Reassessment of the suitable range of water pH for culture of Nile tilapia Oreochromis niloticus L . in eutrophic water

The present work aimed at reassessing the suitable range of water pH for culture of Nile tilapia, Oreochromis niloticus L. juveniles in eutrophic water. Two hundred and forty tilapia juveniles (1.37 ± 0.04 g) were stocked in twenty 250-L polyethylene tanks (12 fish per tank) for eight weeks. In the control tanks, the pH of water was not adjusted at any time, varying freely over the entire study. In the slight acidification treatment, the culture water was acidified daily to reach a water pH between 5.5 and 6.5. In the moderate acidification treatment, there were daily applications of HCl solution to reach a water pH between 4.5 and 5.5. In the alkalinization treatment, tanks received daily applications of Na2CO3 and NaOH to achieve a water pH between 8.5 and 9.5. Acidification of water, regardless the degree, i.e., slight or moderate, was not able to significantly affect final body weight, specific growth rate and yield of fish. It can be concluded that the acidification of water up to pH 5.5 has no negative influence on growth of Nile tilapia fingerlings in eutrophic tanks. Accordingly, the suitable range of water pH for rearing Nile tilapia should be set at 5.5 – 9.0.


Introduction
The pH of water can significantly affect the physiology of aquatic animals.The degree of acidity and basicity of water can stress and disrupt the normal growth of farmed fish and shrimp.Mechanisms of ionic regulation in fish are activated by variations in water pH, seeking the homeostasis and health maintenance.Acid-base disturbances in blood and body fluids can alter important metabolic parameters in fish, such as the concentrations of glycose, glycogen and lactate (Bolner, Copatti, Rosso, Loro, & Baldisserotto, 2014;Garcia, Gutiérrez-Espinosa, Wásquez-Torres, & Baldisserotto, 2014).
Therefore, it is possible to have exceptions to that general rule of suitability of water pH.
According to El-Sherif and El-Feky (2009), the optimal water pH for the culture of Nile tilapia, Oreochromis niloticus L., is 7 -8.These authors, however, have reared tilapia fingerlings in oligotrophic clear waters, in tanks with 100% daily water exchange.Consequently, there were low concentrations of toxic metabolites in the water, such as ammonia and H 2 S.However, Nile tilapia juveniles have grown surprisingly well on acidic organic-matter rich waters (pH < 6) in previous studies carried out in our laboratory (Nobre, Lima, & Magalhães, 2014;Rebouças, Lima, & Cavalcante, 2015;Silva, Santos Lima, Vale, & Carmo, 2013).Colt, Momoda, Chitwood, Fornshell, and Schreck (2011) have also found that O. niloticus could be transferred from pH 6 -7 to as low as pH 4.2 without problems.Due to the discrepancy between these results and those reported by El-Sherif and El-Feky (2009).
The present work was carried out aiming to reassess the suitable range of water pH for culture of Nile tilapia juveniles in eutrophic waters.

