Interactive effects of dietary fishmeal level and feeding frequency on growth performance, nutrient utilization, and waste outputs in juvenile largemouth bass (Micropterus salmoides)

1 Introduction

The rapid expansion of aquaculture has intensified concerns regarding nutrient discharge and environmental sustainability, particularly in intensive production systems where feed inputs represent the primary source of nutrient loading (Cho and Bureau, 2001; Hua et al., 2008). In such systems, the nitrogen (N), phosphorus (P), and carbon (C) released through uneaten feed, fecal egestion, and metabolic excretion constitute the major pathways of nutrient enrichment in surrounding aquatic environments (Amirkolaie, 2011; Bunting and Pretty, 2007; Cho and Bureau, 1997; Yuan et al., 2019). These nutrient outputs originate largely from feed ingredients, its composition, and feeding management practices (Antony Jesu Prabhu et al., 2019; Brinker and Reiter, 2011; Unger and Brinker, 2013), which together determine the nutrient utilization efficiency and the magnitude of waste generation (Bureau et al., 2006; Chowdhury et al., 2013). Consequently, improving feeding strategies and diet formulations has become a central approach to reducing their environmental impacts while maintaining production efficiency in modern aquaculture systems.

Fishmeal remains the principal protein source in feeds for carnivorous fish species due to its high palatability, balanced amino acid profile, and superior digestibility (González-Rodríguez et al., 2016; Karapanagiotidis et al., 2019; Jobling, 2012; Tran-Ngoc et al., 2019; Naylor et al., 2009). Despite its nutritional advantages, fishmeal production relies on capture fisheries, raising concerns regarding resource sustainability and fishing pressure on wild stocks (Naylor et al., 2009; Wang et al., 2013, Wang et al., 2008). Nevertheless, fishmeal is also rich in nitrogen, phosphorus, and carbon, and excessive dietary inclusion can exacerbate nutrient discharge into culture waters (Jobling, 2012; Tran-Ngoc et al., 2019). In addition to environmental concerns, the rising cost and limited availability of fishmeal have stimulated substantial efforts to reduce its inclusion through partial replacement with alternative protein sources, such as poultry by-product meal and plant proteins (Tidwell et al., 2005; Wang et al., 2013; González-Rodríguez et al., 2016; Karapanagiotidis et al., 2019). While such substitutions may reduce the feed costs and phosphorus inputs (Amirkolaie, 2011; Hua et al., 2008), they often compromise growth performance (Francis et al., 2001) and nutrient retention (Kaushik, 1998), if not supported by appropriate feeding strategies.

Feeding management, particularly the feeding frequency, is a critical but frequently under-optimized component of aquaculture nutrition (Bureau and Hua, 2010; Brinker and Rösch, 2005). Feeding frequency directly influences the feed intake (FI) and the digestive efficiency (Riche et al., 2004), nutrient assimilation (Kaushik, 1998), growth performance, and waste production (Deng et al., 2003; Liu et al., 2023; Wang et al., 2007). Inadequate feeding frequency may lead to reduced growth (Liu et al., 2022), increased size heterogeneity (Campeas et al., 2009), and poor feed utilization (Bureau and Hua, 2010; Liu et al., 2018), whereas excessive feeding frequency can increase the feed conversion ratios (FCRs) (Liu et al., 2022) and elevate the dissolved nutrient losses through metabolic excretion (Cho and Bureau, 1997; Cho and Bureau, 1998; Cho and Bureau, 2001). Consequently, identifying feeding frequencies that maximize nutrient retention while minimizing waste output is central to sustainable aquaculture intensification.

Largemouth bass (Micropterus salmoides) is a freshwater carnivorous species of high commercial and recreational value (Biswas et al., 2010). Largemouth bass was selected due to its rapid growth (Victor et al., 2024), high protein requirement, and increasing importance in intensive aquaculture systems, making it an appropriate model species for the evaluation of feeding management strategies (Ren et al., 2017). Previous studies have reported that feeding largemouth bass two to three times daily can enhance growth performance under specific rearing conditions (Liu et al., 2022). However, the majority of existing studies have evaluated feeding frequency in isolation, without explicitly accounting for the dietary composition. In particular, the interactive effects between the fishmeal inclusion level and the feeding frequency on nutrient utilization and waste output remain poorly understood.

Understanding these interactions is essential as the feeding frequency may partially compensate for reduced dietary fishmeal by improving the FI patterns and the digestive efficiency, but may simultaneously increase nutrient losses and environmental loading. Therefore, an integrated assessment that simultaneously considers growth performance, nutrient retention, energy partitioning, and waste output is required to inform practical and sustainable feeding strategies.

The present study was designed to evaluate the combined effects of dietary fishmeal level and feeding frequency on the growth performance, feed utilization, whole body composition, nutrient (N, P, and C) and energy budgets, and solid and dissolved waste outputs in juvenile largemouth bass reared under a flow-through system. Because the feeding frequency regulates the nutrient intake dynamics while the fishmeal level determines nutrient quality, their interaction may strongly influence the nutrient retention efficiency and waste outputs. However, this interaction remains poorly quantified. Specifically, this study aimed to 1) determine the optimal feeding frequency under high- and low-fishmeal diets; 2) assess whether an increased feeding frequency can mitigate the growth reductions associated with a lower fishmeal inclusion; and 3) quantify how these factors interact to influence nutrient retention efficiency and environmental waste production. Thus, optimize feeding strategies that balance productivity, cost, and environmental sustainability in carnivorous fish aquaculture.

