In situ and laboratory bioassays using Poecilia reticulata Peters, 1859 in the biomonitoring of an...
-
Upload
independent -
Category
Documents
-
view
0 -
download
0
Transcript of In situ and laboratory bioassays using Poecilia reticulata Peters, 1859 in the biomonitoring of an...
Chemosphere 73 (2008) 365–370
Contents lists available at ScienceDirect
Chemosphere
journal homepage: www.elsevier.com/ locate /chemosphere
Discriminating the pH toxicity to Poecilia reticulata Peters, 1859 in the Dunas
Lake (Camaçari, BA, Brazil)
0045-6535/$ - see front matter © 2008 Elsevier Ltd
doi:10.1016/j.chemosphere.2008.05.061
Cristiano V.M. Araújo *, Salomão J. Cohin-de-Pinho, Carla B.A. Chastinet, Jéssica S. Santos, Eduardo M. da Silva
Instituto de Biologia, Federal University of Bahia, Campus de Ondina, 40170-115 Salvador, BA, Brazil
a r t i c l e i n f o a b s t r a c t
Article history:
Received 26 September 2007
Received in revised form 28 April 2008
Accepted 27 May 2008
Available online 15 July 2008
Keywords:
Acidification
Acute Toxicity
Bioassays
Biomonitoring
Survival time
* Corresponding author. Tel.: +55 71 3263 6525; f
E-mail address: [email protected] (C.V.M
Toxic potential of the pH reduction to fingerlings of Poecilia reticulata, through acute toxicity bioassays,
as well as the influence of increased pH on the toxicity were assessed. Acid lake samples (Dunas Lake)
were collected during 19 months, and assessed with following treatments: water at local pH (±3.0) and
samples with modified pH to 3.5, 3.8, 4.0, 4.3, 4.6, 5.0, 5.5, 6.0, and 6.5. Culture water samples with pH
reduced to 3.0 were also assessed. Newborn P. reticulata were exposed during 96 h, and dead/immobile
organisms were counted at various time intervals during exposure (short intervals in the beginning and
long towards the end). Mean results of LT50 and confidence intervals from the Dunas Lake and control
water with reduced pH were 1.36 (±0.48) h, and 1.03 (±0.50) h, respectively, with no statistical difference.
Samples with increased pH showed a significant reduction in toxicity, with no toxicity detected at pH 6.0
and higher. Relationship between pH and lethal toxicity for fingerlings of P. reticulata demonstrated that
pH exerted a strong effect on the survival of this species at the Dunas Lake, explaining about 80% of the
toxicity observed.
© 2008 Elsevier Ltd. All rights reserved.
1. Introduction
The knowledge about ecosystem acidification goes back to the
mid-18th century (Gorham, 1998), but, over the past 30 years, sci-
entists have been working to understand the causes and results
of the acidification process, and how acidic deposition might influ-
ence different ecosystems (Driscoll et al., 2003). Acidification is
one of the main causes of environmental degradation in aquatic
ecosystems in the temperate regions, not only due to its toxicity,
but also due to its effect on the speciation, mobility, and bioavail-
ability of other toxicants (Lopes et al., 1999). For instance, many
regions are severely impacted by acidic rain, acid mine drainage,
volcanic activities or by other types of anthropogenic activities
(Geller et al., 1998; Ribeiro et al., 2002), resulting in decreasing pH
values, followed by increasing metal solubility (Geller et al., 1998).
As a whole, acidification in the Brazilian ecosystems is little doc-
umented, although the impact caused on the biodiversity is very
similar to other regions (Jesus, 1996). Rodhe et al. (1988) identi-
fied schematically areas in which acidification might represent a
potential threat on the basis of expected emissions, population
density, and soil sensitivity for south-eastern Brazil. More recently,
Moreira-Nordemann et al. (1988); Ometto et al. (2005) and Mar-
. All rights reserved.
ax: +55 71 3263 6511.
. Araújo).
tinelli et al. (2006) pointed out the risks involved in sugar-cane
burning to the acidification of the same region. An acidified lake,
and its ecotoxicity history, after rehabilitation, has been described
by da Silva et al. (1999a, 2000).
The toxicity of the trace elements, especially metals, does not
only depend on their concentration, but also on their bioavailabil-
ity, which is higher under reduced pH conditions (Wren and Ste-
phenson, 1991; Renoux et al., 2001). Therefore, the degradation
of fish populations is explained primarily by the direct combined
influence of low pH and aluminum ions, which causes biochemi-
cal and physiological disturbances (Moiseenko and Sharova, 2006).
Besides, pH influences the ionic regulation of aquatic species, the
rate of organic matter decomposition, and primary production
(Abel, 1996; Geller et al., 1998). According to Driscoll et al. (2003),
few fish species can survive in pH lower than 4.5. Indeed, most
of the acidified waters throughout the world are fish-free environ-
ments (Van Sickle et al., 1996; Nixdorf et al., 1998); however some
small South American fish can be found in acidic blackwater rivers
(such as Rio Negro, Amazon, Brazil) whose pH can be around 3.5
(Mounier et al., 1999; Matsuo and Val, 2002; Aride et al., 2007).
