Biofloc Formation Strategy Effects on Halophyte Integration in IMTA with Marine Shrimp and Tilapia

Abstract

The incorporation of aquaponics into saline integrated multitrophic aquaculture (IMTA) systems, employing biofloc technology (BFT), relies on the cultivation of halophytes capable of withstanding the physical–chemical conditions created by the unique microbial communities in BFT systems. This study aimed to evaluate the integration of the halophyte Salicornia neei with tilapia (Oreochromis niloticus) and marine shrimp (Litopenaeus vannamei) reared in BFT systems dominated by chemoautotrophic (CHE) and heterotrophic (HET) microorganisms over a period of 84 days in southern Brazil. Each BFT treatment had three replicates, composed of IMTA units. The stocking densities were 400 ind. m⁻³ (17 m³ circular tanks), 44 ind. m⁻³ (4 m³ circular tanks), and 30 ind. m⁻² (4.8 m² hydroponic benches) for shrimp, fish, and halophyte, respectively. The highest S. neei individual shoot production (up to 31 g per 30 days) was observed in the CHE treatment, along with favorable agronomic characteristics, possibly due to overall elevated nitrate (98.41 mg N−NO₃ L⁻¹) and phosphate concentrations (4.62 P−PO₄ L⁻¹). Shrimp in the CHE treatment displayed higher average final weight, specific growth rate, productivity, and survival (11.24 g, 2.88% day⁻¹, 3.86 kg m⁻³, and 90%, respectively) compared to the HET treatment. Results indicated no significant difference in tilapia zootechnical performance between treatments.

1. Introduction

Integrated multitrophic aquaculture (IMTA) relies on interactions between species from different trophic levels to achieve sustainable production. In this system, waste, unconsumed food, organic matter, and inorganic compounds generated by the cultivation of one species are repurposed as food or a source of nutrients for other species [1], thereby promoting economic diversification and sustainability [2,3].

Enhanced productivity in IMTA systems can be attained through the utilization of biofloc technology (BFT) combined with soil−less plant cultivation (aquaponics) [4,5,6]. BFT represents a sustainable approach allowing minimal or no water exchange, achieved by fostering specific communities of microorganisms in the culture medium through manipulation of the carbon and nitrogen ratio of the system [7,8]. These microorganisms form aggregates known as “bioflocs”, which help maintain water quality by absorbing toxic nitrogenous compounds such as ammonia nitrogen and nitrite, converting them into less harmful nitrate [9] and microbial biomass that can be utilized by farmed animals, thereby reducing feed conversion rates and feed costs [8,10]. IMTA systems incorporating BFT technology have shown promise in marine aquaculture conducted in continental areas, particularly for marine shrimp Litopenaeus vannamei (Boone, 1931), utilizing alkaline–saline waters or marine salts dissolved in fresh water [11].

The oldest and most widely used method for biofloc formation involves the stimulation of heterotrophic nitrifying bacteria [7,12]. By introducing a carbon source into the water, these bacteria can rapidly reduce the concentration of toxic ammonia and nitrite, generating protein−rich bacterial biomass [13,14,15]. The total suspended solids content in water from heterotrophic BFT systems can reach hundreds of milligrams per liter due to exuberant microbial growth [8,16], necessitating the control of this material, often achieved through clarifiers and settlers [16].

When BFT technology is integrated into IMTA, part of the suspended solids can be removed by feeding fish with an iliophagous–detritivorous diet, such as Nile tilapia (Oreochromis niloticus) [17,18], filter−feeding mollusks [1], or deposit−eating invertebrates [3]. An alternative strategy for controlling suspended material is the formation of bioflocs using an inorganic carbon source (by adding ammonia and nitrite salts) and stimulating the development of chemoautotrophic nitrifying bacteria. These bacteria obtain energy by catabolizing ammonia into nitrite (ammonia oxidizing bacteria—AOB) and subsequently oxidizing nitrite to nitrate (nitrite oxidizing bacteria—NOB) [12]. Compared to heterotrophic bacteria, chemoautotrophic bacteria develop slowly, take time to stabilize in the system, and form a smaller bacterial biomass, resulting in fewer suspended solids [19]. Additionally, these microorganisms have a lower biochemical oxygen demand but demonstrate lower protein value [13,19,20].

Halophytes of the genus Salicornia have been integrated into aquaculture systems for shrimps and fishes through aquaponic systems, including those with BFT, with the aim of absorbing/controlling dissolved nutrients and producing biomass with high added value [4,5,17,21,22]. According to Poli et al. [17], IMTA using BFT technology with shrimp L. vannamei, Nile tilapia, and Salicornia plants demonstrated a 21% higher yield in total productivity compared to IMTA without Salicornia. However, the integration of aquaponics into BFT systems still faces technological limitations and challenges in finding plants tolerant to water with high levels of suspended material. The clogging of irrigation pipes and plant roots with particulate matter from BFT waters is a significant issue for aquaponics [17,22,23]. Furthermore, the incorporation of aquaponics into saline IMTA systems, employing biofloc technology, relies on the cultivation of halophytes (salt−tolerant vascular plants) capable of withstanding the nutritional conditions created by the unique microbial communities in BFT systems.