2 Materials and methods

2.1 Experimental site and ethics statement

The experiment was conducted at a commercial aquaculture facility at Hengze commercial aquaculture and ecological farm (Hangzhou, China). All experimental and sampling procedures were conducted following the Chinese law on experimental animals and were approved by Zhejiang University, China.

2.2 Experimental design and diet preparation

A two-factorial experiment including two fishmeal levels [350 g kg⁻¹, high fishmeal (HFM) and 210 g kg⁻¹, low fishmeal (LFM)] and three feeding frequencies [once a day (F1), twice a day (F2), and three times a day (F3)] were used. Six treatments (HFMF1, HFMF2, HFMF3, LFMF1, LFMF2, and LFMF3) were examined. A concentration of 1% of chromium trioxide (Cr₂O₃) was added in both test diets as an inert marker to measure digestibility. The proximate composition (in grams per kilogram) of the feed ingredients is shown in Table 1. The diets were prepared to have the same protein and lipid contents, with 500 g kg⁻¹ crude protein and 90 g kg⁻¹ crude lipid. The feed ingredients were first mixed by hand and then thoroughly mixed in a mechanical mixer. Pellets with a diameter of 3 mm and a length of 5 mm were subsequently extruded. The diets were dried at 24°C in an air-conditioned room and stored in plastic bags at −20°C until use.

Table 1. Formulation (g kg⁻¹), proximate composition (g kg⁻¹), and energy content (MJ kg⁻¹) of the test diets expressed as dry matter.

HFM, high fishmeal; LFM, low fishmeal.

Vitamin and mineral premix (VMP): vitamin A, 8,000 IU; vitamin D₃, 2,000 IU; vitamin E, 100 mg; vitamin K₃, 7.5 mg; vitamin B₁, 15 mg; vitamin B₂, 15 mg; vitamin B₆, 12.5 mg; vitamin B₁₂, 0.05 mg; d-biotin, 0.25 mg; d-calcium pantothenate, 40 mg; folic acid, 5 mg; niacinamide, 50 mg; vitamin C, 140 mg; inositol, 120 mg; ethoxyquin, 5 mg; FeSO₄, 40 mg; CuSO₄·5H₂O, 25 mg; MnSO₄·4H₂O, 10 mg; ZnSO₄, 100 mg; MgSO₄·7H₂O, 200 mg; CoCO₃, 0.35 mg; KI, 0.05 mg; and Na₂SeO₃, 0.3 mg.

Ingredients, crude protein, crude lipid, ash, gross energy, and phosphorus are expressed on a dry matter basis. Crude protein, crude lipid, ash, gross energy, and phosphorus are presented as the means of two determinations (n = 2).

2.3 Experimental setting and culture conditions

Juvenile largemouth bass were obtained from Hengze commercial and ecological farm (Hangzhou, China). Before commencing the experiment, fish were reared in in-pond raceways for 2 weeks. During this time, the fish were fed with a commercial feed. At the start of the experiment, 18 tanks (volume, 300 L/tank) were installed with a flow-through system receiving water from a large pond into the tanks. A total of 360 juvenile largemouth bass (unfed for 24 h) with similar body size were randomly distributed among the 18 fish rearing tanks (20 fish/tank). Fish had an initial body weight of 51.0 ± 0.12 g. Each rearing tank was assigned randomly to one of the feeding frequencies in both diets, forming triplicates per feeding frequency in both test diets. The tanks were stationed permanently to avoid deviation or potential mix-up of the treatments. The experimental period was limited to 5 weeks instead of the initially planned 8 weeks to minimize the risk of data loss associated with potential damage to the rearing facilities caused by an approaching typhoon. Previous studies have demonstrated that the growth performance and nutrient balance responses in juvenile largemouth bass can be reliably detected within similar experimental periods (Xia et al., 2023). Thus, a 5-week feeding trial was sufficient to obtain reliable data. In the experiment, the culture conditions and water quality parameters were usually optimal for largemouth bass. The water temperature, dissolved oxygen (DO), and pH were measured daily in the morning and evening and ranged from 20.5°C to 24°C, from 6.80 to 7.20 mg L⁻¹, and from 7.2 to 7.4, respectively. Total ammonia nitrogen was always less than 0.5 mg L⁻¹ throughout the experiment. The water flow in each tank was maintained between 100 and 150 ml s⁻¹ and was allowed to flow continuously, but was stopped during feeding.

The fish were fed according to the experimental design (fed once a day at 6:00 a.m., twice a day at 6:00 a.m. and 5:00 p.m., or three times a day at 6:00 a.m., 12:00 p.m., and 5:00 p.m., for the two test diets). The fish were hand-fed until apparent satiation, and satiation was achieved when no fish was actively feeding at the surface or the bottom of the water. Uneaten feed was minimized through careful observation to avoid overfeeding bias among the feeding frequency treatments. Moreover, the feeding ration was adjusted based on fish appetite and environmental condition. Uneaten feed and feces were collected 2 h after feeding.