Gonzalez et al. (1998) suggested that exceptional acid tolerance is
a characteristic of fish that inhabit acidic ecosystem. A tool adopted
for preliminary environmental evaluation of acidification effects
in aquatic ecosystems is the use of survival time as an endpoint,
allowing the establishment of the species recolonization potential
(Ribeiro et al., 2002). It is reasonable to suppose that, increasing
366 C.V.M. Araújo et al. / Chemosphere 73 (2008) 365–370
pH values should improve water quality of acidic systems, leading
to previous or acceptable ecological conditions, and to reestablish-
ment of previous food webs, therefore it is extremely relevant to
assess biological and chemical conditions of acidified ecosystems
(Tipping et al., 2002; Driscoll et al., 2003).
Poecilia reticulata (guppy) is a tropical species, very abundant
in shallow waters of canals, rivers, lagoons and reservoirs in South
America, and largely employed in ecotoxicological tests. The results
of several studies showed that P. reticulata is a very sensitive spe-
cies when exposed to toxicants (Gallo et al., 1995; Miliou et al.,
1998; Polat et al., 2002; Yilmaz et al., 2004), and, therefore, ade-
quate for biomonitoring programs (Widianarko et al., 2000) and is
recommended as a standard ecotoxicological test organism (OECD,
1992; ABNT, 2002). The sensitivity of P. reticulata in comparison to
other fish species, as Brachydanio rerio (zebra fish), Cyprinus carpio
(common carp), Lepomis macrochirus (bluegill), Pimephales prom-
elas (fathead minnow), Salmo gairdneri (rainbow trout), was stud-
ied by Vittozzi and De Angelis (1991). Additionally, P. reticulata has
been used with success in the biomonitoring of an acidic lake for
several years (da Silva et al., 1999a), including in in situ bioassays
(Araújo et al., 2006). This species has been chosen as a test organ-
ism for the Dunas Lake survey, because it was a common species in
the lake prior to the contamination episode.
According to da Silva et al. (2000), in the late 1980s large quanti-
ties (ca. 34 t) of both industrial and domestic solid waste including
sulphur, iron, titanium dioxide and ilmenite residues were depos-
ited on the dunes adjacent to a small lake (Dunas Lake, Camaçari,
BA, Brazil). These wastes were leached out by rainwater percolat-
ing through the dunes, thus contaminating the ground- and sur-
face-water. There was a decrease in pH to 1.8 of the ground- and
surface-water and an increase in the concentrations of dissolved
iron and sulphate, causing the precipitation of humic acids, leading
to highly transparent waters with a concurrent disruption of the
biological communities in the lake (da Silva et al., 1999b, 2000). It
was not investigated if the toxicity of the water of the Dunas Lake
was due to the direct effect of low pH value, about 3.0, or due to
the ability of low pH to increase the bioavailability of toxic metals.
The methodologies to discriminate a toxic component using serial
dilutions or chelating agents can mask the ecotoxicity and alter
the characteristics of the samples (Lopes et al., 1999; Ribeiro et al.,
2002). In this context, this work aimed to discriminate, through
bioassays with altered pH, the potential toxicity of the acidity to
P. reticulata; identifying possible reduction in the toxicity due to
increased pH; and to determine threshold survival for P. reticulata
in relation to Dunas Lake pH.
Fig. 1. Location o
2. Materials and methods
2.1. Study site
The acidic lake (Dunas Lake) is located in Camaçari (BA, Brazil)
(Fig. 1) between geographic coordinates 12°489090 to 12°48912.30 S and 38°139090 to 38°139140 W, lying within a depression, forming
a narrow and shallow body of freshwater between dunes along the
Atlantic Ocean (da Silva et al., 2000). After contamination, a reha-
bilitation program was carried out (1992–1993) to recover ground-
and surface water quality and reduce contamination (Gomes, 1994;
da Silva et al., 1999b). Initially, the residues were partially removed
and the contaminated dune was sealed with impermeable layers
of clay and topsoil (hydraulically encapsulated). An additional
action was to pump the groundwater to reduce the contaminated
plume (Gomes, 1994; da Silva et al., 1999b, 2000). The results of
the established biomonitoring program, including bioassays with
guppy fingerlings (P. reticulata), were described by da Silva et al.
(1999a, 2000) and Araújo et al. (2006).
2.2. Sampling
Monthly samplings (n = 19) of the Dunas Lake were carried out
from March 2003 to November 2004, except for August and Octo-
ber 2004. Samples were transported to the laboratory and kept at
4.0 ± 1.0 °C until the next day, when the pH was measured. Imme-
diately the samples were distributed in 5 l beakers to alter the pH.
Samples were treated with NaOH 1 M to raise their pH, with the fol-
lowing treatments: original Dunas Lake pH (pH not altered – about
3.0), 3.5, 3.8, 4.0, 4.3, 4.6, 5.0, 5.5, 6.0, and 6.5. These treatments
aimed to assess the toxicity reduction after increasing pH. To assess
the toxic effect of pH, the control water pH (i.e. dechlorinated tap
water) was reduced to the same pH value of the Dunas Lake, using
H2SO4 1 M, because this was the main acid in the acidification
process (da Silva et al., 1999b). A possible difference between the
Dunas Lake water in natura and the control sample with reduced
pH may be attributed to other factors potentially toxic present in
the Dunas Lake that could interfere in the toxicity which may be
influenced by pH.