Material and methods
Masculinized Nile tilapia juveniles with body weight between 1 -2 g were obtained from a regional producer and transported to the laboratory facilities, where they were maintained for four days in one 1,000-L tank for acclimation.In this phase, the animals were fed on four times daily at 0800, 1100, 1400 and 1700 with a commercial diet for omnivorous tropical fish containing 43.4% crude protein at 10% body weight daily.
At the onset of the experiment, two hundred and forty tilapia juveniles (1.37 ± 0.04 g) were stocked in twenty 250-L polyethylene tanks (12 fish per tank) for eight weeks.Fish were fed daily with appropriate commercial diets at 1000, 1300, 1500 and 1700, on feeding rates that ranged from 8.9% (initial) to 3.9% (final) body weight.No mechanical aeration was provided to the tanks throughout the experimental period.There was also no water exchange, just replenishment to maintain the initial water level.The bottom of the tanks was filled with a 5-cm layer of gross sand to allow water-soil interactions.
The experimental design consisted of three treatments and one control group, each one with five replicates.In the control tanks, the pH of water was not adjusted at any time, varying freely over the entire study.In the slight acidification treatment, the culture water was acidified daily with a 3.6 N HCl solution in order to reach a water pH between 5.5 and 6.5.In the moderate acidification treatment, there were daily applications of HCl solution to obtain a water pH between 4.5 and 5.5.Finally, in the alkalinization treatment, tanks received daily applications of Na 2 CO 3 (12 g) and 1 N NaOH (9 mL), in order to achieve a water pH between 8.5 and 9.5.At each day, the dosages of HCl, Na 2 CO 3 and NaOH used were adjusted to reach the designed pH for each treatment (slight acidification, moderate acidification and alkalinization).The acidic and alkaline solutions had their total volumes split in three equal doses, which were delivered at 0800, 0830 and 0900.The water pH at 0730 was used to define the amounts of the acidic or alkaline solutions used on that specific day.A second pH reading was performed daily at 1500.The reported pH of water was the mean value of those two determinations.
The soil pH and organic carbon concentration were determined every other week following the guidelines provided by Boyd, Wood, and Thunjai (2002).In the seventh experimental week, the pH and concentrations of DO, TAN, TS and H 2 S in water were observed on a diel basis.For that, water samples were taken from the culture tanks every two (pH, DO, TS and H 2 S) or four (TAN and NH 3 ) hours.The growth performance variables analyzed were the following: survival (%), final body weight (g), specific growth rate (% day -1 ; SGR = [Ln (final weight) -Ln (weight initial)] -1 days of culture) x 100), fish yield (g m -3 day -1 ) and feed conversion ratio (FCR = feed consumed -1 body weight gain).Metabolic performance of fish was assessed by the respiratory rate using 2.5-L respirometers (Barbieri, Passos, & Garcia, 2005).Weekly, from the third experimental week, one fish from each treatment was allotted to one respirometer.For that, one animal was withdrawn from each tank and those individuals not used for the respirometer assays were discarded.Each respirometer was filled up with 2.0-L filtered water taken from the culture tanks.Initially, air was bubbled into the respirometer water for one hour and the resulting concentration of dissolved oxygen in water was measured.Next, one fish was placed inside the respirometer for four hours.After that, the DO concentration in the respirometer water was measured by the Winkler method with azide modification.The respiratory rate in μg DO g -1 fish h -1 was obtained by the following equation: respiratory rate = [(DO i -DO f ) -1 fish body weight (g) 4h -1 ] x 2 L (respirometer volume), where DO i is the initial concentration of DO and DO f is the concentration of DO in the respirometer water after 4 hours.
The results were analyzed by one-way ANOVA.When a significant difference was detected between the treatments (p < 0.05), the means were compared pairwise by Tukey's test, for equalvariance variables, or Games-Howell's test, for nonequal variance variables.The assumptions of normal distribution (Shapiro-Wilk's test) and homogeneity of variance (Levene's test) were checked before analysis.The SPSS v.15.0 and Windows Excel 2010 software were used for the statistical analyses.