Studies conducted in Brazil have shown that the IMTA system with the native halophyte Salicornia neei Lag. [syn. Salicornia gaudichaudiana Moq., Sarcocornia ambigua (Michx.) M.A. Alonso and M.B. Crespo [21] is advantageous for the production of L. vannamei shrimps in BFT systems [17]. This plant, known as marine asparagus, has been successfully cultivated in marine aquaponics using BFT technology on laminar nutrient flow benches (Nutrient Film Technique—NFT [4,22]) and tanks with floating rafts [24]. Pinheiro et al. [4] reported a 25% increase in the efficiency of nitrogen utilization added to shrimp (L. vannamei) cultivation in BFT systems connected to S. neei aquaponic benches compared to those without halophytes. Additionally, this research achieved a production of 2 kg of S. neei plants for each kilogram of shrimp in the proposed aquaponic system. More recently, Doncato and Costa [22,25] demonstrated that clarified waters from L. vannamei cultivation in BFT systems meet the macro− and micronutritional needs of S. neei.

The nutritional value and biomass productivity of S. neei are influenced by various cultivation conditions, including nutrient availability [24,25], salinity [26], and irrigation timing [27]. Therefore, it is essential to assess S. neei development and its capacity for phytoremediation of nutrients, as well as its impact on other physicochemical parameters of recirculated water, under different biofloc formation strategies applicable to saline IMTA cultures. This study aimed to evaluate the effects of different biofloc formation strategies (chemoautotrophic treatment = CHE, and heterotrophic treatment = HET) on S. neei integration in an IMTA system with marine shrimp L. vannamei and Nile tilapia Oreochromis niloticus. To reach that objective, the impact of this integration was quantified via the nutrient contents and other physicochemical parameters of the recirculated water, as well as via plant development and shrimp and fish yields.

2. Materials and Methods

2.1. Study Location and Organism Origin

The experiment was conducted at the Marine Aquaculture Station of the Federal University of Rio Grande (FURG), situated in Cassino Beach (Rio Grande, Brazil; 32°12′19″ S; 52°10′45″ W), spanning 73 days during the 2022 austral summer–autumn period.

Propagules of S. neei from the BTH2 lineage were produced through vegetative propagation from plants maintained in the germplasm of the Halophyte Biotechnology Laboratory (BTH−EMA−FURG), utilizing a cutting technique [21]. Prior to cultivation using aquaponics, all plants underwent uprooting, and their roots were cleansed with water to remove any substrate.

Sea water previously diluted in chlorinated tap water neutralized with vitamin C [18] was used to formulate water with a salinity of 20 g L⁻¹, and water lost by evaporation was replaced with tap water. The post−larvae (PL 12) of L. vannamei were procured from Aquatec LTDA (Canguaretama, Brazil). Juveniles of Nile tilapia (O. niloticus) were obtained from Piscicultura Dalferth Ltd. (Teutônia, Brazil) and acclimatized for 10 days in bioflocs with 20 g NaCl L⁻¹ before being transferred to their respective tanks. The experimental protocol was approved by FURG’s Ethical and Animal Welfare Committee (process number 23116.005895/2016−42).

2.2. Experimental Design and Systems

The experimental design was defined by two treatments representing different biofloc formation strategies (with three repetitions): IMTA chemoautotrophic (CHE) and IMTA heterotrophic (HET). Six multitrophic systems were established within an unheated greenhouse.

Each IMTA system comprised three main components (Figure 1): (1) a 20 m³ circular tank (with a useful volume of 17 m³) housing the L. vannamei shrimps and biofloc community; (2) a 4 m³ circular tank accommodating Nile tilapia (O. niloticus) fishes: (3) commercial NFT (Nutrient Film Technique) hydroponic benches containing S. neei plants, positioned on the exterior of the greenhouse. Each hydroponic bench comprised 6 PVC pipes 85 mm in diameter and 3 m in length, providing a total planting area of 4.8 m² with 144 plants (approximately 24 plants per square meter).

Within each IMTA system, an aeration system injected air into the two tanks through a blower connected to microporous hoses (aerotubes). Additionally, a submersible pump (75 w, SPA 4000 L/h, BOYU^©, Chaozhou, China) transported water from the shrimp tank to the tilapia tank via PVC pipes and, via gravity, returned it to the shrimp tank (Figure 1). The same submersible pump transferred water from the shrimp tank to the hydroponic bench through connections in the piping and hoses, with an average flow rate of 455.3 L h⁻¹, which then returned to the shrimp tank via gravity.