2.4 Fecal collection and chromium oxide determination

Feces were collected 2 h post-feeding following established protocols, allowing sufficient gastric evacuation while minimizing nutrient leaching (Cho and Kaushik, 1990; Gilannejad et al., 2019). Only intact feces were collected. The collection bottle was kept under ice to minimize bacterial decomposition of feces during collection. The feces were immediately homogenized and frozen at −20°C before subsequent analysis. Feces collection began 2 days after the start of the experiment. The collection lasted until the end of the experiment. It was made sure that the feces were sufficient for subsequent analysis (40 g). The feed and feces were analyzed for the presence of the inert indicator (Cr₂O₃). The contents of Cr₂O₃ in the diets and the feces were analyzed as described in Furukawa and Tsukahara (1966). Approximately 200 mg of the sample was weighed and transferred into digestion tubes. The digestion of samples was in two phases. In phase 1, 5 ml concentrated nitric acid (HNO₃) was added to the digestion tubes. The tubes were placed in a heating digestion chamber for approximately 30 min for solubilization of all organic material. During phase 2, 3 ml of perchloric acid (HClO₄) was added to the digestion tubes and then put back in the heating chamber for 30 min. After cooling to ambient temperature, 10 ml of deionized water was added to the tubes. The concentration of chromic oxide was then measured at 350 nm in a spectrometer.

2.5 Chemical analysis

The sampled fish were autoclaved at 120°C for 20 min, cooled, homogenized with a laboratory grinder, and dried at 105°C. The moisture, crude protein, crude lipid, ash, and phosphorus contents of the fish samples, diets, and feces were analyzed using the AOAC methods (AOAC, 2005), in particular methods 934.01 (moisture), 990.03 (protein), 920.39 (lipid), and 942.05 (ash). The energy content was analyzed using a bomb calorimeter (Parr 1281, Moline, IL, USA), while the carbon and nitrogen contents were analyzed using EA Elemental Analyzer (EuroEA3000, EuroVector, Pavia, Italy).

2.6 Calculations

At the end of the feeding trial, all fish in each replicate tank were individually weighed to determine the final body weight (FBW), and total length was measured for each fish with a measuring ruler. Five fish were randomly selected from each tank for the evaluation of organ indices. These fish were individually dissected to collect the liver and viscera, which were weighed to calculate the hepatosomatic index (HSI) and the viscerosomatic index (VSI). The condition factor (K) was calculated using the measured body weight and corresponding total length. Weight gain (WG) was calculated as the difference between the initial body weight and the FBW. Relative survival (in percent) was determined for each treatment. FI was recorded daily throughout the experimental period. The nitrogen retention efficiency (NRE), the phosphorus retention efficiency (PRE), and the carbon retention efficiency (CRE), together with other growth, nutrient utilization, nutrient budget, and waste output parameters, were calculated using the equations described below.

Relative percent survival (%)=(Initial number of fish Final number of fish)×100

K (g/cm3) =Body weight (g)Total length 3 (cm3)

HSI (%)=[Liver weight (g)Individual fish weight (g)]×100

VSI (%)=[Viscera weight (g)Individual fish weight (g)]×100

Feed intake (% d−1)= [(Total feed fed dailyTime × (Final + Initial fish number2))((Final fish weight Final fish weight )+(Initial fish weight Initial fish weight )/2)]× 100

WG (g fish−1)=[(Final fish weight (g)Final fish number)− (Initial fish weight (g)Initial fish number)]

FCR=Total feed intake (g)[(Final −Initial fish weight)+Weight of dead fish] (g)

NRE (%)=[(Final fish weight×Final fish nitrogen contentFinal fish number)−(Initial fish weight×Initial fish nitrogen contentInitial fish number)(2×Feed intake×Nitrogen content in the dietFinal fish number+Initial fish number)]×100

PRE (%)=[(Final fish weight×Final fish phosphorus contentFinal fish number)−(Initial fish weight × Initial fish phosphorus contentInitial fish number)(2×Feed intake×Phosphorus content in the dietFinal fish number+Initial fish number)]×100

CRE (%) =[(Final fish weight×Final fish carbon contentFinal fish number)−(Initial fish weight×Initial fish carbon contentInitial fish number)(2×Feed intake×Carbon content in the dietFinal fish number+Initial fish number)]×100

ERE(%) =[(Final fish weight×Final fish energy contentFinal fish number)−(Initial fish weight×Initial fish energy contentInitial fish number)(2×Feed intake×Energy content in the dietFinal fish number+Initial fish number)] ×100

The apparent digestibility coefficients (ADCs; in percent) of dry matter (DM), N, C, P, and energy of the diets were calculated as described by Cho et al. (1982) and De La Noüe and Choubert (1986).