2.3. Physical–chemical analysis
Dissolved oxygen content (WTW, Inolab Oxi Level 2), water hard-
ness (APHA, 1998), pH (Digi-Sense, Cole Parmer) and conductivity
(Hanna HI 9033) of the samples were evaluated at the beginning
f the study site.
C.V.M. Araújo et al. / Chemosphere 73 (2008) 365–370 367
and end of each experiment. Mean pH was computed by calculat-
ing the average concentration of hydrogen ions and then calculat-
ing the respective pH. Bioassay results were validated when pH
variation was no higher than 0.5. All measurements were carried
out at 25 ± 1 °C.
Physical–chemical characterization of the Dunas Lake and
control water (dechlorinated tap water) was carried out and it is
summarized in Table 1. Metals were analyzed using inductively cou-
pled plasma atomic emission spectrometry (ICP-AES) and Cr(VI),
nitrate, nitrite, phosphate and sulphate were analyzed by ion chro-
matography (APHA, 1998). Data of control water was provided by
EMBASA (Water and Sanitation Company of the State of Bahia) that
is responsible for drinking water quality control.
2.4. Test organism and acclimation
Guppy (P. reticulata) fingerlings, 10–15 days old (average length
of 1.0 ± 0.2 cm) were obtained from a local aquarist who kept the
fish under standardised conditions. In spite of the recommenda-
tion by OECD (1992) and ABNT (2002) to use adult P. reticulata
in acute and static bioassays, fingerlings in early life stages were
employed since the early life stages of fish are in many cases the
most sensitive to adverse conditions (Farag et al., 1993; Petersen
and Kristensen, 1998), especially under pH changes (Vuorinen
et al., 2003).
The organisms were transported to the laboratory in glass flasks
with sufficient air, and acclimatized in 20 l glass aquaria containing
dechlorinated tap water for 24 ± 2 h prior to the experiments. The
fish were not fed during this period or during the experiment. Accli-
mation and bioassays were performed at 25.0 ± 1.0 °C, in constant
temperature rooms with a photoperiod of 12:12 h (light and dark).
Static acute toxicity tests were carried out according to OECD
(1992) and ABNT (2002). Dechlorinated tap water, in which the
fish had been reared, was used as a control.
2.5. Bioassays
Glass aquaria of 1.2 l capacity, containing 900–1000 ml of water
sample, were used as test vessels. Five replicates of each treat-
ment, containing eight to ten fish in each replicate were tested,
totalling 40–50 organisms exposed to each treatment. For each
treatment 19 bioassays (March 2003 to November 2004, except
for August and October 2004) were carried out. The organisms
were randomly distributed in the test vessels and fish mortality
was checked at reduced time intervals, 10 min in the first 3 h of
the tests, 30 min of the third until the tenth hour, and at longer
time intervals in the following hours. Time was the independent
variable, and samples were not diluted. Dead fish were counted
and removed immediately to avoid adverse effects due to decom-
Table 1
Physical and chemical characterisation of the Dunas Lake and dechlorinated tap
water
Parameters Dunas
Lake
Control
water
Parameters Dunas
Lake
Control
water
SO4 (mg l¡1) 113.7 – Total-Fe (mg l¡1) 0.65 0.26
Cr(VI) (mg l¡1) <0.01 <0.01 Dissolved-Fe (mg l¡1) 0.64 –
NO2 (mg l¡1) <0.1 <0.1 Total-K (mg l¡1) 0.78 –
NO3 (mg l¡1) <0.1 <0.1 Total-Mg (mg l¡1) 2.6 –
Phosphate
(mg PO4 l¡1)
<1.0 <1.0 Total-Mn (mg l¡1) 0.5 <0.005
Total-Al (mg l¡1) 1.5 0.16 Total-Na (mg l¡1) 6.7 –
Total-Ca (mg l¡1) 19.0 – Total-Ni (mg l¡1) <0.016 <0.016
Total-Cd (mg l¡1) <0.005 <0.005 Total-P (mg l¡1) <0.33 <0.33
Total-Co (mg l¡1) <0.01 <0.01 Total-Pb (mg l¡1) <0.14 <0.14
Total-Cr (mg l¡1) <0.01 <0.01 Total-Ti (mg l¡1) 0.05 –
Total-Cu (mg l¡1) 0.014 <0.005 Total-Zn (mg l¡1) <0.01 <0.01
position of the organisms. Behavioural changes such as loss of equi-
librium and swimming disorders were recorded. Organisms were
only considered dead when operculum and gill movements had
ceased and there was no swimming response after stimulation
with a plastic Pasteur pipette.
2.6. Data analysis
Median lethal time (LT50: time at which 50% of the test animals
were dead) of the treatments was determined by Probit Analysis,
considering the 5 replicates. Mean LT50 values were based on the
19 monthly values. To compare mean values the analysis of vari-
ance (one-way ANOVA) was used followed by Tukey multiple com-
parison test (Zar, 1996). Differences were considered significant at
P < 0.05 (Zar, 1996). All values are given as means ± standard devia-
tion. A logarithmic regression model based on pH values against
raw data of the Dunas Lake samples LT50 was plotted to evaluate
the proportion of the median lethal time that is predictable from
pH values.
