of the chemical. To obtain the desired level of protection, the minimum
data requirements for sediment quality criteria for a specific chemical are the octanol-water
partition coefficient, the final chronic value, and sediment toxicity tests. The final chronic value
is obtained from the water quality criteria for specific chemicals, and the octanol-water partition
coefficient is either measured or taken from the literature.
Sediment toxicity tests must be designed to incorporate the assumptions of the
equilibrium partitioning approach, must represent a range of organic carbon content in the
sediments, and must include organisms that are sensitive to the given chemical. As appropriate,
toxicity tests using sediments from specific sites are needed to establish sediment quality criteria
for that site. Work is currently in progress on developing methodologies for establishing
sediment quality criteria for metals.. Sediment quality criteria are at this time only proposed and
not legally binding.
Nonpoint source pollution comes from diffuse sources and enters surface and ground
waters in many ways, often in surges or storm events and often in large quantities. Nonpoint
source pollution is the largest remaining category of contamination threatening water quality in
the United States. Effectively dealing with nonpoint source pollution is a much more difficult
task than correcting point source problems and requires different approaches. Nonpoint source
pollution is contributed by agriculture, urban runoff, atmospheric deposition, highway drainage,
hydromodification (stream channelization, flood prevention, lake drainage), silviculture,
livestock grazing, construction practices, land disposal activities, septic systems, landfills/spills,
and mining.
The Clean Water Act as amended established a national program to control nonpoint
sources of water pollution. Two major new requirements were established: reports from the
50 states describing water impairment due to nonpoint sources, the types of sources causing the
problems, and state and local control programs; and state programs for controlling nonpoint
source pollution including methods and a time frame for remedying problems. EPA established
a three-year grants program in 1990 through which the agency provided funds to states to support
approved state nonpoint source programs. The Coastal Zone Management Act Reauthorization
Amendments of 1990 required that states develop a Coastal Nonpoint Source Control Program.
The Act established a "technology-based" approach to nonpoint sources management, based on
the assumption that best management practices will meet water quality goals.
In the United States much of the regulatory activity directed toward controlling nonpoint
sources pollution is voluntary; that is, landowners and operators like farmers, ranchers, and
foresters are encouraged to adopt best management practices mat help control nonpoint sources
pollution while sustaining production arid profitability. Federal and state government institutions
provide technical and financial assistance in this activity. Legislative mandates addressing
nonpoint source pollution problems cut across many governmental units—local, state, and
334
-------
federal-and across several federal agencies; e.g., United States EPA, United States Department
of Agriculture, United States Geological Survey, United States Department of the Interior,
Federal Highway Administration.
Addressing nonpoint sources pollution with research studies has focused on the
watershed as the unit for assessment, management, and restoration. Problems are identified in
terms of watersheds, the surrounding land and associated land uses that contribute to the aquatic
system.
Environmental quality is affected not only by pollutant chemicals, individually and
collectively, but also by various other stressors caused by man's activities, such as global
warming, habitat loss, and decrease hi biological diversity. Therefore initial attempts have been
made to address the risk posed to ecosystems by both chemical and nonchemical stressors.
A framework for ecological risk assessment has been established by the U.S. EPA (1992), and
this framework is intended to provide a basis for discussion and future development of formal
guidelines for ecological risk assessment. A detailed discussion of risk science and policy is
given in "Understanding Risk" (Stern and Fineberg 1996), and a comparison of different
approaches is provided by Covello and Merkhofer (Covello and Merkhofer 1993).
Figure 6 provides a schematic diagram of risk assessment and risk management and how
they interrelate. Risk assessment is generally taken to include hazard assessment and exposure
assessment. Hazard assessment evaluates or predicts the adverse effects that stressors may have
on components of an ecosystem, such as survival, reproduction, biodiversity, etc. Exposure
assessment evaluates or predicts the stressor's contact with various ecological components;
exposure scenarios include consideration of sources, transport, transformation processes, routes
(e.g., ingestion, dermal), etc. Risk management attempts to prevent environmental degradation
and to remediate or restore areas where adverse impacts have already occurred.
The approach to risk assessment that is being taken by the U.S. EPA (1992) is illustrated
in Figure 7. The approach consists of several steps, and recommends early and continuing
discussion between those conducting a risk assessment and those charged with managing the
risk, based on policy considerations. The approach begins with formulation of the problem to
identify the goals, breadth and focus of the assessment. The stressors, ecosystem at risk,
ecological effects, and exposure must be characterized. Then indicators or endpoints to be
measured are selected and a conceptual model is developed. The conceptual model (NRC 1986)
addresses how the stressor might affect ecological components of the natural environment and
what exposure scenarios are likely. Based on the conceptual model, data on the potential effects
and exposure of the stressor are analyzed, and the likelihood of adverse effects occurring as a.
result of exposure to a stressor are evaluated. The process is intended to integrate environmental
risks, so that potential risks from multiple stressors will be addressed.