Each IMTA system was equipped with a settling tank with a useful volume of 150 L, receiving water from the submerged pump in the shrimp tank via pipes. This clarifier was activated when total suspended solids (TSS) levels exceeded 300 mg L⁻¹. Weekly maintenance involved washing the PVC pipes on the hydroponic benches to remove excess solids and prevent pipe blockages and water overflow. The drained solids were returned to the shrimp tanks.

For the development of the chemoautotrophic system, three shrimp tanks were fertilized over a period of 45 days before stocking shrimps and fishes. Fertilization was conducted once daily with 1 mg L⁻¹ ammonium chloride (NH₄Cl) and 1 mg L⁻¹ sodium nitrite (NaNO₂) to stimulate the growth of nitrifying bacteria, following the methodology proposed by Ferreira et al. [19]. In the remaining three shrimp tanks, the heterotrophic system was established using the methodology defined by Ebeling et al. [13] and Samocha et al. [7]. After stocking the animals in the heterotrophic system, the concentration of total ammonia nitrogen (NAT) was monitored daily until it reached 1.0 mg L⁻¹. At this point, a source of organic carbon (sugarcane molasses) was added, stimulating the growth of bacterial biomass, while considering the carbon/nitrogen ratio (15C:1N), until a reduction in NAT content was observed.

The initial stocking densities of shrimp and tilapia were 400 individuals m⁻³ (average weight 1 g) and 44 individuals m⁻³ (average weight 25 g), respectively. The experiment spanned 84 days. Shrimp and tilapia were fed species−specific commercial dry diets (containing 38% crude protein). Shrimp were fed twice daily (at 8:00 a.m. and 3:00 p.m.) following the methodology described by Jory et al. [28], while tilapia were underfed once daily, consuming approximately 1% of their biomass, thereby encouraging the fish to feed on the bioflocs [18]. The feeding rate for shrimp was adjusted weekly, and for tilapia, it was adjusted biweekly based on growth sampled during this study.

Eleven (11) days after stocking the animals in their tanks, S. neei plants with shoot heights between 10 and 15 cm were placed in 200 mL plastic net pots filled with small gravel and transferred to the hydroponic benches connected to the IMTA systems. The plants were acclimatized on the hydroponic benches for 17 days. Subsequently, all plants underwent a leveling cut of their shoots, with each shoot being cut 4 cm above the upper edge of each cup. The development of the plants was then evaluated based on the regrowth of branch apical meristems, following the methodology described by Costa and Herrera [21] and Doncato and Costa [29]. The experimental evaluation period for S. neei lasted for 56 days (8 weeks), commencing 28 days after the shrimp were stocked in the shrimp tanks.

2.3. Assessment of Meteorological and Physical–Chemical Parameters of Water

Daily meteorological data were obtained from the INMET automatic station on the FURG campus (32°04′43″ S; 52°10′03″ W), accessed on 15 July 2024 https://portal.inmet.gov.br/dadoshistoricos. Throughout the experiment, periodic collections of input and output water from the hydroponic bench of each IMTA system were conducted, respectively, in the shrimp tank and in the bench’s return pipe.

Dissolved oxygen and water temperature were monitored using a model Pro−20, YSI Inc., Yellow Springs, OH, USA multiparameter device. The pH was measured with a Seven2Go S7 Básico, Mettler Toledo, São Paulo, Brazil pH meter, and salinity was determined using a multiparameter probe (HI9829 Hanna instruments, Smithfield, RI, USA). Alkalinity was monitored, ensuring it remained above 150 mg CaCO₃ L⁻¹, by utilizing calcium hydroxide for corrections [30]. Water samples (50 mL) were collected at the specified points, and total suspended solids (TSS) levels were estimated via gravimetry after filtration using pre−weighed Whatman GF/C filters [31]. Concentrations of total ammonia nitrogen (NAT), nitrite, nitrate, and phosphate were determined using spectrophotometry [32,33]. Temperature and dissolved oxygen were quantified three times a week, pH and nutrients were estimated twice a week, and alkalinity, salinity, and TSS were measured once a week during the experiment.

2.4. Assessment of Zootechnical Parameters and Growth and Productivity of S. neei

At the conclusion of the experimental period, all remaining animals (shrimp and fish) were tallied to determine survival. The parameters analyzed included the following: (1) survival (%) calculated as (final number/initial number) × 100; (2) final mean weight (g) determined as the final weight of animals (g) divided by the total number of animals; (3) productivity of animals (kg m⁻³) calculated as [(final biomass (kg) − initial biomass (kg)) × 1000] divided by the useful tank volume (L); and (4) SGR: specific growth rate (% day − 1) calculated as [(ln (final weight (g)) − (ln (initial weight (g)) × 100/days].

Plant growth was evaluated through two cuttings, performed every 28 days of cultivation. During each cutting, all S. neei plants on the benches were identified and trimmed 4 cm above the upper edge of each cup. Following each cutting, the fresh aerial biomass of each plant (shoots with branches) was measured using a precision balance (±0.01 g). The fresh biomass of the roots of each plant, which developed outside the cups, was quantified only at the conclusion of the experiment, also by weighing on a balance. The percentage of fresh biomass allocated to shoots (aerial biomass) was estimated by dividing the shoot biomass values by the total biomass formed (shoots + roots) and multiplying by 100.