3. Results and discussion
Survival in all control assays was greater than 90%, thus accom-
plishing the recommendations from OECD (1992) and ABNT (2002).
Dissolved oxygen concentrations were always above 6.5 mg l¡1 in
the Dunas Lake treatments and above 8.0 mg l¡1 in the control. The
Dunas Lake and control water presented a pH mean value of 3.08
(±0.4), and 7.34 (±0.15), respectively (measurement at 25 ± 1 °C).
Conductivity and hardness total values of all samples are sum-
marized in Table 2. There was no significant difference in conduc-
tivity and hardness mean values of Dunas Lake treatments after
changed pH. Regarding the Dunas Lake treatments, it could be
observed that the highest mean conductivity value measured
in the in natura sample was 343.80 lS cm¡1, and the lowest was
246.50 lS cm¡1 in the samples with pH 5.5 (Table 2). Therefore,
with a pH increase there was a correspondent decreasing in con-
ductivity. In the control samples, with no pH change, the mean con-
ductivity value was 418.55 lS cm¡1, corroborating that this param-
eter was not responsible for mortality during the experiments.
Control samples with reduced pH were significantly different from
all other samples (F11,216 = 48.179; P < 0.05) (Table 2). Mean hardness
of Dunas Lake treatments showed significant difference in relation
to the values registered in the control samples and control with
reduced pH (F11,216 = 26.371; P < 0.05) (Table 2). The mean value of
water hardness in the control samples was 126.25 mg CaCO3 l¡1,
while in the Dunas Lake was 79.00 mg CaCO3 l¡1 (Table 2). The
influence of the hardness on the bioavailability of some chemical
Table 2
Mean values and confidence intervals (95%) of conductivity and hardness of Dunas
Lake samples and control at the start of the experiment
Samples Parameters
Conductivity (lS cm¡1) Hardness (mg CaCO3 l¡1)
Control 418.55 (±67.32) A 126.25 (±20.49) A
Control (reduced pH
to Dunas Lake)
548.86 (±66.14) B 129.56 (±22.95) A
Dunas Lake (in natura) 343.80 (±12.14) AC 79.00 (±4.07) B
Dunas Lake (pH 3.5) 276.80 (±41.28) C 76.87 (±9.47) B
Dunas Lake (pH 3.8) 265.20 (±24.29) C 76.88 (±12.20) B
Dunas Lake (pH 4.0) 289.09 (±23.95) C 77.05 (±6.17) B
Dunas Lake (pH 4.3) 263.57 (±16.92) C 79.38 (±6.33) B
Dunas Lake (pH 4.6) 249.00 (±24.53) C 79.30 (±10.43) B
Dunas Lake (pH 5.0) 251.00 (±29.39) C 75.30 (±7.18) B
Dunas Lake (pH 5.5) 246.50 (±43.17) C 81.77 (±17.69) B
Dunas Lake (pH 6.0) 258.00 (±8.52) C 75.10 (±7.38) B
Dunas Lake (pH 6.5) 251.50 (±15.23) C 79.90 (±7.64) B
368 C.V.M. Araújo et al. / Chemosphere 73 (2008) 365–370
compounds and the toxicity is unarguable (Akkanen and Kukko-
nen, 2001), especially due to its capacity to modulate the pH effect
(Ribeiro et al., 2002). In our study, however, water hardness values
for the Dunas Lake did not change with pH increase, therefore the
changed pH of samples with NaOH did not show any influence in
this variable.
Mean survival time of the organisms in the Dunas Lake with
no altered pH was 1.36 (±0.48) h (Fig. 2). In the study of da Silva
et al. (1999a) with P. reticulata, the LT50 of Dunas Lake was <1 h
(pH lower than 3.0), with the majority of the values lower than
30 min. The results of the present study show a better toxicologi-
cal condition of the lake, in comparison to the former study, due
to increased pH, but ecologically there has been no improvement
(i.e. the tested organisms are still not able to survive in the waters
of the Dunas Lake). In relation to other treatments of the Dunas
Lake there was a significant statistical difference, except for sam-
ples with pH 3.5 (Fig. 2). With gradual pH increase there was a
toxicity reduction, resulting in higher LT50, although no significant
statistical difference was noted between treatments with pH 4.3 to
pH 5.5. The pH variation (0.5 units) for the bioassays may explain
some of the lack of significance between these pH groups. The high-
est LT50 mean was registered in the samples with pH 5.5 (74.30 h).
Samples with pH 6.0 and 6.5 did not show any mortality during
96 h, with results being expressed as >96 h. The relative increase
in survival time due to the pH increase can be considered a toxic-
ity measure that allows only the prediction of the ecotoxicological
effects to guppy fingerlings and to assess the possibility of recoloni-
zation of this species during the rehabilitation of the Dunas Lake.
