Litopenaeus vannamei, commonly known as the Pacific white shrimp, is a euryhaline species valued for its high meat yield, strong stress tolerance, and rapid growth. It is one of the most important shrimp species farmed in China. Currently, the primary farming models for L. vannamei in China include outdoor ponds, small greenhouse ponds, and high-level ponds. However, domestic production still cannot meet market demand, necessitating significant imports. Moreover, the rapid expansion of models like small greenhouse farming has exposed issues such as an incomplete technical framework, frequent disease outbreaks, and challenges in treating effluent wastewater. Against the backdrop of advocating for resource conservation and sustainable development, the Recirculating Aquaculture System (RAS), recognized as an intensive, efficient, and environmentally friendly farming model, has garnered widespread attention in the industry in recent years.
RAS employs industrial methods to actively regulate the water environment. It features low water consumption, a small footprint, minimal environmental pollution, and yields high-quality, safe products with fewer diseases and higher stocking densities. Its production is largely unrestricted by geography or climate. This model boasts high resource utilization efficiency and is characterized by high investment and high output, representing a crucial pathway toward the sustainable development of the aquaculture industry. Currently, domestic farming of L. vannamei is concentrated in coastal areas, primarily utilizing natural seawater. Inland regions, constrained by water source availability and environmental regulations, face a significant mismatch between supply and consumer demand. Exploring RAS using artificial seawater in inland areas holds great significance for supplying local markets and promoting regional economic development. This experiment successfully constructed an indoor RAS for L. vannamei in an inland setting and conducted a successful cultivation cycle. The methods and data concerning system construction, artificial seawater preparation, and farm management can serve as a reference for inland L. vannamei farming.
1. Materials and Methods
1.1 Materials
The trial was conducted at the Sichuan Province Leiocassis longirostris Original Breeding Farm. The post-larval L. vannamei (P5 stage) were sourced from the Huanghua Base of Qingdao Hainen Aquatic Seed Industry Technology Co., Ltd., and were in good health. The feed used was the "Xia Gan Qiang" brand from Tongwei Group Co., Ltd. Its main components were: crude protein ≥44.00%, crude fat ≥6.00%, crude fiber ≤5.00%, and crude ash ≤16.00%.
1.2 Artificial Seawater Preparation
Groundwater from a well was used as the source water. It was sequentially treated with disinfection (bleaching powder 30 mg/L, aerated for 72 h), residual chlorine removal (sodium thiosulfate, 15 mg/L), and detoxification [Ethylenediaminetetraacetic acid (EDTA), 10–30 mg/L] before being used for artificial seawater preparation.
Artificial seawater with a salinity of 8 was prepared using sea salt crystals as the main ingredient; its primary components are listed in Table 1. Food-grade CaCl₂, MgSO₄, and KCl were used to supplement Ca, Mg, and K elements. After preparation, food-grade NaHCO₃ was used to adjust the total alkalinity to 250 mg/L (as CaCO₃), and NaHCO₃ along with citric acid monohydrate were used to adjust the pH to 8.2–8.4.
1.3 RAS Construction
1.3.1 Overall Design Concept
Combining independent design with integrated application, an RAS for L. vannamei was constructed utilizing multi-stage physical treatment and biofiltration. Corresponding system operation strategies, water quality adjustment protocols, and scientific feeding strategies were implemented according to the shrimp's growth requirements at different stages, aiming for stable operation, economic input, and efficient output.
1.3.2 Main Process Flow and Technical Parameters
An existing container-based fish farming system was modified to establish the L. vannamei RAS, consisting of culture tanks, a composite shell/particulate collection device (three-way drainage), biofilter, circulation pumps, etc. The process flow is shown in Figure 1.
The system's total designed water volume was 750 m³, with a water treatment system volume of 150 m³ and an effective culture volume of 600 m³. The designed culture load was 7 kg/m³. Key technical parameters are listed in Table 2.
1.3.3 Structural Design
The six octagonal culture tanks were arranged in two rows. Considering management convenience, environmental stability, and investment cost, the main structure of the tanks was brick-concrete. Dimensions were: length 10.0 m, width 10.0 m, depth 1.2 m, with cut edges of 3.0 m. The effective water volume per tank was 100 m³. The tank bottom had a slope (16%) towards the central drain (Figure 2).
The three-way drainage device consisted of a central collector (for dead shrimp, shells, and large particles), a vertical flow sedimentation collector (for broken shells, medium particles, feces), and a siphon side-drain collection box (for fine shells and small-to-medium particles) (Figure 2).