The shoot biomass of 10 plants from each bench was dried in an oven (60 °C for 48 h) and weighed on a precision scale (±0.01 g), allowing for the estimation of dry biomass. Shoot succulence was determined by calculating the percentage difference between fresh and dry biomass. Biomass production per bench area was also estimated by summing the individual fresh biomass of all plants on the bench and subsequently dividing this value by the bench area (4.8 m²). Additionally, the height of the shoot (in cm) and the number of its branches longer than 10 cm in length for each plant were quantified.

2.5. Statistical Analysis

All collected water quality data were evaluated using two−factor ANOVAs of repeated measures (within−subject; bench input and output) [34], with treatment strategies for biofloc formation (CHE and HET) and the collection date (week) as fixed between−subject factors.

Regarding plant development, S. neei regrowth data after repeated shoot cutting (shoot height, number of branches of the cut shoot ≥ 10 cm, fresh aerial biomass, and shoot succulence) were compared between the biofloc formation strategies (fixed between−subject factor) through one−way ANOVA of repeated measures (within−subject; 1st and 2nd cutting of each plant) [34]. Final root biomass data, percentage allocation of biomass to shoots, and the survival rate of plants from the two biofloc formation strategies were compared using Student’s t−test. The zootechnical performance of shrimp and fish was also evaluated using Student’s t−test.

Physicochemical parameters of water and animal growth, as well as biometric and biomass values of S. neei plants, were tested for normality (Kolmogorov–Smirnov test) and homoscedasticity (Levene test). Variables that did not exhibit normality and/or homoscedasticity were mathematically transformed. In this experiment, the variable shoot height was transformed using the Log10(×) function. For ANOVAs and t−tests, differences between means were considered significant at a probability level of p < 0.05. In the ANOVAs, means were compared using the Tukey HSD post hoc test (p < 0.05) [34].

3. Results

3.1. Environmental Conditions and Water Quality

The air temperature outside the greenhouse ranged from 9.0 to 32.5 °C (average ± standard error = 22.0 ± 0.4 °C), with a maximum solar radiation observed at 3.9 J m⁻² s⁻¹ (3900 W m⁻²; average daily radiation = 17.87 ± 0.76 MJ m⁻² day⁻¹).

The waters from the CHE treatment exhibited global averages of nitrate and phosphate significantly higher than those in the HET treatment (

Table 1. Average (±SE) of water quality parameters in IMTA systems with chemoautotrophic (CHE) and heterotrophic (HET) biofloc formation strategies during 56 days of S. neei evaluation. Overall average values before (input = IN) and after water passage (output = OUT) through the hydroponic benches with S. neei plants are presented for each biofloc formation strategy. Legend: DO = dissolved oxygen; NAT = total ammonia nitrogen; TSS = total suspended solids. The parameters DO, alkalinity, NAT, nitrite, nitrate, phosphate, and TSS are presented in mg L⁻¹. Temperature is presented in °C and salinity in g L⁻¹. Different lowercase letters in the lines indicate significantly different averages (p < 0.05) according to the Tukey test.

Parameter	Treatment
	CHE						HET
	IN			OUT			IN			OUT
DO	5.78	±	0.05 a	5.69	±	0.05 a	6.21	±	0.06 c	6.10	±	0.05 b
Temperature	25.37	±	0.24 c	25.03	±	0.25 ab	25.09	±	0.26 bc	24.78	±	0.27 a
pH	7.78	±	0.02 a	7.82	±	0.02 ab	7.83	±	0.03 ab	7.88	±	0.02 b
Alkalinity	158.13	±	3.86	157.71	±	4.75	165.83	±	4.12	163.75	±	3.96
NAT	0.14	±	0.01 b	0.16	±	0.01 c	0.11	±	0.01 a	0.14	±	0.01 b
Nitrite	1.06	±	0.14 a	1.04	±	0.12 a	5.42	±	1.18 b	5.16	±	1.14 b
Nitrate	95.09	±	5.70 b	101.78	±	6.52 b	59.80	±	5.40 a	64.61	±	5.87 a
Phosphate	4.49	±	0.36 b	4.74	±	0.35 b	3.32	±	0.30 a	3.27	±	0.26 a
TSS	350.83	±	16.64 b	343.54	±	17.98 b	305.00	±	19.60 a	316.26	±	22.32 a
Salinity	20.74	±	0.35 a	20.76	±	0.34 a	24.44	±	0.31 b	24.37	±	0.31 b

). These differences between IMTA treatments were more pronounced before the first cutting of S. neei plants (Figure 2 and Figure 3). NAT and SST also showed their highest values in the CHE treatment. In contrast, the highest average values of pH, salinity, nitrite, and dissolved oxygen were observed in the HET treatment (Table 1).