ADC of DM (%) =[1−(Marker in the feed)(Marker in the feces)]×100

ADC of Nutrient (N,P and C as %)=[1−(Marker in the feedMarker in the feces)×(Nutrient in fecesNutrient in the feed)]×100

The N, P, C, and energy balance values were calculated per tank and expressed as grams N per kilogram of fish produced, grams P per kilogram of fish produced, grams C per kilograms of fish produced, and kilojoules per kilogram of fish produced, respectively, and were calculated as follows:

Gross nitrogen intake (GN)=Total feed intake (g)×Nitrogen content in the diet

Gross phosphorus intake (GP)=Total feed intake (g)×Phosphorus content in the diet

Gross carbon intake (GC)=Total feed intake (g)×Carbon content in the diet

Gross energy intake (GE)=Total feed intake (g)×Energy content in the diet

Digestible nitrogen (DN)= GN×Apparent digestibility coefficient of crude protein (%)100

Digestible phosphorus (DP)=GP×Apparent digestibility coefficient of phosphorus (%)100

Digestible carbon (DC)=GC×Apparent digestibility coefficient of carbon (%)100

Digestible energy (DE)=GE×Apparent digestibility coefficient of energy (%)100

Fecal nitrogen losses (FN)=GN−DN

Fecal phosphorus losses (FP)=GP−DP

Fecal carbon losses (FC)=GC−DC

Fecal energy losses (FE)=GE−DE

Retained nitrogen (RN)=[(Final fish weight × Final fish nitrogen content6.25)−(Initial fish weight × Initial fish nitrogen content6.25)]

Retained phosphorus (RP)=[(Final fish weight×Final fish phosphorus content)−(Initial fish weight×Initial fish phosphorus content)]

Retained carbon (RC)=[(Final fish weight×Final fish carbon content)−(Initial fish weight×Initial fish carbon content)]

Retained energy (RE)=[(Final fish weight×Final fish energy content)−(Initial fish weight×Initial fish energy content)]

Branchial and urinary nitrogen losses (BUN)=DN−RN

Branchial and urinary phosphorus losses (BUP)=DP−RP

Branchial and urinary carbon losses (BUC)=DC−RC

Branchial and urinary energy losses (BUE)=BUN×24.91000

Where 24.9 kJ N g⁻¹ is the energy concentration of NH₃–N (Bureau et al., 2003).

Metabolizable energy (ME)=DE−BUE

The nutrient waste output was calculated according to Bureau et al. (2003). Apparent feed waste (AFW) was assumed constant based on established nutrient budget models (Cho and Bureau, 1998), as follows:

Total solid waste (TSW)=[DM consumed by the fish ×(1

−ADC of DM of feed)]+(Apparent feed waste (AFW, dry basis)]

Solid (N or P or C) waste (SNW/SPW/SCW)=[N or P or C consumed×(1−ADC of N or P or C of feed)]+N or P or C in AFW

Dissolved (N or P or C) waste (DNW/DPW/DCW)=(N or P or C consumed×ADC of N or P or C of feed) − N or P or C retained by fish=(N or P or C consumed×ADC of N or P or C of feed)−N or P or C retained by fish

AFW was recorded to be 5% based on the collected waste feeds. The dissolved N, P, and C and the retained nitrogen (RN), retained carbon (RC), and retained phosphorus (RP) were calculated according to Tran-Ngoc et al. (2019).

2.7 Statistical analysis

Statistical analyses were performed using IBM SPSS 26 (SPSS Inc., Chicago, IL, USA). Digestibility was subjected to one-way ANOVA. The growth parameters, body composition, waste outputs, and nutrient budget were analyzed using two-way ANOVA. The results were considered statistically significant when the p-value was less than 0.05. Tukey’s test was applied for pairwise comparisons of means, where appropriate. Tukey’s post-hoc test was then utilized for specific comparisons on significant differences among the experimental treatments at a significance level of p < 0.05.

3 Results

3.1 Digestibility

The ADCs of DM, nutrients, and energy in the experimental diets are shown in Table 2. The digestibility of DM, protein, and energy in the HFM diet was higher than that in the LFM diet (p < 0.05). The digestibility of phosphorus and carbon was higher in the LFM diet than in the HFM diet. For the HFM and LFM diets, the digestibility values were, respectively, 75.23% and 74.60% for DM, 88.17% and 86.30% for protein, 90.10% and 92.40% for phosphorus, 74.40% and 77.13% for carbon, and 90.11% and 89.16% for energy.

Table 2. Digestibility (in percent) of the test diets formulated with different levels of fish meal.

HFM, high fishmeal; LFM, low fishmeal.

3.2 Growth performance

Feeding fish under the HFMF1, HFMF2, HFMF3, LFMF1, LFMF2, and LFMF3 treatments resulted in relative percentage survival of 100%, 95%, 100%, 100%, 91%, 85%, and 95%, respectively, with significantly lower relative percentage survival observed in LFMF2 (p < 0.05). Two-way ANOVA indicated that the fishmeal level, feeding frequency, and their interaction significantly affected FCR and FI (p < 0.05) (Table 3). FBW, WG, and PRE were significantly influenced by both the fishmeal level and feeding frequency, whereas NRE and CRE were affected only by the fishmeal level (p < 0.05) (Table 3).

Table 3. Growth, feed intake, feed utilization, and survival of largemouth bass at two fishmeal levels and three feeding frequencies.

Treatments consisted of high fishmeal (HFM; 350 g kg⁻¹ fishmeal) and low fishmeal (LFM; 210 g kg⁻¹ fishmeal) combined with feeding frequencies F1 (once a day, 0600 hours), F2 (twice a day, 0600 and 1700 hours), and F3 (three times a day, 0600, 1200, and 1700 hours), forming six treatments: HFMF1, HFMF2, HFMF3, LFMF1, LFMF2, and LFMF3. Data are presented as the mean ± SD (n = 3). Within a column, values not sharing the same letter differ significantly according to Duncan’s multiple range test (p < 0.05). Lowercase letters indicate differences among treatments, uppercase letters indicate differences between dietary fishmeal levels, and Greek letters indicate differences among feeding frequencies.

