Water Renewal Rate and Temperature on the Growth Performance and Physiology of Piaractus brachypomus in a Recirculating Aquaculture System (RAS)

Abstract

This study evaluated the effects of water renewal rate and temperature on the growth performance and physiological responses of juvenile Piaractus brachypomus reared in a recirculating aquaculture system (RAS). A total of 336 fish (1.35 ± 0.24 g) were distributed in six RAS units under two water renewal rates (42 and 128 L h⁻¹) and three temperatures (26, 29, and 32 °C) for 45 days. Temperature was the main factor affecting growth, with higher final weight and total length at 29 and 32 °C throughout the experimental period. Water renewal rate significantly influenced feeding efficiency and energy allocation. Higher renewal (128 L h⁻¹) increased dissolved oxygen and daily feed intake and resulted in higher hemoglobin levels and hepatic lipid deposition, particularly at 32 °C, indicating greater metabolic activity. Conversely, the lower renewal rate (42 L h⁻¹) was associated with better feed conversion ratios at 29 °C and higher muscle lipid content at 26 °C, suggesting reduced energy expenditure. Hematocrit, total plasma protein, and cholesterol were primarily influenced by temperature, with higher values at 29 and 32 °C, while glucose, triglycerides, and liver enzymes were unaffected. Overall, temperatures of 29–32 °C optimized growth, while water renewal rate modulated feed utilization, physiological responses, and lipid deposition. These findings highlight the importance of jointly optimizing temperature and water renewal rate in RAS to enhance growth performance and metabolic balance in juvenile P. brachypomus.

1. Introduction

Pirapitinga (Piaractus brachypomus) is a native fish species of South America where it occurs in the Orinoco, Amazon, and Solimões river basins [1]. It plays a significant role in the region, especially in the Amazon basin [2]. Recognized as a crucial species in aquaculture in several countries, such as Colombia, Brazil, Peru, Venezuela, Vietnam, Thailand, Malaysia, and Bangladesh [3], the species stands out for its ability to tolerate a wide range of variations in water quality parameters, exhibiting accelerated growth, superior meat quality, and omnivorous feeding habits [4].

RAS is increasingly used in aquaculture due to its various advantages, such as higher productivity, improved biosecurity, and the possibility of recycling waste [5], as well as its efficiency in controlling water temperature and physicochemical quality and reducing environmental impacts [6,7]. Among the main characteristics of RAS, water renewal rate is important for the proper functioning of the system [8]. According to [8], tilapia (Oreochromis niloticus) presented lower feed consumption when grown in tanks with lower water renewal rates. Furthermore, an improvement in water quality was achieved with a low renewal rate for rainbow trout (Oncorhynchus mykiss), especially in weeks with a higher feeding rate [9]. A decrease in water exchange rate affected the growth of juvenile lumpfish (Cyclopterus lumpus L.) [10]. Therefore, studies related to water renewal rate within RAS tanks are of great importance.

Temperature plays a crucial role in fish health and can have both positive and negative effects [11]. Fish species differ in their temperature preferences, and exposure to values outside their ideal range can be stressful and limit growth and survival [12]. For example, variation in ideal temperature directly impacted the growth and physiology of pirapitinga (P. brachypomus) [13], pacamã (Lophiosilurus alexandri) [14], and tambaqui (Colossoma macropomum) [15]. Temperature can also influence lipid accumulation [16], which may occur due to low metabolic activity at low temperatures [17], with direct impacts on fish welfare. Decreased cholesterol levels may indicate stress due to low water temperature, leading to increased energy consumption by fish [18], which can have negative effects on endogenous cholesterol synthesis [19]. Recent studies have demonstrated how P. brachypomus successfully adapts to recirculating aquaculture systems (RAS) [13,20]. The aim of the present study, therefore, was to evaluate the effect of different water renewal rates and water temperatures on the growth performance and physiology of juvenile P. brachypomus kept in RAS.

2. Materials and Methods

2.1. Fish Cultivation and Conditions

The present study was carried out for 45 days at the Laboratório de Aquicultura of the Universidade Federal de Minas Gerais and was approved by the ethics committee on the use of animals at UFMG (Protocol 309/2023, CEUA/UFMG, Belo Horizonte, Brazil).

2.2. Acclimation to the Experimental Environment

P. brachypomus larvae were acquired from the company Biofish^® (Biofish Alevinos, Porto Velho, Brazil). After the larval rearing period carried out in the same laboratory as the present study, the juveniles were acclimated to the experimental conditions for 14 days, maintaining the physicochemical conditions of the water within the appropriate range for the species, and fed the same commercial diet used in the experiment.

2.3. Experimental Protocol

A total of 336 juvenile P. brachypomus, with an average weight of 1.35 ± 0.24 g and an average length of 3.97 ± 0.40 cm, were used in the experiment. The animals were distributed in tanks with 28 L of useful volume at a density of 1 juvenile/2 L of water. The tanks were distributed among six RAS with 4 tanks/RAS (totaling 24 tanks), as described by [21]. The tanks were kept at a controlled temperature, with supplemental aeration and a photoperiod of 12 L:12D (digital timer, Key West DNI Group, São Paulo, SP, Brazil).

The relationship between water renewal rate and temperature was tested using two water renewal rates (42 L/h and 128 L/h, referred to as Q_42L/h and Q_128L/h, respectively) and three temperatures (26, 29, and 32 °C, referred to as T_26°C, T_29°C, and T_32°C, respectively) in a 2 × 3 factorial design, with four replicates each. The renewal rates of 42 L/h and 128 L/h correspond to 1.5 and 4.5 tank volume changes per hour, respectively.