Results and discussion
Water and soil quality The temperature and concentration of dissolved oxygen in water were not significantly affected by water acidification (slight and moderate) and alkalinization (Table 1).The average temperature of water at 0800 and 1600 were 27.6 ± 0.40°C (27.0 -28.6°C) and 30.8 ± 1.14°C (28.2 -33.3°C), respectively.In the last week, the average concentration of DO in water was 4.5 ± 1.9 mg L -1 (1.8 -7.6 mg L -1 ).Therefore, it seems that the acidification and alkalinization procedures have not impaired the release of O 2 by photosynthesis to the water.The acidification of water has increased the concentrations of free CO 2 in the tanks.The highest concentration of free CO 2 (23.7 mg L -1 ) was found in the moderately acidified tanks.On the other hand, the alkalinization of water has reduced the concentrations of free CO 2 in water (Table 1).It is accepted that concentrations of free CO 2 in water above 20 mg L -1 may be stressful to fish (Danley, Kenney, Mazik, Kiser, & Hankins, 2005).However, it has probably not affected Nile tilapia fingerlings because they have reached the highest final body weight only in the moderately acidified tanks (23.4 ± 1.3 g; Table 2).
The specific conductance (SC) of water increased with the acidification and, mainly, alkalinization of water.The ionic concentration or salinity of water is the main factor responsible for the water SC.However, once the isosmotic point for Nile tilapia is near 12 g L -1 (Hassan et al., 2013) and the highest TDS (total dissolved solids) of water in the present work was 870 gm L -1 , these minor variations in SC of water have probably not affected the tilapia physiology.
The acidification of water has decreased the total alkalinity (TA) of water in direct proportion with the level of acidification (Table 1).The average TA of water was 13.5 ± 2.0 and 5.4 ± 1.7 mg L -1 CaCO 3 eq.for the slight and moderate acidification, respectively.
A minimum TA of 20 mg L -1 CaCO 3 eq. is required for an acceptable water pH buffering in aquaculture tanks (Boyd et al., 2016).In the present work, however, since the pH of water was deliberately controlled to reach certain levels, the effect of low TA on water pH was null.While the acidification of water has increased the total hardness (TH), the alkalinization has reduced it (Table 1).The highest TH of 223.4 mg L -1 CaCO 3 was found in one of the moderately acidified tanks; the lowest TH of 102.7 mg L -1 CaCO 3 was found in one of the tanks subjected to alkalinization.Boyd et al. (2016) recommended TH between 40 -300 mg L -1 CaCO 3 for aquaculture production.Therefore, the values of TH for both the acidified and alkaline tanks have remained within the appropriate range.
The acidification of water has not significantly affected the concentrations of nitrite in water when compared to the control tanks (Table 1), which, in general, were very low (0.04 -0.12 mg L -1 ).Interestingly, the concentrations of nitrite have been zeroed in the alkalinized tanks.Yanbo, Wenju, Weifen, and Zirong (2006) have determined the 96h LC 50 nitrite for Nile tilapia at 28.2 mg L -1 .Therefore, the safe concentration of nitrite for tilapia in freshwater is 0.3 mg L -1 (1% 96-h LC 50 ).Accordingly, the concentrations of nitrite found herein (0.12 mg L -1 ) have probably caused no damage to fish.
There were significant increases in the concentrations of reactive phosphorus and dissolved iron in water by the moderate acidification implemented.The slight acidification and alkalinization of water had no expressive effect on the phosphorus and Fe 2+ concentrations (Table 1).The water acidification has led to soil acidification: the pH of soil in the control and moderately acidified tanks were 7.9 ± 0.5 (7.3 -8.9) and 6.4 ± 1.1 (4.6 -7.6), respectively.The increase in the concentrations of phosphorus and iron in water in the moderately acidified tanks may be explained by the release of these ions from the acidic soils into the water (Falagán, Sánchez-España, & Johnson, 2014).The concentrations of organic carbon in soil were not significantly affected by the treatments (ANOVA p < 0.05).At the end, these concentrations were 0.35% ± 0.16 (control), 0.33% ± 0.09 (slight acidification), 0.33% ± 0.10 (moderate acidification) and 0.30% ± 0.08 (alkalinization).
The acidification of water, either slight or moderate, promoted an increase in H 2 S concentrations of water (Table 1).The alkalinization of water had no significant effect on that variable.It is possible that the ions Fe +2 and S -2 have been released together to the water along the soil acidification (Lahav, Ritvo, Slijper, Hearne, & Cochva, 2004).While the average concentrations of H 2 S in water were 0.73 ± 0.31 mg L -1 and 0.62 ± 0.28 mg L -1 for the slightly-and moderatelyacidified tanks, respectively, H 2 S levels as high as 0.96 mg L -1 were detected in the acidified tanks.Exposure to 0.1 -0.5 mg H 2 S L -1 caused severe biochemical and physiological damages in channel  ncentration of L -1 (El-Shafai in alkalinized H 3 in some in els were very s he diel monit increased from 0.75 mg L -1 at .91 ± 0.78 mg nced increase (midnight), 7.9 mg L -1 lfide concentr 800, when no s Figure 3).Fo moderate), r have matche de over the on of H2S in er drops (Boyd l  ).On the H 2 S were reg 0.42 mg L -1 ) an was detected i

Conclusion
The acidification of water up to pH 5.5 has not affected the growth performance of Nile tilapia fingerlings in eutrophic tanks.Accordingly, the suitable range of water pH for rearing Nile tilapia should be extended from 6.5 -9.0 to 5.5 -9.0.
Figure 1 of Nile ti DO modera those o the con signific the oth a maxim at 1400 L -1 at th but no since th Figure NH 3 i T for fi 2004) migh along V betwe The t ± 0.5 and d 1800.5.49 conce deter fell ag was acidif conce conce moni sulfid

Figure
Figure 3. Diel m Nile tilapia tanks Respirometer p Except for fish in the al than fish sto (284.4 and 12 the consumpt respirometers treatments and (Figure 4).

Table 1 .
Water quality in Nile tilapia outdoor tanks after eight weeks of culture (mean ± SD; n = 5).