On the other hand, sublethal levels of acidity can be detrimental
affecting the weight and length of the fish, oocyte development
earlier during vitellogenesis and delay of spermatogenesis (Vuori-
nen et al., 2003). When fish are subjected to acid stress, blood
pH decreases possibly as the result of flux of H+ ions across gill
membranes into the blood and an acidemia (lowering of the pH
of the blood) may occur, decreasing the capacity of hemoglobin
to transport oxygen (Fromm, 1980). Moreover, the ability of fish
uptake oxygen decreases with a decrease in water pH due to gill
mucus coagulation, and this may represent the primary cause of
death of fishes (Fromm, 1980). In a study with guppy, Dunson et
al. (1977) found mortality in all organisms after 11 days at pH 4.75,
and it was associated with loss of large amounts of body sodium.
Fig. 2. LT50 mean values to P. reticulata, confidence intervals (95%) of the Dunas Lake
(DL) samples and control with reduced pH (CR). Mean values followed by the same
letter in the same column do not differ significantly by Tukey’s test (F8,162 = 109.89;
P < 0.0001).
Undoubtedly, the toxic action of increased ambient H+ ions concen-
tration on freshwater fish is variable (Fromm, 1980).
Dunas Lake in natura samples were compared to the control sam-
ples with reduced pH to assess whether the observed effect (LT50)
by action of the pH were similar to the observed effect in the Dunas
Lake samples. Differences in survival time between these two sam-
ples could show an additional toxicity of Dunas Lake, unexplained
by the pH, that will allow to discriminate the toxicity due to pH
and due to other elements (Lopes et al., 1999). However, there was
no significant statistical difference between the LT50 mean values
of these two samples. Dunas Lake presented a mean LT50 of 1.36
(±0.48) h, while control water with reduced pH showed a mean
LT50 of 1.03 (±0.50) (Fig. 2). According to these results, it is possible
to attest that pH is the main responsible for toxicity results from
Dunas Lake and that the fish survival will increase with pH value
increases to 6.5.
The exponential regression shows that the model is highly
significant as evident from the P-value (P < 0.0001) and from the
coefficient of determination (r2 = 0.80), indicating that pH is able
to explain ca. 80% of the toxicity observed. Similar data was also
obtained when toxicity tests results using natural lake water,
between 2005 and 2007 (P < 0.0001; r2 = 0.83), were integrated in
the original data (Fig. 3). In this model, as well as in Fig. 2, there is
a high variation of the LT50 results of the toxic moderately samples
(e.g.: pH 3.8–5.0). According to Salvadó et al. (1995), the relation
between some parameters, such as pH, and a species does not fol-
low a linear model, because in general there is an optimal range,
in which below and above of this range the species can be affected
with density reduction or disappearance. For instance, the range of
pH considered safer to fish is between 5 and 9, and extreme values
as 3.0–3.5 were considered unlikely to survive of any fish (EIFAC,
1969). According to Fromm (1980), it appears that the pH depres-
sion no-effect level for successful reproduction of fish is around
6.5; however, an avoidance behavior could be observed at a pH of
5.5. Thus, the highest pH value tested was 6.5, because this is the
maximum pH value registered in the water bodies of this region
(da Silva et al., 1999b, 2000), moreover, values above the neutral
pH can be toxic due to the alkalinity (Aride et al., 2007). When
the data generated in this work was plotted together with P. reticu-
lata toxicity data (monthly results, between the years 2005–2007),
collected from natural Dunes Lake water (Chastinet, unpublished
results) (black circle in Fig. 3), a good agreement of the results was
pH values
2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
LT50
(h)
0
12
24
36
48
60
72
84
96
Black and white circles (n=195)
+
=
0.282
3.898x
1
81.12y
e
+
=
0.243
3.913x
1
79.73y
e
0.832 =r
0.802 =r
White circles (n=171)
Fig. 3. Relationship between pH and lethal toxicity for fingerlings of P. reticulata.
White circles and continuous line represent results from the Dunas Lake modified
pH (this work). Black circles represent the same kind of test, using natural Dunas
Lake water between years 2005–2007 (Chastinet, unpublished results). The dotted
line represents the integration of black and white circles, as data from the two stud-
ies did not differ statistically (Mann-Whitney test: P < 0.05; U9 = 18400).
C.V.M. Araújo et al. / Chemosphere 73 (2008) 365–370 369
obtained, confirming the predictability of the results found when
pH changes were induced in the laboratory with the same water.
Moreover, the Mann-Whitney test (P < 0.05; U9 = 18400), indicated
that the two models did not differ statistically with respect to fish
toxicity and that pH is, indeed, the major factor responsible for
toxicity. The higher variation observed in the LT50 results of the
samples with pH 4.0–4.6 may be due to the moderately toxic con-
ditions. At moderately impacted sites, where ambient toxicity is
relatively low, it will be more difficult to establish a relationship
between toxicity and response (Del Valls et al., 1999). Treatments
with high toxicity (pH lower than 4.0) may exert their effects more
severely. On the other hand, from pH 4.0 there was higher proba-
bility of survival of the organisms due to reduced toxicity. At the
same time, in the treatments with pH values from 5.5, the favour-
able conditions lead to a more stable survival. Notwithstanding,
it is important to take into account that under reduced pH, metal
bioavailability is increased together with toxicity (Wren and Ste-
phenson, 1991; Renoux et al., 2001); however, de Santana (2004),
working with water, plants and sediments from the Dunas Lake,
noticed that metal availability was low, and suggested that this
was due to a possible metal removal to the groundwater due to
its high solubility under low pH. On the other hand, the toxicity
that P. reticulata underwent was mainly controlled by pH, as the
evidence indicated.