The fish were fed to apparent satiation three times a day (08:00, 12:00, and 16:00 h) with an extruded commercial diet measuring 1.8 mm in diameter and containing the following: 40% (min.) crude protein, 7% (min.) ether, 140 g/Kg (max.) mineral matter, 50% (max.) crude fiber, 10 g/Kg (min.) calcium, 25 g/Kg (max.) calcium, 400 mg/Kg (min.) vitamin E, and 1000 mg/Kg (min.) vitamin C (Wean Prime, Bernaqua, Belo Horizonte, Brazil). Upon reaching satiety, leftover feed was collected, dried in an oven (Nova Ética/Ethik, Vargem Grande Paulista, Brazil) at 55 °C and weighed to calculate consumption.

2.4. Water Quality

After the experimental period, temperature, pH, and conductivity were measured every day after the first feeding, while dissolved oxygen and total ammonia were measured on weekends. Dissolved oxygen and temperature were measured using an oximeter (YSI 6920VZ2 multiparameter probe, Yellow Springs, OH, USA), while pH and conductivity were measured using a conductivity meter (Hanna Instruments HI9146, Barueri, Brazil). Total ammonia was measured by colorimetric test using a commercial kit (Labcon, Alcon Pet, Camboriú, Brazil).

2.5. Growth Performance Indices

Biometrics were performed every 15 days of culture. The animals were anesthetized with 50 mg/L of eugenol [21] and measured for weight using a precision scale (Marte Precision Scale, Ad5002, Santa Rita do Sapucaí, Brazil) and length using a digital caliper (Starret^®-799 series, São Paulo, Brazil). Data for feed consumption, weight, and length were used to calculate the following:

-

Daily weight gain (g/day) (DWG) = weight gain/days of experiment;

-

Initial body weight (IW);

-

Final body weight (FW);

-

Total length (cm) (TL);

-

Daily feed intake (DFI);

-

Feed conversion ratio (FCR) = feed intake/biomass gain;

-

Specific growth rate (%/day) (SGR) = SGR = 100 (ln FW − ln IW)/ΔT, where IW is initial weight, FW is final weight, and Δt is the number of days between samplings.

2.6. Blood Analyses

After 45 days, blood was collected from two animals of each tank (n = 8 per treatment). Blood samples were dispensed into microtubes containing sodium heparin anticoagulant (10%) for hemoglobin determination using a commercial colorimetric kit (Quibasa-Bioclin, Belo Horizonte, Brazil) and hematocrit using the microhematocrit method [22] and capillary tubes. Total plasma protein was determined with an analog refractometer (0 to 90% Brix)—RHB0-90 after rupture of the microhematocrit tube. The number of erythrocytes was determined by diluting 10 µL of whole blood in 2 mL of citrate formalin and counting in a Neubauer chamber. The remaining whole blood was centrifuged (4000 RPM for 10 min) to separate the plasma and determine glucose, triglycerides, cholesterol, alanine aminotransferase (ALT), and aspartate aminotransferase (AST) by colorimetric and enzymatic methods using commercial kits (Quibasa-Bioclin, Belo Horizonte, Brazil), with readings by spectrophotometer (semi-automatic analyzer Bioclin (Bioclin-Quibasa, Belo Horizonte, Brazil): the hematimetric indices of mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), and mean corpuscular hemoglobin concentration (MCHC), were calculated according to the formulations established by [23]: MCV (fL) = (hematocrit × 10)/(number of erythrocytes (× 10⁶/µL)); MCH (pg) = (hemoglobin concentration × 10)/(number of erythrocytes (× 10⁶/µL)); MCHC (g/dL) = (hemoglobin concentration × 100)/hematocrit.

2.7. Tissue Assays and Biometric Indexes

After blood collection, the same animals (2 fish from each tank; n = 8 per treatment) were euthanized with an anesthetic overdose (285 mg/L eugenol) [24]. The liver, a dorsal portion of muscle, and adipose tissue were removed. Liver and muscle samples were frozen for further analysis of lipid concentrations, following the methodology described by [25]. The liver and adipose tissue were weighed for subsequent calculation of the following indexes:

-

Hepatosomatic index (%) = (liver weight/body weight) × 100;

-

Mesenteric fat index (%) = (adipose tissue weight/body weight) × 100.

2.8. Statistical Analysis

Sampling for data collection was performed as follows:

-

Water quality: Samples were collected from all tanks.

-

Growth performance indices: All juveniles were sampled for weight and length collection and subsequent calculation of zootechnical indices, totaling the experimental n of 56 juveniles per treatment.

-

Blood analyses: 2 juveniles per tank were randomly selected for blood sample collection, totaling a sample n of 8 juveniles per treatment.

-

Tissue assays and biometric indexes: The same 2 juveniles from each tank used for blood analysis were euthanized and used for tissue collection. Sample n of 8 juveniles per treatment.

All data were initially assessed for normal distribution (Shapiro–Wilk) and homogeneity of variance (Levene). Two-way analysis of variance (ANOVA) was performed to evaluate differences among treatments, followed by Tukey’s post hoc test (at the 5% level) to compare means. Kruskal–Wallis test was performed for the water quality analysis (dissolved oxygen, total ammonia, and electrical conductivity). Values are expressed as mean ± standard deviation (SD). Data were analyzed using InfoStat Statistical Software version 2020 (Universidad Nacional de Córdoba, Argentina).