IBW, initial body weight; FBW, final body weight; WG, weight gain; FI, feed intake; FCR, feed conversion ratio; NRE, nitrogen retention efficiency; PRE, phosphorus retention efficiency; ERE, energy retention efficiency; FM, fishmeal; FF, feeding frequency.

Fish fed the HFM diet (350 g kg⁻¹) exhibited higher FBW, WG, NRE, PRE, and CRE compared with those fed the LFM diet (210 g kg⁻¹; p < 0.05). Increasing the feeding frequency improved growth performance, with fish fed two or three meals per day (F2 and F3, respectively) showing greater FBW and WG than those fed once a day (F1; p < 0.05). However, NRE was higher in fish fed once or twice per day compared with those fed three times a day (p < 0.05).

Among treatments, HFMF3 produced the highest FBW, WG, and FI, whereas a poorer feed efficiency (higher FCR) was observed in LFMF1 and LFMF3. PRE was generally higher under HFM treatments, while CRE was not significantly affected by treatment (p > 0.05). Detailed pairwise comparisons are presented in Table 3.

3.3 Whole body composition and nutrient retention

The fishmeal level and feeding frequency significantly affected the energy content (two-way ANOVA: p < 0.05) (Table 3). The fishmeal level significantly affected HSI and VSI (two-way ANOVA: p < 0.05) (Table 4). The feeding frequency significantly affected the phosphorus content (two-way ANOVA: p < 0.05) (Table 3). However, K, moisture, crude protein, crude lipid, ash, and carbon were independent of the fishmeal level, feeding frequency, and their interaction (two-way ANOVA: p > 0.05). Fish cultured on 350 g kg⁻¹ fishmeal level (HFM) had higher HSI and VSI compared with those cultured on 210 g kg⁻¹ fishmeal level (LFM; p < 0.05). Fish cultured on treatment LFMF3 had significantly higher energy content (p < 0.05) than those cultured on HFMF1, HFMF2, HFMF3, and LFMF1, but not those cultured on LFMF2 (p > 0.05).

Table 4. Body composition of largemouth bass at two fishmeal levels and three feeding frequencies.

Treatments included high fishmeal (HFM; 350 g kg⁻¹ fishmeal) and low fishmeal (LFM; 210 g kg⁻¹ fishmeal) combined with the feeding frequencies F1 (once a day, 0600 hours), F2 (twice a day, 0600 and 1700 hours), and F3 (three times a day, 0600, 1200, and 1700 hours), forming six treatments: HFMF1, HFMF2, HFMF3, LFMF1, LFMF2, and LFMF3. Crude protein, crude lipid, phosphorus, ash, and gross energy are expressed on a wet weight basis. Data are presented as the mean ± SD (n = 3). Within a column, values not sharing the same differ significantly according to Duncan’s multiple range test (p < 0.05). Lowercase letters indicate differences among treatments, uppercase letters indicate differences between dietary fishmeal levels, and Greek letters indicate differences among feeding frequencies.

K, condition factor; HSI, hepatosomatic index; VSI, viscerosomatic index; FM, fishmeal; FF, feeding frequency.

3.4 Nitrogen, phosphorus carbon, and energy balance

The fishmeal level, feeding frequency, and their interaction significantly affected fecal nitrogen loss (FN) (two-way ANOVA: p < 0.05) (Table 5). The fishmeal level and feeding frequency significantly affected the gross nitrogen intake (GN), digestible nitrogen (DN), RN, and branchial and urinary nitrogen losses (BUN) (two-way ANOVA: p < 0.05) (Table 5). The fishmeal level significantly affected RN/DN (two-way ANOVA: p < 0.05) (Table 5). Fish cultured at 350 g kg⁻¹ fishmeal level (HFM) had higher GN, DN, RN, and RN/DN, but lower BUN, than those cultured on 210 g kg⁻¹ fishmeal level (LFM; p < 0.05). Moreover, fish fed three times per day (F3) had higher GN, DN, RN, and BUN than those on the other feeding frequencies (p < 0.05). The interaction effect indicated that a higher feeding frequency intensified the N fluxes under both fishmeal levels.

Table 5. Nitrogen balance of largemouth bass at two fishmeal levels and three feeding frequencies.

Treatments included high fishmeal (HFM; 350 g kg⁻¹ fishmeal) and low fishmeal (LFM; 210 g kg⁻¹ fishmeal) combined with feeding frequencies F1 (once a day, 0600 hours), F2 (twice a day, 0600 and 1700 hours), and F3 (three times a day, 0600, 1200, and 1700 hours), forming six treatments: HFMF1, HFMF2, HFMF3, LFMF1, LFMF2, and LFMF3. Data are presented as the mean ± SD (n = 3). Within a column, values not sharing the same letter differ significantly according to Duncan’s multiple range test (p < 0.05). Lowercase letters indicate differences among treatments, uppercase letters indicate differences between dietary fishmeal levels, and Greek letters indicate differences among feeding frequencies.