Several behavioural changes were observed before dead
occurred: (i) the fish, on the whole, tended to gather at the sur-
face, motionless, with some respiratory difficulties; (ii) there was
some colour change in the abdominal area, which became more
whitish; and (iii) there was also a loss of equilibrium, shown by
spiral swimming behaviour and loss of vertical orientation in the
water. In general, these changes were similar to those observed by
Polat et al. (2002); Viran et al. (2003), and Yilmaz et al. (2004) in
other studies.
4. Conclusions
Dunas Lake pH showed high toxicity potential to P. reticulata
fingerlings and the assays were able to detect a change in water
quality, as well as to discriminate the pH toxic action. The discrim-
inating model here presented showed high capacity of prediction,
explaining about 80% of the Dunas Lake toxicity to P. reticulata fin-
gerlings, moreover, data replicates the natural toxicity of the eco-
system.
Although of the effects already well-known of the total hard-
ness and conductivity on the toxicity, especially in acidic aquatic
environments, these parameters do not seem to influence the toxic-
ity to P. reticulata in the Dunas Lake.
According to the historical data collected at the Dunas Lake
(da Silva et al., 1999a), an increase in survival time of the test organ-
isms has been recorded, but this does not imply a toxicity reduc-
tion of this ecosystem in relation to the pH, as the fish continue to
die and are not able to survive in the actual conditions. The results
here presented attest that the survival time of P. reticulata finger-
lings increased with the increase of the pH, however, only from pH
6.0 the organisms were capable to survival for more than 96 h.
In spite of the results reported here, the Dunas Lake has been
showing signs of rehabilitation, albeit at a slow pace, and the colo-
nisation of some micro-habitats by several biological communities
certify this (Reis, 2004). It is reasonable to suppose that the inter-
nal processes (i.e. photosynthesis and respiration), may change the
acidic feature of the water and enable other life forms to colonise
the lake. Several rehabilitation measures have been suggested for
acidic lakes in the literature (Fredmann, 1989; George and Dav-
ison, 1998; Wendt-Potthoff and Neu, 1998), but in this study no
intervention has been proposed, so the natural succession could be
assessed and monitored, thus reflecting the actual rehabilitation
capacity of the lake. The recovery of P. reticulata population in the
Dunas Lake will only attain a sustainable level, when pH values
reach neutral pH values, about 6.0.
Acknowledgements
The authors are grateful to Lyondell Inc. (Brazil) and the Brazil-
ian Research Council (CNPq) for supporting this study (Grant #
620151/2004-8). C.V.M. Araújo and C.B.A. Chastinet received schol-
arships from the Brazilian Coordination of Improvement of Person-
nel of Superior Level (CAPES) and CNPq, respectively. The authors
are also thankful to R. Ribeiro, A.L. Fonseca, R. Dacosta and to two
reviewers for improving the original manuscript.
References
Abel, P.D., 1996. Water Pollution Biology, second ed. Taylor & Francis Ltd, London, p. 286.
ABNT – Associação Brasileira de Normas Técnicas, 2002. Água – Ensaio de toxicid-ade aguda com peixes – Parte I – Sistema estático (NBR 12714). ABNT, Rio de Janeiro.
Akkanen, J., Kukkonen, V.K., 2001. Effects of water hardness and dissolved organic material on bioavailability of selected organics chemicals. Environ. Toxicol. Chem. 20, 2303–2308.
APHA, 1998. Standard Methods for the Examination of Water and Wastewater. 20 Ed. American Public Health Association, Port City Press, Baltimore, Maryland.
Araújo, C.V.M., Cohin-de-Pinho, S.J., Santos, J.S., Delgado, F., Santana, L.C.S., Chasti-net, C.B.A., da Silva, E.M., 2006. In situ and laboratory bioassays using Poecilia reticulata Peters, 1859 in the biomonitoring of an acidic lake at Camaçari, BA, Brazil. Chemosphere 65, 599–603.
Aride, P.H.R., Roubach, R., Val, A.L., 2007. Tolerance response of tambaqui Colossoma macropomum (Cuvier) to water pH. Aquat. Res. 38 (6), 588–594.
da Silva, E.M., Navarro, M.F.T., Barros, A.F., Mota, M.F.V., Chastinet, C.B.A., 1999a. The utilisation of Poecilia reticulata as a biomonitor in the environmental recovery of an aquatic ecosystem. Ecotoxicol. Environ. Restor. 2, 51–55.
da Silva, E.M., Barros, A.F., Navarro, M.F.T., Mota, M.F.V., Cotsifis, P., Chastinet, C.B.A., 1999b. Rehabilitation following industrial contamination: Jauá Lake, a coastal wetland in Camaçari, Bahia, Brazil. In: Streever, W. (Ed.), An International Per-spective on Wetland Rehabilitation. Kluver Academic Press, Netherlands, pp. 197–203.
da Silva, E.M., Navarro, M.F.T., Barros, A.F., Mota, M.F.V., Chastinet, C.B.A., 2000. Met-als in the sediments of Jauá Lake (Camaçari, Bahia, Brazil) following an episode of industrial contamination. Aquat. Ecosys. Health Manage. 3, 509–514.