Water flow through the halophyte bench promoted small (but significant; p < 0.001) reductions in the global average values of dissolved oxygen and water temperature (1.2–1.8%; joint data from both BFT treatments). Additionally, this passage increased the concentrations of NAT (F = 22.33; p < 0.001), nitrate (F = 7.65; p < 0.01), and pH (F = 4.39; p < 0.001) (0.5–27.3%).

All water parameters, except NAT, alkalinity, and salinity, exhibited marked and significant (p < 0.01) variations throughout the experiment (between weeks). The average values of nitrate (Figure 2; F_Week = 91.1, p = < 0.01), phosphate (Figure 3; F_week = 21.99, p = < 0.01), and TSS (Figure 4; F_week = 6.12, p = < 0.01) increased throughout the cultivation, while nitrite values (Figure 2; F_week = 13.6, p =< 0.01) decreased, and other parameters fluctuated between the weeks of cultivation (e.g., dissolved oxygen had values inversely related to water temperature, varying, respectively, from 7.86 to 4.74 mg L⁻¹ and from 20.3 to 28.6 °C). These temporal variations were similar in the CHE and HET treatments, except for nitrite content (biofloc strategy × week interaction; F = 12.18; p < 0.001). During the initial three weeks after plants underwent the leveling cut of their shoots, nitrite concentrations in the HET tanks exceeded the safety level for shrimp (15.2 mg L⁻¹ in 25 g NaCl L⁻¹; [35]). To mitigate this, 30% of the water in the tanks was renewed, and the feed supply was reduced (by 30 to 50%) following Valencia−Casteñeda et al.’s [36] approach to acute toxicity of nitrite, finally dropping the nitrite concentrations at the third week (Figure 2). In the CHE treatment, nitrite was already stabilized when S. neei was cut, remaining constant throughout the experiment (Figure 2).

3.2. Growth of S. neei in the Waters of the Two IMTA Biofloc Formation Strategies

The average values of survival rates of S. neei plants (85.7–91.5%) and shoot succulence (88–90% water content) showed no statistical differences between treatments (

Table 2. Performance of S. neei plants cultivated with waters from IMTA with different biofloc formation strategies (chemoautotrophic = CHE; heterotrophic = HET) during 56 days of S. neei evaluation. Averages (±SE) for each biometric parameter in the two consecutive cuts of the shoots are presented. Different lowercase letters in the lines indicate significantly different averages (p < 0.05) according to Student’s t−tests (parameters evaluated only at the 2nd cut) and Tukey tests (after ANOVA).

Parameter	Treatment
	CHE						HET
	1st Cut			2nd Cut			1st Cut			2nd Cut
Survival rate (%)	−	−	−	91.52	±	0.29	−	−	−	85.67	±	2.88
Shoot height (cm)	15.60	±	0.28 b	17.53	±	0.23 c	12.60	±	0.27 a	16.97	±	0.25 c
Branch ≥ 10 cm (per shoot)	6.93	±	0.37 b	10.43	±	0.36 c	3.73	±	0.26 a	7.97	±	0.29 b
Shoot biomass (g)	15.80	±	0.77 b	30.99	±	0.81 d	9.63	±	0.53 a	27.01	±	0.79 c
Shoot productivity (kg m−2)	0.34	±	0.08 ab	0.67	±	0.03 c	0.19	±	0.04 a	0.53	±	0.03 bc
Shoot succulence (%)	89.20	±	0.34 ab	88.84	±	0.64 ab	88.10	±	0.25 a	89.95	±	0.16 b
Root biomass (g)	−	−	−	2.21	±	0.08	−	−	−	2.38	±	0.09
Shoot allocation (%)	−	−	−	95.42	±	0.14 a	−	−	−	93.87	±	0.16 b

and

Table 3. Results of repeated measures factorial ANOVAs for S. neei shoot development and production, as well as “t” tests for survival, root biomass, and biomass allocation for shoot formation. Averages were considered significantly different at the p < 0.05 level. F = Fisher test; p = significance values for the biofloc formation strategies (t), cut stage (c), and their interaction (t X c).

Parameter	t−test		ANOVA
	t	p	Ft	p	Fc	P	FtXc	p
Survival rate (%)	2.02	0.11
Shoot height (cm)	−	−	37.92	<0.001	196.08	<0.001	29.91	<0.001
Branch ≥ 10 cm (per shoot)	−	−	56.50	<0.001	204.05	<0.001	1.83	0.18
Shoot biomass (g)	−	−	31.73	<0.001	913.88	<0.01	3.99	0.05
Shoot productivity (kg m−2)	−	−	4.97	0.09	94.79	<0.001	0.04	0.85
Shoot succulence (%)	−	−	<0.001	1.00	5.89	0.02	14.04	<0.001
Root biomass (g)	−1.38	0.17
Biomass allocation–Shoot (%)	7.34	<0.001

). In the first cutting, the averages for shoot height, number of branches ≥ 10 cm, shoot biomass, and biomass allocation into shoot formation were significantly higher in the CHE treatment than in the HET treatment (Table 2). Except for shoot height, in the second cutting, statistical differences between treatments remained. Within each biofloc treatment, the parameters mentioned above were significantly higher in the second cutting (Table 2 and Table 3).