3. Results

3.1. Water Quality

Dissolved oxygen showed a significant effect of Q (p = 0.0001), T (p = 0.0023), and the Q × T interaction (p < 0.0001) (

Table 1. Water quality parameters (mean ± standard deviation) for juvenile Piaractus brachypomus exposed to different water renewal rates and different temperatures after 45 days of cultivation in a recirculating aquaculture system (RAS).

Statistic	p-Value
	Dissolved Oxygen (mg/L) 2	pH 1	Total Ammonia (mg/L) 2	Electrical Conductivity (µS/cm) 2
Q	0.0001 **	0.0841 ns	0.1422 ns	0.0193 *
T	0.0023 **	<0.0001 **	0.1738 ns	0.0075 **
Q × T	<0.0001 **	0.6470 ns	0.3138 ns	0.0024 **
Treatments	Means for Q
Q42L/h	5.72 ± 0.17 b	6.74 ± 0.15	0.21 ± 0.10	0.48 ± 0.03 a
Q128L/h	6.01 ± 0.12 a	6.70 ± 0.13	0.15 ± 0.13	0.45 ± 0.03 b
Treatments	Means for T
T26°C	5.89 ± 0.25 a	6.85 ± 0.08 a	0.15 ± 0.13	0.45 ± 0.04 b
T29°C	5.72 ± 0.16 b	6.75 ± 0.05 b	0.17 ± 0.12	0.49 ± 0.02 a
T32°C	5.99 ± 0.08 a	6.56 ± 0.04 c	0.23 ± 0.07	0.46 ± 0.02 b
Treatments	Means for Q × T
Q42L/hT26°C	5.66 ± 0.10 c	6.88 ± 0.10	0.17 ± 0.13	0.47 ± 0.04 ab
Q42L/hT29°C	5.57 ± 0.08 c	6.77 ± 0.04	0.21 ± 0.10	0.49 ± 0.02 a
Q42L/hT32°C	5.92 ± 0.02 ab	6.56 ± 0.04	0.25 ± 0.00	0.48 ± 0.01 a
Q128L/hT26°C	6.12 ± 0.05 a	6.83 ± 0.06	0.13 ± 0.14	0.43 ± 0.03 b
Q128L/hT29°C	5.86 ± 0.03 bc	6.73 ± 0.04	0.13 ± 0.14	0.49 ± 0.01 a
Q128L/hT32°C	6.06 ± 0.05 a	6.55 ± 0.04	0.21 ± 0.10	0.44 ± 0.01 b

** p < 0.01, * p < 0.05. ns—not significant. ^1. ANOVA followed by Tukey’s test (5%). ² Kruskal–Wallis test (5%). Different lowercase letters in the same column are significantly different.

). The highest values were recorded at Q128 L/h at 26 °C and 32 °C, while the lowest occurred at Q42 L/h at 26 °C and 29 °C (p < 0.05). pH was influenced only by temperature (p < 0.0001), with the highest value at 26 °C and the lowest at 32 °C (p < 0.05). Electrical conductivity was affected by Q, T, and the Q × T interaction (p < 0.05), with the highest values at Q42 L/h at 29 °C and 32 °C, and at Q128 L/h at 29 °C (p < 0.05). Total ammonia showed no significant effect from Q, T, or the interaction between these factors (p > 0.05).

3.2. Growth Performance

After 15 days of cultivation, the final weight (FW) was influenced only by temperature (p < 0.05) (

Table 2. Growth performance and survival (mean ± standard deviation) for juvenile Piaractus brachypomus exposed to different water renewal rates and temperatures after 15 days of cultivation in a recirculating aquaculture system (RAS).

Statistic	p-Value
	FW (g)	TL (cm)	DWG (g/Day)	DFI (g)	FCR	SGR (%/Day)	Survival (%)
Q	0.6163 ns	0.2902 ns	0.4574 ns	0.0001 **	0.0002 **	0.8478 ns	0.6109 ns
T	<0.0001 **	<0.0001 **	<0.0001 **	<0.0001 **	<0.0001 **	<0.0001	0.8457 ns
Q × T	0.0862 ns	0.0087 **	0.0062 **	0.0096 **	0.0049 **	0.0061 **	0.2152 ns
Treatments	Means for Q
Q42L/h	7.24 ± 0.97	6.78 ± 0.31	0.36 ± 0.02	0.38 ± 0.04 b	1.06 ± 0.08 ab	10.42 ± 0.31	95.83 ± 6.43
Q128L/h	7.12 ± 1.34	6.69 ± 0.49	0.35 ± 0.03	0.44 ± 0.03 a	1.26 ± 0.07 a	10.44 ± 0.64	96.43 ± 8.34
Treatments	Means for T
T26°C	5.89 ± 0.64 c	6.30 ± 0.31 c	0.32 ± 0.02 b	0.39 ± 0.03 b	1.21 ± 0.14 b	9.90 ± 0.33 b	98.21 ± 3.31
T29°C	7.54 ± 0.76 b	6.83 ± 0.24 b	0.36 ± 0.02 a	0.40 ± 0.06 b	1.09 ± 0.14 a	10.66 ± 0.42 a	95.54 ± 7.58
T32°C	8.11 ± 0.53 a	7.08 ± 0.13 a	0.38 ± 0.01 a	0.45 ± 0.03 a	1.18 ± 0.07 b	10.73 ± 0.16 a	94.64 ± 9.92
Treatments	Means for Q × T
Q42L/hT26°C	6.33 ± 0.50	6.53 ± 0.20 b	0.34 ± 0.01 b	0.37 ± 0.02 cd	1.08 ± 0.04 b	10.14 ± 0.19 cd	96.43 ± 4.12
Q42L/hT29°C	7.26 ± 0.73	6.72 ± 0.22 ab	0.35 ± 0.02 ab	0.34 ± 0.01 d	0.97 ± 0.02 a	10.41 ± 0.26 ab	92.86 ± 10.10
Q42L/hT32°C	8.13 ± 0.70	7.09 ± 0.19 a	0.38 ± 0.01 a	0.43 ± 0.02 ab	1.14 ± 0.06 bc	10.70 ± 0.21 ab	98.21 ± 3.57
Q128L/hT26°C	5.44 ± 0.43	6.07 ± 0.21 c	0.31 ± 0.01 c	0.41 ± 0.02 bc	1.33 ± 0.02 d	9.66 ± 0.24 d	100.00 ± 0.00
Q128L/hT29°C	7.81 ± 6.87	6.93 ± 0.22 ab	0.37 ± 0.02 a	0.45 ± 0.03 ab	1.21 ± 0.08 c	10.90 ± 0.44 a	98.21 ± 3.57
Q128L/hT32°C	8.10 ± 0.39	7.08 ± 0.06 a	0.38 ± 0.003 a	0.46 ± 0.02 a	1.23 ± 0.04 c	10.77 ± 0.09 a	91.07 ± 13.52