Del Valls, T.A., Forja, J.M., Gómez-Parra, A., 1999. An integrative study of environ-mental quality of sediments from the Bay Cadiz. Ecotoxicol. Environ. Restor. 2, 26–33.
de Santana, L.C.S., 2004. Avaliação da reabilitação e da ecotoxicidade atual da Lagoa das Dunas, Camaçari-Bahia, Brasil. MSc. Dissertation, Federal University of Bahia, p. 104.
Driscoll, C.T., Driscoll, K.M., Mitchell, M.J., Raynal, D.J., 2003. Effects of acidic depo-sition on forest and aquatic ecosystems in New York State. Environ. Pollut. 123, 327–336.
Dunson, W.A., Swarts, F., Silvestri, M., 1977. Exceptional tolerance to low pH of some tropical blackwater fish. J. Exp. Zool. 201, 157–162.
EIFAC, 1969. Water Quality Criteria for European freshwater fish – extreme pH val-ues and inland fisheries. Water Research Pergamon Press 3, 593-611.
Farag, A.M., Woodward, D.F., Little, E.E., 1993. The effects of low pH and elevated alu-minum on yellowstone cutthroat trout (Oncorhynchus clarki bouvieri). Environ. Toxicol. Chem. 12, 719–731.
Fredmann, B., 1989. Environmental Ecology: The Impacts of Pollution and Other Stresses on Ecosystem Structure and Function. Academic Press, San Diego, p. 424.
Fromm, P.O., 1980. A review of some physiological and toxicological responses of freshwater fish to acid stress. Environ. Biol. Fish. 5 (1), 79–93.
Gallo, D., Merendino, A., Keizer, J., Vittozzi, L., 1995. Acute toxicity of two carba-mates to the guppy (Poecilia reticulata) and zebrafish (Brachydanio rerio). Sci. Total Environ. 171, 131–136.
Geller, W., Klapper, H., Schultze, M., 1998. Natural and anthropogenic sulfuric acid-ification of lakes. In: Geller, W., Klapper, H., Salomons, W. (Eds.), Acidic Mine Lakes: Acid Mine Drainage, Limnology and Reclamation. Springer, Berlin, pp. 3–14.
George, D.G., Davison, W., 1998. Managing the pH of an acid lake by adding phos-phate fertilizer. In: Geller, W., Klapper, H., Salomons, W. (Eds.), Acidic Mine Lakes: Acid Mine Drainage, Limnology and Reclamation. Springer, Berlin, pp. 366–384.
Gomes, D.C., 1994. Poluição de aquífero costeiro de Arembepe-BA por ácido sulfú-rico e compostos inorgânicos oriundos da produção de dióxido de titânio. MSc. Dissertation, Universidade de São Paulo, 116 pp.
Gonzalez, R.J., Wood, C.M., Wilson, R.W., Patrick, M.L., Bergman, H.L., Narahara, A., Val, A.L., 1998. Effects of water pH and calcium concentration on ion balance in fish of the Rio Negro, Amazon. Physiol. Zool. 71, 15–22.
370 C.V.M. Araújo et al. / Chemosphere 73 (2008) 365–370
Gorham, E., 1998. Acid deposition and its ecological effects: a brief history of research. Environ. Sci. Policy 1, 153–166.
Jesus, E.F.R., 1996. A importância do estudo das chuvas ácidas no contexto da aborda-gem climatológica. Sitientibus 14, 143–153.
Lopes, I., Gonçalves, F., Soares, A.M.V.M., Ribeiro, R., 1999. Discriminating the eco-toxicity due to metals and to low pH in acid mine drainage. Ecotoxicol. Environ. Saf. 44, 207–214.
Miliou, H., Zaboukas, N., Moraitou-Apostopoulou, M., 1998. Biochemical composi-tion, growth, and survival of the guppy, Poecilia reticulata, during chronic suble-thal exposure to cadmium. Arch. Environ. Contam. Toxicol. 35, 58–63.
Martinelli, L.A., Howarth, R.W., Cuevas, E., Filoso, S., Austin, A., Lara, L.B.S.L., Medina, E., 2006. Sources of reactive nitrogen affecting ecosystems in Latin America and the Caribbean: current trends and future perspectives. Biogeochemistry 79, 3–24.
Matsuo, A.Y., Val, A.L., 2002. Low pH and calcium effects on net Na+ and K+ fluxes in two catfish species from the Amazon River (Corydoras: Callichthyidae). Braz. J. Med. Biol. Res. 35, 361–367.
Moiseenko, T.I., Sharova, O.N., 2006. Physiological mechanisms of degradation of fish populations in acidified water bodies. Russ. J. Ecol. 37, 257–263.
Moreira-Nordemann, L.M., Forti, M.C., Di Lascio, V.L., do Espírito Santo, C.M., Dan-elon, O.M., 1988. Acidification in Southeastern Brazil. In: Rodhe, H., Herrera, R. (Eds.), Acidification in Tropical Countries. John Wiley & Sons, Chichester, pp. 257–296.