The average productivity of S. neei fresh shoot biomass per bench area ranged from 0.19 to 0.67 kg m⁻² per 30 days, being higher in the second cutting in both biofloc treatments (Table 2 and Table 3). The productivity of S. neei per unit of bench area showed no significant difference (p = 0.09) between the CHE and HET treatments (Table 3), a statistical result that may reflect the low degree of freedom for this comparison (n = 3; df = 2).

3.3. Zootechnical Development of Shrimp and Tilapia in Waters of the Two IMTA Biofloc Formation Strategies

Both L. vannamei shrimp and tilapia had survival rates greater than 75%, which did not differ between biofloc formation strategies. In the CHE treatment, shrimp had a higher average weight, specific growth rate (SGR), and, consequently, greater productivity per tank volume than in the HET treatment (

Table 4. Performance (Averages ± SE) of L. vannamei and O. niloticus tilapia during the experimental period in two different biofloc formation strategies (chemoautotrophic = CHE; heterotrophic = HET). Different lowercase letters in the lines indicate significantly different averages (p < 0.05) according to Student’s t−tests).

Parameter	Treatment
	CHE			HET			t	p
L. vannamei
Survival (%)	90.50	±	3.75	85.09	±	4.21	1.17	0.31
Final average weight (g)	11.24	±	1.39 a	8.25	±	1.06 b	10.42	<0.01
Yield (kg m−3)	3.86	±	0.08 a	2.83	±	0.08 b	10.42	<0.01
SGR (% day−1)	2.88	±	0.02 a	2.51	±	0.03 b	10.20	<0.01
O. niloticus
Survival (%)	75.05	±	17.27	83.81	±	6.31	−0.58	0.59
Final average weight (g)	171.60	±	12.59	180.28	±	6.53	−0.79	0.47
Yield (kg m−3)	1.11	±	0.26	1.29	±	0.13	−0.77	0.48
SGR (% day−1)	2.29	±	0.07	2.35	±	0.04	−0.82	0.46

). Tilapia’s zootechnical parameters showed no statistical differences between treatments.

4. Discussion

4.1. Water Quality Parameters in CHE and HET Treatments

In both BFT treatments, the levels of dissolved oxygen, NAT, nitrate, phosphate, pH, salinity, and alkalinity were within adequate limits for the survival and growth of L. vannamei [7,8,17] and O. niloticus [14,17]. The average water temperature (25 °C) was close to the ideal limits for both shrimp and tilapia [14,37]. Water TSS content also remained within the tolerance limits of cultivated animals [16,17], and phosphate concentrations in BFT treatments were in the range of average values cited for saline cultures with bioflocs (2–4 mg P−PO₄ L⁻¹ [4,18,22,26,38]). However, the highest TSS and phosphate values in the CHE treatment differ from other studies. Ray et al. [20], Ferreira et al. [19], and Soares et al. [38] reported higher concentrations of suspended solids in heterotrophic BFT systems than in chemoautotrophic systems and observed that phosphate levels in a heterotrophic BFT were associated with the addition of molasses (phosphorus source). The lowest TSS and phosphate levels in HET in the present study can be partially explained by the water renewal and reduction in feed supply due to nitrite peaks during the first three weeks after the leveling cut of S. neei plants.

A faster process of nitrification occurred during the first stage of S. neei cultivation in the CHE treatment compared to the HET treatment, which had a marked impact on the development of halophyte plants. The nitrification process through the oxidation of ammonia to nitrite and then to nitrate is mainly carried out by chemoautotrophic bacteria [13,15,19]. Therefore, the rapid accumulation of nitrate indicates greater efficiency of the nitrification process in the CHE treatment. Similar results were obtained in previous studies comparing cultures in chemoautotrophic and heterotrophic BFT systems [19,20,38], demonstrating that the nitrification process occurred more slowly and reduced when heterotrophic pathways were favored. The connection of the hydroponic benches occurred at the beginning of animal cultivation and encompassed the entire period of biofloc development. Over time, the accumulation of organic particles from animal excreta and the decomposition of unused feed can favor the development of heterotrophic bacteria, even in an initially chemoautotrophic environment [12,15]. These heterotrophic bacteria have higher reproduction rates than chemoautotrophic ones, and competition for substrate, nutrients, and/or oxygen between groups of bacteria leads to a reduction in nitrification rates as a mixed community is established [19]. This process of maturation of the biofloc microbial community seems to explain similar nutritional conditions of CHE and HET treatment waters towards the end of the experiment. Other evidence of this maturation process and formation of a predominantly heterotrophic mixed community are the high TSS levels observed (>300 mg L⁻¹) and the need to connect clarifiers during the final phase of cultivation in tanks of both treatments.