** p < 0.01. ns—not significant. ANOVA followed by Tukey’s test (5%). Different lowercase letters in the same column are significantly different.

), with the highest value at 32 °C and the lowest at 26 °C. Total length (TL) and daily weight gain (DWG) were affected by temperature and the Q × T interaction (p < 0.05). The highest TL values occurred at Q42 L/h and Q128 L/h at 32 °C. DWG was highest at Q42 L/h at 32 °C and at Q128 L/h at 29 °C and 32 °C (p < 0.05). DFI was influenced by Q, T, and the interaction between the factors (p < 0.05), with the lowest value at Q42 L/h at 29 °C. Feed conversion ratio (FCR) was affected by Q, T, and the Q × T interaction (p < 0.05), with the worst value at Q128 L/h at 26 °C. SGR was influenced by temperature and the interaction of the factors (p < 0.05), with higher values at Q128 L/h at 29 °C and 32 °C. Survival did not differ between treatments (p > 0.05).

After 30 days of cultivation, final weight (FW) was influenced by temperature (p < 0.05) (

Table 3. Growth performance (mean ± standard deviation) for juvenile Piaractus brachypomus exposed to different water renewal rates and temperatures after 30 days of cultivation in a recirculating aquaculture system (RAS).

Statistic	p-Value
	FW (g)	TL (cm)	DWG (g/Day)	DFI (g)	FCR	SGR (%/Day)	Survival (%)
Q	0.7714 ns	0.4605 ns	0.8445 ns	<0.0001 **	0.0042 **	0.8875 ns	0.7899 ns
T	0.0002 **	<0.0001 **	0.0021 **	0.6218 ns	0.0014 **	0.9108 ns	0.0839 ns
Q × T	0.3991 ns	0.2558 ns	0.5930 ns	0.2482 ns	0.4064 ns	0.6958 ns	0.6077 ns
Treatments	Means for Q
Q42L/h	19.46 ± 3.86	9.24 ± 0.44	0.83 ± 0.14	0.78 ± 0.09 b	0.95 ± 0.17 a	6.67 ± 0.45	91.67 ± 10.48
Q128L/h	19.74 ± 2.97	9.15 ± 0.61	0.82 ± 0.18	0.93 ± 0.04 a	1.19 ± 0.28 b	6.70 ± 0.46	90.48 ± 12.31
Treatments	Means for T
T26°C	16.16 ± 1.89 b	8.62 ± 0.30 b	0.69 ± 0.09 b	0.86 ± 0.10	1.29 ± 0.28 b	6.73 ± 0.41	92.86 ± 10.10
T29°C	20.42 ± 2.54 a	9.37 ± 0.35 a	0.86 ± 0.14 a	0.83 ± 0.14	0.99 ± 0.20 a	6.63 ± 0.56	96.43 ± 5.40
T32°C	22.23 ± 2.28 a	9.59 ± 0.31 a	0.94 ± 0.12 a	0.87 ± 0.08	0.93 ± 0.12 a	6.70 ± 0.39	83.93 ± 13.63
Treatments	Means for Q × T
Q42lL/hT26°C	17.23 ± 1.65	8.83 ± 0.24	0.73 ± 0.08	0.81 ± 0.11	1.11 ± 0.13	6.67 ± 0.25	91.07 ± 13.52
Q42L/hlT29°C	20.05 ± 3.12	9.35 ± 0.46	0.85 ± 0.18	0.72 ± 0.09	0.87 ± 0.18	6.73 ± 0.80	96.43 ± 4.12
Q42L/hT32°C	21.95 ± 2.21	9.55 ± 0.28	0.92 ± 0.10	0.80 ± 0.05	0.88 ± 0.04	6.61 ± 0.11	87.50 ± 12.20
Q128L/hT26°C	15.1 ± 1.62	8.42 ± 0.18	0.64 ± 0.09	0.92 ± 0.05	1.46 ± 0.30	6.79 ± 0.57	94.64 ± 6.84
Q128L/hT29°C	20.78 ± 2.21	9.40 ± 0.28	0.86 ± 0.10	0.95 ± 0.03	1.11 ± 0.15	6.52 ± 0.26	96.43 ± 7.14
Q128L/hT32°C	22.52 ± 2.66	9.63 ± 0.38	0.96 ± 0.16	0.93 ± 0.05	0.99 ± 0.15	6.79 ± 0.57	80.36 ± 15.84

** p < 0.01. ns—not significant. ANOVA followed by Tukey’s test (5%). Different lowercase letters in the same column are significantly different.