335
-------
Problem Identification & Formulation
o Discuss Drivers (Environmental Justice, Regulatory
Compliance, Socio-Economic, Technical, Cultural,
Stakeholder, Aesthetic)
o Analyze Management Options & Scenarios
o Build Decision Analysis (Assessment Endpoints)
o Assemble & Validate Conceptual Model
o Identify DQOs for Measurement Endpoints
A
N
A
T
Y
I
^~i
^-^-"
Environmental Characterization
Measure &
Predict
Effects
and
Stressor-
Rcsponsc
Profile
Measure &
Predict
Stressor
Properties
and
Exposure
Profile
V
£
Risk Characterization
o Integrate Exposure and Effects
o Quantify Uncertainty
o Interpret Human, Ecological, & Economic
Significance
o Develop Risk Communication Tools
DATA
M
O
N
I
T
O
R
I
N
G
M
O
E>
E
L
E
E>
1
N
P
U
T
S
Risk Management Decisions
o Is the question answered?
o Are there derivative questions?
o Is more data collection cost-effective?
o Reformulate the question?
Adaptive Management
o Monitor Results
o Analyze Outcomes
o Design Follow-on Options & Scenarios,
Figure 6. Framework for ecological and human health risk assessment and management.
(Courtesy of L. A. Burns)
336
-------
Discussion
Between the
Risk Assessor
and
Risk Manager
(Planning)
Ecological Risk Assessment
PROBLEM FORMULATION
A
N
A
L
Y
S
Characterization ' Characterization
of | of
Exposure . Ecological
1 Effects
RISK CHARACTERIZATION
Discussion Between the
Risk Assessor and Risk Manager
(Results)
I
; Risk Management
Z¥
O
I
o
03
o.
o
-------
In the years since passage of the Clean Water Act in the United States, much progress has
been made in decreasing pollution by sewage and industrial wastes of the nation's rivers, lakes,
and coastal waters. Water pollution problems from point sources, such as municipal and
industrial outfalls and other sources that are identified as coming from a clearly defined location,
still exist and work is continuing in this area. However, it is important to concentrate more
efforts on reducing and controlling nonpoint source pollution and contamination of sediments.
Another area requiring additional research is risk assessment and risk management at ecosystem,
watershed, and regional scales. It is also recognized that various "stakeholders" should have
some voice in water quality regulations; and, increasingly, states, municipalities, national and
local environmental organizations, industries, Native American tribes, etc., are participating in
discussions of water quality protection in the United States.
REFERENCES
Alabaster, J.S., and R. Lloyd. 1980. Water quality criteria for freshwater fish. Food and
Agriculture Organization of the United Nations, Butterworths, London.
APHA (American Public Health Association), American Water Works Association, and Water
Pollution Control Federation. 1985. Standard methods for the examination of water and
wastewater. 16th ed. American Public Health Association, Washington, D.C.
Ankley, G.T., D.M. Di Toro, D.J. Hansen, and W.J. Berry. 1996. Technical basis and proposal
for deriving sediment quality criteria for metals. Submitted to Environmental Toxicology
and Chemistry.
Brune, D.E., and J.R. Tomasso. (Eds.) 1991. Aquaculture and water quality. The World
Aquaculture Society, Louisiana State University, Baton Rouge, Louisiana.
Covello, V.T., and M.W. Merkhofer. 1993. Risk Assessment Methods: Approaches for
Assessing Health and Environmental Risks. Plenum Press, New York, New York.
Hollis, E.H., and R.E. Lennon. 1987. The Toxicity of 1,085 Chemicals to Fish. United States .
Fish and Wildlife Service, Kearneysville, West Virginia. EPA-560/6-87-002. Part 2..
MacPhee, C., and F.F. Cheng. 1989. Part 1. Lethal Effects of 964 Chemicals upon Steelhead
Trout and Bridgelip Sucker. Part 2. Lethal Effects of 2,014 Chemicals upon Sockeye
Salmon, Steelhead Trout and Threespine Stickleback. Department of Wildlife and
Fishery Resources, College of Forestry, Wildlife, and Range Sciences, University of
Idaho, Moscow, Idaho. EPA 560/6-89-001.
338
-------
Mayer, F.L. Jr., and M.R. Ellersieck. 1986. Manual of Acute Toxicity: Interpretation and Data
Base for 410 Chemicals and 66 Species of Freshwater Animals. Fish and Wildlife
Service, United States Department of the Interior, Columbia, Missouri. Resource
Publication No. 160.
McKee, J.E., and H.W. Wolf. (Eds.) 1963. Water Quality Criteria. 2nd ed. Resources Agency of
California, State Water Resources Control Board, Publication No. 3-A.
NAS (National Academy of Sciences), National Academy of Engineering. 1973. Water Quality
Criteria 1972. Report of the Committee on Water Quality Criteria,'Environmental
Studies Board. EPA/R3-73-033.
NRC (National Research Council). 1986. Ecological Knowledge and Environmental Problem-
solving: Concepts and Case Studies. National Research Council, National Academy
Press, Washington, D.C.
Rand, G.M., and S.R. Petrocelli. (Eds.) 1985. Fundamentals of Aquatic Toxicology: Methods
and Applications. Hemisphere Publishing Corporation, Washington, D.C.