4.2. Effect of Inserting Aquaponic Benches with Halophytes on IMTA Water Quality

The passage of water from the IMTA systems through the hydroponic benches had minimal effect on the magnitude of water quality parameters. Generally, slight reductions in temperature and dissolved oxygen were observed, along with increases in pH, dissolved NAT, and nitrate levels after passage. The small temperature decrease in output water can be attributed to the location of hydroponic benches outside the greenhouse, while the reduction in dissolved oxygen may be related to the contact of water with biofloc accumulated in the pipes and plant roots. These solids were removed by weekly washing and did not affect the development of S. neei plants, with root growth observed over the deposited material in the pipes.

This result apparently contrasts with findings by Santos et al. [38], who reported impaired performance of Salicornia ambigua (syn. S. neei) plants due to the accumulation of bioflocs in hydroponic benches. However, the same authors describe a reduced nitrification process in their heterotrophic BFT systems and an average nitrate value of just 0.41 mg N−NO₃ L⁻¹. This low availability of nitrate, more than the solids’ accumulation in the pipes, likely limited the development of S. neei in their study. In laboratory conditions, Doncato and Costa [25] estimated that S. neei prefers absorbing nitrogen in the form of nitrate, and that nitrate levels below 10 mg L⁻¹, particularly in aquaculture waters with the presence of ammonia nitrogen, can limit plant development.

The water contact with S. neei hydroponic benches resulted in a small but significant increase (+7.4%) in the global average nitrate concentration of output water (83.20 mg N−NO₃ L⁻¹; joint data from CHE and HET treatments) compared to the input water from the shrimp tank (77.45 mg N−NO₃ L⁻¹). Despite the vigorous growth of S. neei and its assimilation of nitrate into plant biomass, this absorption process was unable to counteract the formation of this nutrient due to the intense nitrification process in the bioflocs. Previous aquaponic cultivations of S. neei have reported similar results, indicating a significant contribution of this species to nitrogen recovery from intensive shrimp farming, albeit with limited reduction in the high levels of nitrate formed in BFT systems [4,17,26]. However, nitrate concentrations were maintained within safe levels for farmed animals [9,14,35]. Since the plant density used in this study did not significantly affect water quality or the performance of the animals, expanding the S. neei cultivation area could potentially increase plant and overall IMTA system productivity while reducing nitrogenous compounds.

The cultivation of S. neei during the experiments exhibited high survival rates (92%), similar to previous cultivations of this species in BFT systems (92% [17]; 87–95% [38]; 100% [22]), indicating the species’ good adaptation to systems irrigated with waters containing high levels of particulate matter. However, lower survival rates of S. neei (67.5 to 70.4%) have been reported when cultivated in a system with BFT clarified water and feed reduction [39]. Despite variations in survival rates across studies, the productivity of S. neei per bench area achieved in our study compares favorably with previous cultivations of this species with saline waters from BFT systems (

Table 5. Performance of S. neei in cultivations with saline water from BFT systems. D = plant density; * = regrowth values (after leveling cut); # = sum of two cuttings during the cultivation cycle.

Reference	Individual Shoot (g)	Production (kg m−2)	Time (Week)	Productivity (kg m−2 mês−1)	D (m−2)	Experimental Evaluation
[4]	85.1	8.10	10.4	3.31	100	Aquaponics with shrimp.
[26]	12.2−21.4	0.38−0.61	8.1	0.19−0.30	40	Aquaponics with shrimp; different salinities.
[38]	13.0−41.2	0.49−1.69	5.7	0.34−1.19	40	Aquaponics with shrimp; BFT strategies.
[27]	10.8−19.0	1.10−1.90	10.0	0.44−0.76	100	Aquaponics with shrimp; irrigation time.
[39] *	9.4–11.5	0.67−0.77	12.0	0.24–0.28	100	Aquaponics with shrimp; animal feeding rate.
[17]	23.0	2.27	8.1	1.12	97	IMTA with BFT.
Present study *	37.0−46.8	0.72−1.01 #	8.0	0.19–0.67	24	IMTA; BFT strategies.