), with higher values at 29 °C and 32 °C. Total length (TL) and dry weight gain (DWG) were affected only by temperature (p < 0.05), being higher at 29 °C and 32 °C. Daily feed intake (DFI) was influenced by flow rate (p < 0.05), with the highest value at Q128 L/h. Feed conversion ratio (FCR) was affected by flow rate and temperature (p < 0.05), with the best value at Q42 L/h and at 29 °C and 32 °C. Survival and SGR did not differ between treatments (p > 0.05).

After 45 days of cultivation, final weight (FW) and total length (TL) were influenced only by temperature (p < 0.05) (

Table 4. Growth performance (mean ± standard deviation) for juvenile Piaractus brachypomus exposed to different water renewal rates and temperatures after 45 days of cultivation in a recirculating aquaculture system (RAS).

Statistic	p-Value
	FW (g)	TL (cm)	DWG (g/Day)	DFI (g)	FCR	SGR (%/Day)	Survival (%)
Q	0.7874 ns	0.8788 ns	0.8809 ns	0.9245 ns	0.8156 ns	0.8830 ns	0.8724 ns
T	0.0300 *	0.0010 **	0.5754 ns	0.0003 **	0.0133 **	0.0012 **	0.7899 ns
Q × T	0.3617 ns	0.6522 ns	0.4940 ns	0.0259 **	0.851 ns	0.7439 ns	0.8724 ns
Treatments	Means for Q
Q42L/h	32.85 ± 3.67	12.20 ± 0.58	0.87 ± 0.15	1.10 ± 0.10	1.29 ± 0.26	3.43 ± 0.61	91.67 ± 10.48
Q128L/h	32.42 ± 5.03	12.18 ± 0.67	0.86 ± 0.17	1.10 ± 0.20	1.32 ± 0.39	3.46 ± 0.66	90.48 ± 12.31
Treatments	Means for T
T26°C	29.40 ± 3.17 b	11.51 ± 0.3 b	0.88 ± 0.11	0.96 ± 0.12 b	1.10 ± 0.16 a	3.99 ± 0.30 a	93.75 ± 9.69
T29°C	33.98 ± 3.95 ab	12.49 ± 0.47 a	0.90 ± 0.14	1.12 ± 0.10 a	1.26 ± 0.19 b	3.40 ± 0.42 ab	91.07 ± 9.92
T32°C	34.53 ± 4.15 a	12.56 ± 0.38 a	0.82 ± 0.21	1.22 ± 0.12 a	1.56 ± 0.39 b	2.93 ± 0.59 b	91.07 ± 8.32
Treatments	Means for Q × T
Q42L/hT26°C	31.14 ± 2.84	11.62 ± 0.29	0.93 ± 0.11	1.05 ± 0.07 bc	1.14 ± 0.15	3.95 ± 0.35	91.07 ± 13.52
Q42L/hT29°C	33.92 ± 5.73	12.50 ± 0.68	0.92 ± 0.20	1.09 ± 0.09 bc	1.22 ± 0.26	3.49 ± 0.42	87.50 ± 12.20
Q42L/hT32°C	33.49 ± 1.57	12.49 ± 0.15	0.77 ± 0.09	1.16 ± 0.11 ab	1.52 ± 0.20	2.84 ± 0.47	96.43 ± 7.14
Q128L/hT26°C	27.65 ± 2.70	11.39 ± 0.30	0.84 ± 0.09	0.88 ± 0.10 c	1.06 ± 0.17	4.04 ± 0.29	96.43 ± 4.12
Q128L/hT29°C	34.04 ± 1.88	12.49 ± 0.24	0.88 ± 0.07	1.15 ± 0.11 ab	1.30 ± 0.11	3.31 ± 0.45	94.64 ± 6.84
Q128L/hT32°C	35.57 ± 5.90	12.64 ± 0.54	0.87 ± 0.30	1.28 ± 0.11 a	1.60 ± 0.55	3.02 ± 0.76	85.71 ± 5.83

** p < 0.01, * p < 0.05. ns—not significant. ANOVA followed by Tukey’s test (5%). Different lowercase letters in the same column are significantly different.

), with the highest values at 29 °C and 32 °C. DFI was affected by temperature and the Q × T interaction (p < 0.05), with the lowest value at Q128 L/h at 26 °C. Feed conversion ratio (FCR) and SGR were influenced only by temperature (p < 0.05), with better FCR and higher SGR at 26 °C. Daily weight gain (DWG) and survival did not differ between treatments (p > 0.05).

3.3. Hematological and Biochemical Parameters

After 45 days, hemoglobin was influenced by Q and T (p < 0.05) (

Table 5. Hematological parameters (mean ± standard deviation) for juvenile Piaractus brachypomus exposed to different water renewal rates and temperatures after 45 days of cultivation in a recirculating aquaculture system (RAS).