Mounier, S., Braucher, R., Benaïm, J.Y., 1999. Differentiation of organic matter’s prop-
erties of the Rio Negro basin by cross-flow ultra-filtration and UV-spectrofluo-rescence. Water Res. 33 (10), 2363–2373.
Nixdorf, B., Wollmann, K., Deneke, R., 1998. Ecological potentials for planktonic
development and food web interactions in extremely acidic mining lakes in Lusatia. In: Geller, W., Klapper, H., Salomons, W. (Eds.), Acidic Mine Lakes: Acid Mine Drainage, Limnology and Reclamation. Springer, Berlin, pp. 147–167.
OECD – Organisation Economic Co-operation and Development, 1992. Guideline for testing of chemicals. Fish Acute Toxicity Test. OECD Guideline for Testing of Chemicals, Paris.
Ometto, A.R., Mangabeira, J.A.C., Hott, M.C., 2005. Mapeamento de potenciais de impactos ambientais da queima de cana-de-açúcar no Brasil. In: Proceedings of the 12th Simpósio Brasileiro de Sensoriamento Remoto, Goiânia, Brazil, 16-21 April 2005, INPE, pp. 2297–2299.
Petersen, G., Kristensen, P., 1998. Bioaccumulation of lipophilic substances in fish early life stage. Environ. Toxicol. Chem. 17, 1385–1395.
Polat, H., Erkoç, F.Ü., Viran, R., Koçak, O., 2002. Investigation of acute toxicity of beta-cypermethrin on guppies Poecilia reticulata. Chemosphere 49, 39–44.
Reis, I.S., 2004. Monitoramento da evolução das comunidades biológicas da Lagoa das Dunas após episódios de contaminação ácida, Camaçari – Bahia – Brasil. MSc. Dissertation, Universidade Federal da Bahia, 73 p.
Renoux, A.Y., Tyagi, R.D., Samson, R., 2001. Assessment of toxicity reduction after metal removal in bioleached sewage sludge. Water Res. 35, 1415–1424.
Ribeiro, R., Lopes, I., Pereira, A.M.M., Gonçalves, F., Soares, A.M.V.M., 2002. Survival time of Ceriodaphia dubia in acid waters with metal contamination. Bull. Envi-ron. Contam. Toxicol. 64, 130–136.
Rodhe, H., Cowling, E., Galbally, I., Galloway, J., Herrera, R., 1988. Acidification and regional air pollution in the tropics. In: Rodhe, H., Herrera, R. (Eds.), Acidificat-ion in Tropical Countries. SCOPE 36. John Wiley and Sons, Chichester, pp. 3–39.
Salvadó, H., Gracia, M.P., Amigó, J.M., 1995. Capability of ciliated protozoa as indica-tors of effluent quality in activated sludge plants. Water Res. 29, 1041–1050.
Tipping, E., Bass, J.A.B., Hardie, D., Haworth, E.Y., Hurley, M.A., Wills, G., 2002. Bio-logical responses to the reversal of acidification in surface waters of the English Lake District. Environ. Pollut. 116, 137–146.
Van Sickle, J., Baker, J.P., Simonin, H.A., Baldigo, B.P., Kretser, W.A., Sharpe, W.E., 1996. Episodic acidification of small streams in the northeastern United States: fish mortality in field bioassays. Ecol. Appl. 6, 408–421.
Viran, R., Erkoç, F.Ü., Polat, H., Koçak, O., 2003. Investigation of acute toxicity of delta-methrin on guppies (Poecilia reticulata). Ecotoxicol. Environ. Saf. 55, 82–85.
Vittozzi, L., De Angelis, G., 1991. A critical review of comparative acute toxicity data on freshwater fish. Aquat. Toxicol. 19, 167–204.
Vuorinen, P.J., Keinänen, M., Peuranen, S., Tigerstedt, C., 2003. Reproduction, blood
and plasma parameters and gill histology of vendace (Coregonus albula L.) in long-therm exposure to acidity and aluminum. Ecotoxicol. Environ. Saf. 54, 255–276.
Wendt-Potthoff, K., Neu, T.R., 1998. Microbial processes for potential in situ reme-diation of acidic lakes. In: Geller, W., Klapper, H., Salomons, W. (Eds.), Acidic Mine Lakes: Acid Mine Drainage, Limnology and Reclamation. Springer, Berlin, pp. 269–284.
Widianarko, B., Van Gestel, C.A.M., Verwij, R.A., Van Straalen, N.M., 2000. Associa-tions between trace metals in sediment, water, and guppy, Poecilia reticulata (Peters), from urban streams of Semarang, Indonesia. Ecotoxicol. Environ. Saf. 46, 101–107.
Wren, C.D., Stephenson, G.L., 1991. The effect of acidification on the accumulation and toxicity of metals to freshwater invertebrates. Environ. Pollut. 71, 205–241.
Yilmaz, M., Gül, A., Karaköse, E., 2004. Investigation of acute toxicity and effect of cadmium chloride (CdCl2 · H2O) metal salt on behavior of the guppy (Poecilia reticulata). Chemosphere 56, 375–380.
Zar, J.H., 1996. Biostatistical Analysis, third ed. Prentice-Hall Int, Upper Saddle River,
N.J., USA.