Russo, R.C., and A. Pilli. 1984. AQUIRE: Aquatic Information Retrieval Toxicity Data Base.
Environmental Research Laboratory, Office of Research and Development, Duluth,
Minnesota. EPA-600/8-84-021. AQUIRE User Update. 1993. Scientific Outreach
Program, United States EPA, Environmental Research Laboratory, Duluth, Minnesota.
Russo, R;C., L.A. Mulkey, R. Carlson, and A. Fairbrother. 1992. Nonpoint Sources Research
Plan FY93-FY97. Environmental Research Laboratory, United States Environmental
Protection Agency, Athens, Georgia.
Stephan, C.E., D.I. Mount, D.J. Hansen, J.H. Gentile, G.A. Chapman, and W.A. Brungs. 1985.
Guidelines for Deriving Numerical National Water Quality Criteria for the Protection of
Aquatic Organisms and Their Uses. United States Environmental Protection Agency,
Office of Research and Development, Washington, DC.
Stern, P.C., and H.V. Fineberg (Eds.). 1996. Understanding Risk. Committee on Risk
Characterization, Commission on Behavioral and Social Sciences and Education,
National Research Council. National Academy Press, Washington, D.C.
Thurston, R.V.,R.C. Russo, C.M. Fetterolf Jr., T.A. Edsall, and Y.M. Barber Jr. (Eds.) 1979. A
Review of the EPA Red Book: Quality Criteria for Water. Water Quality Section,
American Fisheries Society, Bethesda, Maryland.
339
-------
U.S. EPA (United States Environmental Protection Agency). 1976. Quality Criteria for Water.
Office of Water Planning and Standards, United States Environmental Protection Agency,
Washington, D.C.
U.S. EPA (United States Environmental Protection Agency). 1979. Methods for Chemical
Analysis of Water and Wastes. Environmental Monitoring and Support Laboratory,
Office of Research and Development, Cincinnati, Ohio. Publication No. EPA-600/4-79-
020.
U.S. EPA (United States Environmental Protection Agency). 1987. Quality Criteria for Water
1986. Office of Water Regulations and Standards, Washington, DC. Publication No.
EPA/440/5-86-001.
U.S. EPA (United States Environmental Protection Agency). 1990. Office of Water
Environmental and Program Information Systems Compendium. Information Resources
Management: Tools for Making Water Program Decisions. Office of Water, Washington,
D.C. Publication No. EPA/9-90-002.
U.S. EPA (United States Environmental Protection Agency). 1991a. Technical Support
Document for Water Quality-Based Toxics Control. Office of Water, Washington, DC.
Publication No. EPA/505/2-90-001.
U.S. EPA (United States Environmental Protection Agency). 1991b. Methods for Measuring the
Acute Toxicity of Effluents to Aquatic Organisms. 4th ed. Office of Research and
Development, Cincinnati, Ohio. Publication No. EPA-600/4-90-027.
U.S. EPA (United States Environmental Protection Agency). 1991c. Short-term Methods for
Estimating the Chronic Toxicity of Effluents and Receiving Waters to Freshwater
Organisms. 3rd Ed. Office of Research and Development, Cincinnati, Ohio. Publication
No. EPA-600/4-91/002.
U.S. EPA (United States Environmental Protection Agency). 199Id. Short-term Methods for
Estimating the Chronic Toxicity of Effluents and Receiving Waters to Marine and
Estuarine Organisms. 2nd Ed. Office of Research and Development, Cincinnati, Ohio.
Publication No. EPA-600/4-91/003.
U.S. EPA (United States Environmental Protection Agency): 1992. Framework for Ecological
Risk Assessment. Risk Assessment Forum, Office of Research and Development,
Washington, D.C. Publication No. EPA/630/R-92-001.
340
-------
U.S. EPA (United States Environmental Protection Agency). 1993. Technical Basis for
Deriving Sediment Quality Criteria for Nonionic Organic Contaminants for the
Protection of Benthic Organisms by Using Equilibrium Partitioning. Office of Water,
Washington, DC. Publication No. EPA/822-R-93 -Oil.
U.S. EPA (United States Environmental Protection Agency). 1994a. Water Quality Standards
Handbook: Second Edition. Office of Water, Washington, DC. Publication No.
EPA/823-B-94-005a. :
U.S. EPA (United States Environmental Protection Agency). 1994b. Water Quality Standards
Handbook: Second Edition, Appendixes. Office of Water, Washington, DC. Publication
No. EPA/823-B-94-005b.
U.S. EPA (United States Environmental Protection Agency). 1996. Strategic Plan for the Office
of Research and Development. EPA/600/R-96/059. Office of Research and
Development, U.S. Environmental Protection Agency, Washington, D.C. 60 p.
Wood, E.M. 1987. The Toxicity of 3,400 Chemicals to Fish. United States Fish and Wildlife
Service, Kearneysville, West Virginia. EPA-560/6-87-002. Parti.
341
•&U.S. GOVERNMENT PRINTING OFFICE: 1997 - 549.001/60165
-------
-------