Statistic	p-Value
	Hemoglobin (g/dL)	Hematocrit (%)	Erythrocytes (106/µL)	MCV (fL)	MCH (pg)	MCHC (g/dL)
Q	0.0079 **	0.4765 ns	0.8089 ns	0.6822 ns	0.3325 ns	0.3875 ns
T	<0.0001 **	<0.0001 **	0.5575 ns	0.3792 ns	0.3853 ns	0.0935 ns
Q × T	0.1985 ns	0.5017 ns	0.8609 ns	0.8224 ns	0.6034 ns	0.5575 ns
Treatments	Means for Q
Q42L/h	7.23 ± 0.68 b	27.46 ± 4.64	1.16 ± 0.28	249.06 ± 75.48	64.62 ± 11.99	26.82 ± 3.69
Q128L/h	7.74 ± 0.99 a	28.14 ± 3.88	1.14 ± 0.27	260.99 ± 67.62	71.26 ± 18.12	27.69 ± 3.05
Treatments	Means for T
T26°C	6.73 ± 0.53 b	24.73 ± 2.87 b	1.10 ± 0.28	243.38 ± 66.76	65.13 ± 17.02	27.65 ± 4.06
T29°C	7.68 ± 0.84 a	26.87 ± 2.42 b	1.20 ± 0.28	234.43 ± 50.75	66.12 ± 11.98	28.35 ± 2.20
T32°C	8.06 ± 0.65 a	31.73 ± 3.95 a	1.16 ± 0.25	286.07 ± 85.68	72.56 ± 17.05	25.67 ± 3.28
Treatments	Means for Q × T
Q42L/hT26°C	6.69 ± 0.39	24.25 ± 2.19	1.09 ± 0.24	233.63 ± 57.71	63.72 ± 11.16	27.87 ± 3.93
Q42L/hT29°C	7.24 ± 0.56	26.00 ± 2.07	1.21 ± 0.35	230.11 ± 58.73	63.29 ± 13.24	27.93 ± 2.33
Q42L/hT32°C	7.78 ± 0.62	32.13 ± 4.76	1.20 ± 0.25	283.45 ± 99.54	66.84 ± 12.80	24.66 ± 3.97
Q128L/hT26°C	6.76 ± 0.66	25.29 ± 3.59	1.11 ± 0.33	254.53 ± 79.01	66.54 ± 22.17	27.39 ± 4.51
Q128L/hT29°C	8.13 ± 0.87	27.86 ± 2.54	1.20 ± 0.21	239.37 ± 43.96	68.96 ± 10.68	28.84 ± 2.11
Q128L/hT32°C	8.34 ± 0.59	31.29 ± 3.09	1.12 ± 0.27	289.07 ± 74.50	78.28 ± 19.63	26.83 ± 1.93

** p < 0.01. ns—not significant. ANOVA followed by Tukey’s test (5%). Different lowercase letters in the same column are significantly different.

); increased values were observed at Q128 L/h and for T29 °C and T32 °C (p < 0.05). Hematocrit was influenced by T (p < 0.05) (p > 0.05), with higher levels at T32 °C (p < 0.05).

Regarding biochemical parameters (

Table 6. Blood biochemistry (mean ± standard deviation) for juvenile Piaractus brachypomus exposed to different water renewal rates and temperatures after 45 days of cultivation in a recirculating aquaculture system (RAS).

Statistic	p-Value
	Glucose (mg/dL)	Total Plasma Protein (g/dL)	Triglycerides (mg/dL)	Cholesterol (mg/dL)	ALT (U/L)	AST (U/L)
Q	0.4964 ns	0.9421 ns	0.3941 ns	0.3946 ns	0.3642 ns	0.6983 ns
T	0.1164 ns	0.0204 *	0.6205 ns	0.0102 *	0.1040 ns	0.3430 ns
Q × T	0.5069 ns	0.8253 ns	0.5966 ns	0.0625 ns	0.0639 ns	0.4658 ns
Treatments	Means for Q
Q42lL/h	77.32 ± 12.56	4.97 ± 0.28	227.01 ± 68.40	122.63 ± 18.12	9.95 ± 4.48	135.60 ± 45.22
Q128L/h	79.78 ± 12.79	4.97 ± 0.33	207.65 ± 59.46	127.64 ± 19.44	11.53 ± 5.53	128.50 ± 61.88
Treatments	Means for T
T26°C	76.20 ± 12.61	4.80 ± 0.30 b	224.68 ± 57.81	114.27 ± 15.46 b	10.42 ± 3.78	137.23 ± 45.99
T29°C	75.51 ± 12.60	4.99 ± 0.28 ab	222.78 ± 72.19	130.16 ± 21.24 a	12.71 ± 6.16	145.23 ± 68.22
T32°C	83.93 ± 11.55	5.11 ± 0.25 a	203.45 ± 62.93	131.45 ± 15.06 a	8.93 ± 4.16	115.00 ± 43.13
Treatments	Means for Q × T
Q42L/hT26°C	72.49 ± 6.42	4.83 ± 0.29	214.43 ± 37.84	113.77 ± 17.30	11.29 ± 4.42	135.71 ± 30.05
Q42L/hT29°C	74.09 ± 15.70	5.00 ± 0.30	243.30 ± 91.39	128.54 ± 21.03	9.57 ± 5.09	138.33 ± 71.46
Q42L/hT32°C	85.37 ± 10.87	5.08 ± 0.21	224.22 ± 73.07	126.33 ± 14.59	9.00 ± 4.24	133.14 ± 36.07
Q128L/hT26°C	79.91 ± 16.38	4.77 ± 0.34	232.37 ± 70.90	114.76 ± 14.57	9.20 ± 2.59	139.00 ± 63.14
Q128L/hT29°C	76.94 ± 9.45	4.97 ± 0.27	207.39 ± 55.51	131.57 ± 22.77	15.86 ± 5.76	151.14 ± 70.46
Q128L/hT32°C	82.49 ± 12.76	5.14 ± 0.30	179.71 ± 42.19	136.57 ± 14.59	8.86 ± 4.41	96.86 ± 44.29

* p < 0.05. ns—not significant. ANOVA followed by Tukey’s test (5%). Different lowercase letters in the same column are significantly different.