). The highest recorded productivity of S. neei in aquaponic cultivation was reported by Pinheiro et al. [4] (3.3 kg m⁻² month⁻¹), achieved with high plant density in hydroponic benches (100 plants m⁻²) and under high nutrient availability and optimal tropical conditions for S. neei [21]. Other productivity values found in the literature for this species are at least 1–2−times lower than the one mentioned above. Some benches’ productivities were most probably limited by low nitrate concentrations (e.g., [38,39]). Our study’s best average performance was observed in the CHE treatment (0.67 kg m⁻² month⁻¹), which can be attributed to the higher concentrations of nitrate and phosphate in this system. The greater differences in plant performance between biofloc formation strategies were observed in the 1st cutting, following a period of greater nitrification processes and phosphate accumulation in the CHE treatment, which had an overall nitrate and phosphate concentrations of 98.41 mg N−NO3 L⁻¹ and 4.62 P−PO4 L⁻¹, respectively. The better nutritional conditions in the CHE treatment resulted not only in high individual biomass and productivity per bench area but also in desirable agronomic traits (sensu Ventura et al. [40]), such as high allocation of primary production to shoot biomass and tall plants with a large number of long branches (≥10 cm). The succulence of S. neei plants in our study was similar to other studies [25], reflecting good growing conditions and desirable characteristics for culinary purposes. The greater allocation of shoot biomass in the CHE treatment can be attributed to higher concentrations of nitrate, which stimulate shoot formation, provided that high levels of phosphorus are also available [24,25].

The average shoot heights obtained in the present study after cultivation periods of 4 weeks (1st cutting = 15.61 cm; 2nd cutting = 17.53 cm) align with values between 14 and 20 cm previously reported in aquaponic production of S. neei with clarified water from 454 shrimp BFT systems, using the same cutting interval [22,29]. Laboratory cultivation of S. neei in enriched saline media has also produced similar or taller shoots over longer periods (37−42 cm after 15 weeks, [24]; 25−30 cm after 7 weeks, [25]; 15−16 cm after 22 weeks, [41]), as seen with Salicornia persica (18−24 cm after 16 weeks, [40]). Typically, the total number of branches formed in S. neei shoots is quantified, rather than just the number of branches with lengths ≥ 10 cm (considered commercially viable), making comparisons with other studies challenging. However, the average values of the CHE treatment (≈7 and 10.5 branches ≥ 10 cm per shoot, in the 1st and 2nd cutting, respectively) closely resemble the number of commercially viable branches produced 4 weeks after shoot cutting by Doncato and Costa [29] (8.5 branches per shoot). The branch numbers in the CHE treatment fall within an intermediate range compared to the total numbers of branches produced by S. neei plants (1.3–60 branches per shoot) in previous studies [25,41], indicating excellent performance concerning the commercial value of the plants produced.

4.3. Zootechnical Development of Shrimp and Fish

Shrimp performance yielded the highest final averages of survival, individual weight, tank productivity, and specific growth rate (SGR) in the CHE treatment (90%, 11.24 g, 3.86 kg m⁻³, and 2.88% day⁻¹, respectively). This performance aligns with observations by the authors of [4], who cultivated L. vannamei in BFT with S. neei aquaponics for 10 weeks (using a stocking density of 250 ind. m⁻²), obtaining an average weight and shrimp productivity of 11.6 g and 2.15 kg m⁻³, respectively. The lower final shrimp weight and productivity in the HET treatment may be attributed to the high nitrite levels observed during the first weeks of cultivation. Vinatea at al. [42] reported an inverse relationship between shrimp growth rate and nitrite concentrations ranging from 0.72 to 9.49 mg N−NO₂ L⁻¹. Santos et al. [38] also observed lower performance of L. vannamei in heterotrophic compared to chemoautotrophic BFT systems, associating it with high TSS concentrations in the former. In contrast, Ferreira et al. [19] did not observe statistical differences in the zootechnical performance of L. vannamei cultivated in heterotrophic and chemoautotrophic systems. The tilapia O. niloticus exhibited similar performance in IMTA with different biofloc formation strategies, and SGR values were close to other studies. Tilapia SGR in the present study was lower than Poli et al.’s (2019; 4,41% day⁻¹) result in IMTA integrated with S. neei aquaponics, which used small fishes and, as a consequence, obtained high SGR. However, Carvalho et al. (2023), studying the production of a IMTA with tilapia, shrimp, and macroalgae, found fish SGR values ranging between 0.38 and 0.84% dia⁻¹ (8.0–9.9 g week⁻¹). These results confirm the rusticity of O. niloticus, capable of tolerating peaks of toxic nitrogen and utilizing nutrients from bioflocs. A recent study highlights greater tilapia productivity in BFT system and BFT system coupled with aquaponics (5.86 and 6.10 kg m⁻³, respectively) compared to RAS aquaponics (4.71 kg m⁻³) [6].

5. Conclusions

The aquaponic cultivation of the native halophyte S. neei within IMTA systems featuring marine shrimp (L. vannamei) and Nile tilapia (O. niloticus) using biofloc technology proves to be viable. This method yields plants with excellent agronomic characteristics. While S. neei can be cultivated in both chemoautotrophic and heterotrophic biofloc formation strategies, superior productivity (e.g., individual shoot production up to 31 g per 30 days) is achieved under chemoautotrophic treatment characterized by elevated concentrations of dissolved nitrate and phosphate. Importantly, integrating S. neei plants into IMTA systems had minimal impact on water quality and nutrient concentrations, suggesting the potential for scaling up plant production within these systems.