GN, gross nitrogen intake; DN, digestible nitrogen; FN, fecal nitrogen losses; RN, retained nitrogen; BUN, branchial and urinary nitrogen losses; FM, fishmeal; FF, feeding frequency.

The fishmeal level, feeding frequency, and their interaction significantly affected digestible phosphorus (DP; two-way ANOVA: p < 0.05) (Table 6). The fishmeal level and feeding frequency significantly affected gross phosphorus intake (GP), fecal phosphorus losses (FP), RP, branchial and urinary phosphorus losses (BUP), and RP/DP (two-way ANOVA: p < 0.05) (Table 6). Fish fed 350 g kg⁻¹ fishmeal level (HFM) had lower GP and BUP, but higher FP, RP, and RP/DP, than those fed 210 g kg⁻¹ fishmeal level (LFM; p < 0.05). Moreover, fish fed three times per day (F3) had higher GP, FP, RP, and BUP, but lower RP/DP, than those on the other feeding frequencies (p < 0.05). The interaction indicated that an increased feeding frequency enhanced the P turnover irrespective of the dietary fishmeal level.

Table 6. Phosphorus balance of largemouth bass at two fishmeal levels and three feeding frequencies.

Treatments included high fishmeal (HFM; 350 g kg⁻¹ fishmeal) and low fishmeal (LFM; 210 g kg⁻¹ fishmeal) combined with feeding frequencies F1 (once a day, 0600 hours), F2 (twice a day, 0600 and 1700 hours), and F3 (three times a day, 0600, 1200, and 1700 hours), forming six treatments: HFMF1, HFMF2, HFMF3, LFMF1, LFMF2, and LFMF3. Data are presented as the mean ± SD (n = 3). Within a column, values not sharing the same letter differ significantly according to Duncan’s multiple range test (p < 0.05). Lowercase letters indicate differences among treatments, uppercase letters indicate differences between dietary fishmeal levels, and Greek letters indicate differences among feeding frequencies.

GP, gross phosphorus intake; DP, digestible phosphorus; FP, fecal phosphorus losses; RP, retained phosphorus; BUP, branchial and urinary phosphorus losses; FM, fishmeal; FF, feeding frequency.

The fishmeal level, feeding frequency, and their interaction significantly affected the fecal carbon losses (FC) (two-way ANOVA: p < 0.05) (Table 7). The fishmeal level and feeding frequency significantly affected the gross carbon intake (GC), digestible carbon (DC), RC, and the branchial and urinary carbon losses (BUC; two-way ANOVA: p < 0.05) (Table 7). The fishmeal level significantly affected RC/DC (two-way ANOVA: p < 0.05) (Table 7). Fish cultured at 350 g kg⁻¹ fishmeal level (HFM) had higher GC, DC, RC, and RC/DC, but lower BUC, than those cultured on 210 g kg⁻¹ fishmeal level (LFM; p < 0.05). Moreover, fish fed three times per day (F3) had higher GC, DC, RC, and BUC than those on the other feeding frequencies (p < 0.05). The interaction effect showed that the feeding frequency amplified the C fluxes under both fishmeal levels.

Table 7. Carbon balance of largemouth bass at two fishmeal levels and three feeding frequencies.

Treatments included high fishmeal (HFM; 350 g kg⁻¹ fishmeal) and low fishmeal (LFM; 210 g kg⁻¹ fishmeal) combined with feeding frequencies F1 (once a day, 0600 hours), F2 (twice a day, 0600 and 1700 hours), and F3 (three times a day, 0600, 1200, and 1700 hours), forming six treatments: HFMF1, HFMF2, HFMF3, LFMF1, LFMF2, and LFMF3. Data are presented as the mean ± SD (n = 3). Within a column, values not sharing the same letter differ significantly according to Duncan’s multiple range test (p < 0.05). Lowercase letters indicate differences among treatments, uppercase letters indicate differences between dietary fishmeal levels, and Greek letters indicate differences among feeding frequencies.

GC, gross carbon intake; DC, digestible carbon; FC, fecal carbon losses; RC, retained carbon; BUC, branchial and urinary carbon losses; FM, fishmeal; FF, feeding frequency.

The fishmeal level, feeding frequency, and their interaction significantly affected the fecal energy losses (FE) and the retained energy (RE)/digestible energy (DE) (two-way ANOVA: p < 0.05) (Table 8). The fishmeal level and the feeding frequency significantly affected the gross energy intake (GE), DE, metabolizable energy (ME), RE, and the branchial and urinary energy losses (BUE; two-way ANOVA: p < 0.05) (Table 8). Fish cultured at 350 g kg⁻¹ fishmeal level (HFM) had higher GE, DE, ME, and RE, but lower BUE, than those cultured at 210 g kg⁻¹ fishmeal level (LFM; p < 0.05). Moreover, fish fed three times per day (F3) had higher GE, DE, ME, RE, and BUE than those on the other feeding frequencies (p < 0.05). The interaction indicated that a higher feeding frequency enhanced the overall energy throughput regardless of the dietary fishmeal level.

Table 8. Energy balance of largemouth bass at two fishmeal levels and three feeding frequencies.