), no significant differences were observed for glucose, triglycerides, and ALT and AST enzymes. However, total protein was influenced by T (p < 0.05), with the highest value for T32 °C (p < 0.05). Cholesterol was also influenced by temperature (T) (p < 0.05), with the highest values for T29 °C and T32 °C (p < 0.05).

3.4. Tissue Lipid Content

Liver lipid content was altered by Q and by the Q × T interaction, with the highest value for Q128 L/hT32 °C (p < 0.05) (

Table 7. Liver and muscle lipids (mean ± standard deviation) for juvenile Piaractus brachypomus exposed to different water renewal rates and temperatures after 45 days of cultivation in a recirculating aquaculture system (RAS).

Statistic	p-Value
	Lipid in Liver (mg/g of Tissue)	Lipid in Muscle (mg/g of Tissue)
Q	0.0070 **	0.3123 ns
T	0.2614 ns	0.0134 **
Q × T	0.0009 **	0.0063 **
Treatments	Means for Q
Q42L/h	34.97 ± 9.97 b	11.79 ± 4.64
Q128L/h	43.07 ± 11.10 a	10.46 ± 5.59
Treatments	Means for T
T26°C	36.46 ± 8.21 b	13.39 ± 4.77 a
T29°C	29.84 ± 10.43 a	8.42 ± 4.59 b
T32°C	43.59 ± 15.15 a	11.35 ± 5.08 ab
Treatments	Means for Q × T
Q42L/hT26°C	31.87 ± 2.59 b	16.24 ± 2.13 a
Q42L/hT29°C	45.03 ± 13.02 ab	10.64 ± 3.95 abc
Q42L/hT32°C	29.56 ± 7.52 b	7.94 ± 3.62 bc
Q128L/hT26°C	41.05 ± 9.54 ab	11.01 ± 5.19 abc
Q128L/hT29°C	37.24 ± 8.66 b	5.11 ± 3.55 c
Q128L/hT32°C	54.82 ± 8.89 a	14.08 ± 4.58 ab

** p < 0.01. ns—not significant. ANOVA followed by Tukey’s test (5%). Different lowercase letters in the same column are significantly different.

). Muscle lipid content was influenced by T and by the Q * T interaction (p < 0.05), with the highest value for Q42 L/hT26 °C.

4. Discussion

4.1. Water Quality

Temperature and water renewal rate are factors of RAS that can affect fish growth, survival, and physiology [26,27,28]. Furthermore, these factors are easily controlled, considering that they are directly linked to the RAS environment. Our results showed that these factors can influence the growth, survival, and physiology of juvenile P. brachypomus in different ways. Survival was not affected after the experiment, similar to what was observed in [29] when evaluating the effects of different water flow rates on the production of juvenile L. alexandri, and to that reported in [30] for C. macropomum grown at different temperatures.

The present study found that the evaluated treatments influenced the measured water quality parameters. Dissolved oxygen levels were highest for Q_128L/hT_26°C and Q_128L/hT_32°C. Electrical conductivity was lowest for Q_128L/hT_26°C, while pH was highest for T_26°C and the lowest for T_32°C. The higher dissolved oxygen levels may be associated with the observed increase in oxygenation of the systems with higher water flow rates [31]. In addition, the more acidic pH observed at T_32°C may be related to an increased metabolic rate, which intensifies fish respiration and the release of CO₂ [32]. In turn, this gas dissolves in water, forming carbonic acid (H₂CO₃), which contributes to water acidification [33]. However, despite the changes observed in the present study, the values found are within those recommended for the cultivation of Neotropical fish species [34].

4.2. Growth Performance

After 15 days of cultivation, final weight was directly influenced by temperature, with the highest value at T_32°C and the lowest at T_26°C. These results are similar to those found by [5], who investigated the effects of the temperatures of 28 °C and 24 °C on the growth of juvenile P. brachypomus and also found effects of temperature after 20 days of cultivation, with the highest growth values being associated with higher temperatures. The interaction of the factors revealed that Q_128L/hT_26°C presented the lowest TL, DWG, and SGR and worse FCR. Ref. [35] reported that the use of high flow rates can reduce growth rates and worsen feed conversion ratio, similar to that observed in the present study. In contrast, ref. [36] observed greater growth in Scophthalmus maximus kept with high water flow rates. Furthermore, high water flow rates can exceed the optimal swimming capacity of the species, requiring animals to swim more actively and increase energy consumption, directly affecting growth performance [37], while for Q_42L/hT_26°C the lower flow rates seem to be more related to temperature than to water flow, since temperature can decrease consumption and growth rates [26].

DFI was also affected by water flow and temperature, with the lowest values for Q_42L/hT_26°C and Q_42L/hT_29°C. Similar results were found for the growth performance of S. maximus, with observed differences in feed consumption, with the lowest values for the lowest water flow [38]. However, it is worth mentioning that determining the best water flow rate may be directly related to the behavior of the species in its natural environment, because studies carried out with Lateolabrax maculatus [39] and Oreochromis niloticus [40] demonstrated a positive effect of water flow on growth. However, the results observed for DFI may also be related to temperature, since other studies have also shown that increased water temperature is responsible for increasing feed consumption, as previously observed by [5] with juvenile P. brachypomus.

Despite the effects of water flow observed after 15 days of culture, growth performance after 30 days showed differences only for FCR and DFI, which were influenced by water flow, while FW, TL, and DWG were highest for T_29°C and T_32°C, while the FCR presented the best values at these temperatures. After 45 days of culture, TL was highest and the FCR was worse for T_29°C and T_32°C, while SGR was highest for T_26°C and T_29°C. DFI was highest for Q_42L/hT_32°C, Q_128L/hT_29°C, and Q_128L/hT_32°C. In general, these findings are consistent with those reported by [41], who evaluated different water flow rates on the performance of juvenile Oncorhynchus mykiss reared for six months in recirculating aquaculture systems. The present results also suggest that water flow should be carefully monitored during fish growth.

The effects of temperature on growth parameters are well known, and it has been reported that increased temperature enhances metabolic rates and thus increases food intake and animal growth [42]. Although P. brachypomus is a species that inhabits environments where temperature varies between 25 and 34 °C [43], the best growth is observed with water temperatures between 27 and 28 °C [5,44]. Therefore, our results agree with the known effects of temperature on the growth of this species.

4.3. Hematological and Biochemical Parameters

Hematological responses portray important information regarding the physiological state of fish raised in RAS [45]. The current study found an effect of higher temperature on hemoglobin and hematocrit, with no changes in MCV, MCH, and MCHC. It is known that changes in hemoglobin are indicative of a greater need for oxygen supply to tissues and organs [46]. The increase in hemoglobin with higher temperature observed in the present study is similar to that observed for P. mesopotamicus [47] and Clarias gariepinus × C. macrocephalus [48]. However, the hemoglobin results observed here are contrary to those observed for O. niloticus, for which the authors suggested that the chronic use of different water flows can produce a compensatory response at the end of eight weeks [10]. One hypothesis for the differences obtained between the studies may be linked to a species-specific response or due to differences in experiment duration. The absence of differences in MCV, MCH, and MCHC demonstrate that the temperature or water flow used in the current study were not sufficient to affect the morphology and properties of red blood cells [49].

Blood biochemical parameters are used as indicators of an animal’s physiological state [50]. In the present study, higher temperatures resulted in an increase in total plasma protein, similar to that observed in [51] during the cultivation of juvenile grass carp (Ctenopharyngodon idella) at temperatures of 23 °C and 28 °C. The highest values of cholesterol observed here were also for the highest temperatures, which allows this response to be related to the higher feed consumption observed at these temperatures, as observed in O. niloticus [52]. Increased activity of the enzymes ALT and AST are indicative of the presence of damage to the liver, bile ducts, and tissues [53]. Since the temperature and water flow rates evaluated in the present study did not influence these parameters, we can suggest that the health of the liver and other tissues was preserved. Likewise, the absence of changes in triglycerides and glucose also directly reflect energy metabolism and allow us to state that the factors evaluated in the study did not represent a stressful condition for the animals.

4.4. Tissue Lipid Content

The tissue lipid content in juvenile Piaractus brachypomus demonstrates that water temperature and flow rate jointly modulated energy intake, expenditure, and storage. Higher hepatic lipid deposition in Q42 L/hT29 °C, Q128 L/hT26 °C, and Q128 L/hT32 °C indicates that these environmental factors influence lipid accumulation patterns, while increased muscle lipid content in Q42 L/hT26 °C reflects differences in energy partitioning among tissues.

The liver is the primary site of lipid storage and metabolic regulation in fish, and hepatic lipid deposition is closely linked to temperature-dependent metabolic rates and the balance between energy intake and expenditure. Elevated temperatures generally increase metabolic rate and feed intake, favoring lipid synthesis and storage, whereas lower temperatures may reduce lipid mobilization [42,54]. Similar temperature-mediated effects on lipid metabolism and tissue allocation have been reported for pacu species and other freshwater fish [18,19,26].

Water flow rate affects lipid deposition by altering swimming activity and energetic demand. Higher flow rates increase energy expenditure and may reduce lipid storage; however, under favorable thermal conditions, increased feed intake can offset these costs, explaining the elevated hepatic lipid content observed in Q128 L/hT32 °C [37].

The higher muscle lipid content observed at Q42 L/hT26 °C likely reflects reduced metabolic demand under cooler temperatures and lower flow rates. Although muscle lipids serve as energy reserves, excessive accumulation may negatively affect filet nutritional and sensory quality [55].

Overall, these results indicate that temperature primarily regulates metabolic rate and lipid synthesis, while water flow modulates energy expenditure, jointly determining lipid storage and energy allocation in juvenile P. brachypomus.

5. Conclusions

Temperatures of 29 and 32 °C promote better growth performance of P. brachypomus juveniles in RAS and influence certain hematological and biochemical parameters, such as hemoglobin concentration, hematocrit, total plasma protein, and cholesterol. Overall, a higher water renewal rate in the RAS has a positive impact on growth and physiological variables but also leads to greater hepatic lipid deposition in the fish.