Treatments included high fishmeal (HFM; 350 g kg⁻¹ fishmeal) and low fishmeal (LFM; 210 g kg⁻¹ fishmeal) combined with feeding frequencies F1 (once a day, 0600 hours), F2 (twice a day, 0600 and 1700 hours), and F3 (three times a day, 0600, 1200, and 1700 hours), forming six treatments: HFMF1, HFMF2, HFMF3, LFMF1, LFMF2, and LFMF3. Data are presented as the mean ± SD (n = 3). Within a column, values not sharing the same letter differ significantly according to Duncan’s multiple range test (p < 0.05). Lowercase letters indicate differences among treatments, uppercase letters indicate differences between dietary fishmeal levels, and Greek letters indicate differences among feeding frequencies.

GE, gross energy intake; DE, digestible energy; ME, metabolizable energy; FE, fecal energy losses; RE, retained energy; BUE, branchial and urinary energy losses; FM, fishmeal; FF, feeding frequency.

3.5 Solid and dissolved waste outputs of nitrogen, phosphorus, and carbon

The fishmeal level, feeding frequency, and their interaction significantly affected the solid nitrogen waste (SNW; two-way ANOVA: p < 0.05) (Figure 1). The solid carbon waste (SCW) and solid phosphorus waste (SPW) were significantly influenced by both fishmeal level and feeding frequency, whereas the total solid waste (TSW) was significantly affected only by the feeding frequency (p < 0.05) (Figure 1). The interaction effect indicated that increasing the feeding frequency resulted in progressively higher solid waste outputs under both fishmeal levels, with the greatest values observed in the HFMF3 and LFMF3 treatments.

The fishmeal level and feeding frequency significantly affected the dissolved nitrogen waste (DNW) and dissolved phosphorus waste (DPW), while the dissolved carbon waste (DCW) was influenced only by the fishmeal level (two-way ANOVA: p < 0.05) (Figure 2). Similarly, their interaction showed that a higher feeding frequency increased the dissolved nutrient losses irrespective of the dietary fishmeal level, leading to the highest DNW and DPW values in the HFMF3 and LFMF3 treatments, whereas DCW was not significantly affected by treatment (p > 0.05).

4 Discussion

The present study provides a comprehensive evaluation of how dietary fishmeal level and feeding frequency interact to influence the growth performance, nutrient utilization, nutrient budget, and waste output in juvenile largemouth bass. By integrating digestibility, nutrient balance, and waste partitioning analyses, the findings offer valuable insights into both the biological and environmental consequences of feeding management strategies in carnivorous fish culture.

4.1 Digestibility and dietary effects

The digestibility for DM, protein, and energy was significantly lower for the LFM diet and high for the HFM diet. The higher ADCs (in percent) of DM, protein, and energy observed in the HFM diet compared with the LFM diet are consistent with previous reports highlighting the superior digestibility of fishmeal-based diets (Zhou and Yue, 2012; Tran-Ngoc et al., 2019; Hien et al., 2017). Fishmeal provides highly available protein and lipid fractions (Bureau and Hua, 2010), which enhance the digestive efficiency and nutrient absorption (Zhou and Yue, 2012). Conversely, partial replacement of fishmeal with a poultry by-product meal in the LFM diet altered the ingredient composition and amino acid balance, contributing to the reduced protein digestibility (Ren et al., 2017). The feeding frequency did not significantly affect nutrient digestibility, confirming earlier findings that digestibility is primarily determined by the diet composition rather than the feeding regime (Cho and Kaushik, 1990; Wang et al., 2007). Thus, feeding frequency influences nutrient utilization primarily through intake dynamics and metabolic partitioning rather than through changes in the digestive capacity (Cho and Kaushik, 1990; Wang et al., 2007).

4.2 Growth performance and feed utilization

In the present study, both the feeding frequency and the fishmeal levels significantly affected the growth performance and feed utilization of the fish. WG was enhanced significantly by increasing the feeding frequency across both fishmeal levels, with fish fed two or three times daily exhibiting higher FBW and WG than those fed once daily. These results align with previous studies demonstrating that an increased feeding frequency improves feed accessibility and reduces competitive exclusion, particularly in aggressive carnivorous species (Wang et al., 1998; Lee et al., 2000; Dwyer et al., 2002; Sun et al., 2016). On the other hand, fish fed the HFM diet consistently outperformed those fed the LFM diet, reflecting the nutritional advantages of a higher fishmeal inclusion. Nevertheless, increasing the feeding frequency partially compensated for the reduced growth observed under the LFM diet, indicating that feeding management can mitigate, although not fully eliminate, the performance penalties associated with fishmeal reduction. This finding has important implications for feed cost optimization, suggesting that moderate fishmeal reduction may be feasible when supported by appropriate feeding regimes (Sun et al., 2016). Despite the higher FBW and WG observed at three meals per day, the FCR did not improve proportionally and, in some cases, deteriorated, particularly under the LFM diet. The FCRs were observed to be 0.93 and 0.95 for two and three meals per day, respectively, when fish were fed HFM and were 0.99 and 1.23 for two and three meals per day, respectively, when fish were fed LFM. This suggests diminishing returns at higher feeding frequencies (Tsevis et al., 1992; Sun et al., 2016), likely due to increased metabolic losses and inefficient nutrient utilization. From both economic and ecological perspectives, feeding twice daily emerged as the most balanced strategy, providing high growth with improved feed efficiency.

4.3 Fish composition, nutrient retention, and metabolic partitioning

Fish body composition indicates fish quality and has been reported to be affected by the quality of feed and the feeding frequency (Liu et al., 2023, 2022). In this study, fish composition was not affected by the feeding frequency and the fishmeal levels, which is consistent with Kayano et al. (1993) and Lee et al. (2000), respectively. The feeding frequency affected phosphorus and gross energy, while the fishmeal level affected HSI, VSI, and gross energy. The N, P, and C retention efficiencies were consistently higher in fish fed the HFM diet, reflecting improved protein quality and nutrient availability. In contrast, the lower retention efficiencies under the LFM diet likely resulted from imbalanced amino acid profiles and the increased reliance on protein catabolism for energy (Tidwell et al., 2005; Tran-Ngoc et al., 2019). Increasing the feeding frequency led to higher absolute nutrient intake and retention; however, the retention efficiencies declined with increasing feeding frequency. The elevated BUN observed at three meals per day indicate increased amino acid oxidation and protein use as an energy source, rather than for somatic growth (Tran-Ngoc et al., 2019). Similar trends have been reported in other species, where excessive feeding intensity increases the metabolic excretion despite higher growth rates (Azevedo et al., 1998; Wang et al., 2007). These findings highlight a critical trade-off that a higher feeding frequency promotes growth, but also accelerates nutrient turnover and metabolic waste production, thereby reducing the overall nutrient use efficiency.

4.4 Waste output and environmental implications

Waste output is closely related to feed composition (Wang et al., 2007; Yuan et al., 2019), digestibility (Zhou and Yue, 2012), feeding practice (Cho and Bureau, 1997; Dauda et al., 2019; Luo et al., 2018), feed ration (Kokou and Fountoulaki, 2018; Zhang et al., 2011), and feeding frequency (Azevedo et al, 1998). Waste output analysis revealed that feeding frequency was the dominant driver of both solid and dissolved nutrient wastes (Luo et al., 2018; White, 2013). TSW, SNW, DNW, and phosphorus wastes increased markedly at higher feeding frequencies, regardless of the dietary fishmeal level. This observation is consistent with bioenergetic models linking increased FI (Kaushik, 1998) to proportional increases in fecal production and metabolic excretion (Cho and Bureau, 1998; Bureau et al., 2003; Fernandes et al., 2007). The dietary fishmeal level influenced the form rather than the magnitude of waste output. The HFM diet produced higher solid waste but lower dissolved nutrient losses (Yu et al., 2019), whereas the LFM diet generated comparatively lower solid waste but higher dissolved nitrogen and phosphorus outputs (Luo et al., 2018). Dissolved wastes are of particular environmental concern due to their direct bioavailability and contribution to eutrophication (Hua et al., 2008; Janes-Bassett et al., 2020; Luo et al., 2018; Montanhini Neto and Ostrensky, 2013). Although a higher feeding frequency enhanced growth, it substantially increased the total nutrient discharge per unit of fish produced. Therefore, from an environmental sustainability standpoint, feeding twice daily represents the most favorable compromise, minimizing waste output while maintaining high productivity.

4.5 Overall implications for sustainable feeding strategies

Collectively, the results demonstrated strong interactive effects between the dietary fishmeal level and the feeding frequency on growth, nutrient utilization, and waste production. While a higher fishmeal inclusion and an increased feeding frequency enhanced growth performance, they also intensified the nutrient loading and metabolic losses. Importantly, the study showed that the feeding frequency can be strategically adjusted to partially offset the reductions in dietary fishmeal; however, excessive feeding intensity undermines nutrient efficiency and environmental performance. Based on biological, economic, and ecological considerations, a feeding frequency of two meals per day, combined with moderate to high fishmeal inclusion, represents the most sustainable feeding strategy for juvenile largemouth bass reared in flow-through systems. Although tank replication followed standard nutritional trial practices, additional replication could further capture variability under commercial conditions. These findings provide a robust framework for optimizing feeding management and contribute to broader efforts aimed at reducing the environmental footprint of intensive aquaculture.

5 Conclusion

In summary, this study evaluated the effects of feeding frequency and dietary fishmeal content on feed digestibility, growth performance, whole body composition, waste output, and the balance of nitrogen, phosphorus, carbon, and energy. Based on the current results, we therefore conclude that 350 g kg⁻¹ of fishmeal in a diet and a feeding frequency of two meals per day are still the best option to sustainably enhance growth performance and improve the balance of N, C, P, and energy while reducing waste outputs in largemouth bass farming. The findings of this study therefore shed light and increase awareness for further studies on the effects of feed and its ingredients and the feeding regimes on waste and the balance of nitrogen, phosphorus, and carbon in the cause of waste management for a more sustainable aquaculture.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. The author declares that generative artificial intelligence (AI) was used during the preparation of this manuscript. Specifically, the corresponding author used ChatGPT 5.2 to assist with language editing and improving clarity and readability. The author reviewed, revised, and approved the final content and takes full responsibility for the accuracy and integrity of the work.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Gouranga Biswas, Central Institute of Fisheries Education (ICAR), India

Reviewed by: Nam Hoang Tran, Tokushima University, Japan

Nhien Nguyen Thi, Vietnam National University of Agriculture, Vietnam