One side of the conditioning tank contained a plastic brush media frame for collecting and removing shells and particles from the tank discharge. Adjustments for calcium, magnesium, total alkalinity, and pH could be made in this tank. Tank volume was 20 m³, with a hydraulic retention time of 0.13 h.
The circulation pump was located on the other side of the conditioning tank, using a single-stage pump for energy efficiency. Based on shrimp ecology and load, the recirculation rate was designed at 2–6 times/day. Pump flow rate was 150 m³/h, head 10 m, power 5.5 kW.
The brush filter was equipped with several filter bags. The bags were connected via pipe fittings to the filter inlet, secured with clamps. Effluent entered the bags via pipes. The bags were made of Polypropylene (PP), filled with plastic brush media, effectively intercepting particles larger than 0.125 mm. The elastic media tank consisted of the tank body (rectangular, depth 2 m), grid frames (parallel to the surface), and elastic media installed on the frames (Figure 3). The media comprised numerous double-ring plastic rings with polyester filaments, forming fiber bundles distributed throughout the tank. Its working principle involved creating a slow-flow sedimentation effect via the media's interception and utilizing the biofilm formed on its surface to absorb, decompose, and transform inorganic nitrogen and phosphorus.
The biofilter included the tank body (rectangular, depth 2 m), aeration components, and bio-media (Figure 4). The aeration assembly included air distribution pipes. Air entered from the top and was released from the bottom, creating a completely mixed flow pattern. The tank was filled with Moving Bed Biofilm Reactor (MBBR) media. By targeted nitrifier enhancement and alkalinity adjustment, large numbers of nitrifying bacteria attached to the media, consuming organic matter and achieving ammonia and nitrite removal, thus constructing a nitrifying biofilter. Inlet and outlet pipes were on opposite sides, with an outlet screen on the inner wall. In this trial, the biofilter effective volume was set at 25% of the system culture volume, with a media fill ratio of 30%, using K5 media.
System aeration combined mechanical and pure oxygen methods. When Dissolved Oxygen (DO) was high, mechanical aeration was primary: using a high-pressure vortex blower and high-quality microporous tubes as diffusers to maximize O₂ transfer efficiency and reduce noise. When DO was low, pure oxygen aeration supplemented: using an oxygen generator + micro-bubble water propeller. The oxygen generator output O₂ concentration above 90%, dispersed via a nano-ceramic disk in the propeller. Under high load, an oxygen generator + oxygen cone combination served as auxiliary aeration, using a booster pump to create oxygen-supersaturated water in the cone.
1.4 Water Quality Measurement
Ammonia and nitrite (as N) concentrations were measured using an Aokedan multi-parameter water analyzer. Total Suspended Solids (TSS) were measured using a Hach DR 900 multi-parameter analyzer.
1.5 Farm Management and System Operation
The trial began on August 8, 2022, lasting 74 days. All six tanks were stocked. Stocking size was 961 individuals/kg, density approximately 403 individuals/m³, totaling 241,800 post-larvae. Feeding frequency was 6 times/day, with daily ration decreasing from about 7.0% (early) to 2.5% (late) of estimated biomass.
System circulation started 3 days post-stocking, initially at 2 cycles/day, increasing to 4 cycles/day later. Early in the trial, daily draining occurred, only replenishing water lost to drainage and evaporation. Later, draining followed each feeding (1 hour after), with daily water exchange below 10% of the early-stage replenishment volume.
Mechanical aeration (vortex blower) was used initially. Due to increased system load later, a combination of mechanical aeration, oxygen generator + nano-ceramic disk, and oxygen generator + oxygen cone was used.
DO, temperature, pH, ammonia, and nitrite in the tanks were measured regularly. Shrimp growth and feeding were observed and recorded.
1.6 Data Processing and Analysis
Data were organized using WPS Office Excel. Graphs were created using Origin 2021.
The following formulas were used to calculate water exchange rate (R), feed conversion ratio (FCR), and survival rate (RS):
R = 100% × V₁ / (V × t) ... (1)
FCR = W / (Wₜ − W₀) ... (2)
RS = 100% × S / N ... (3)
Where: R is daily water exchange rate (%/d); V₁ is total exchanged water volume (m³); V is total system water volume (m³); t is culture days (d). FCR is feed conversion ratio; W is total feed input (kg); Wₜ and W₀ are final harvest mass and initial stocking mass (kg). RS is survival rate (%); S is total number harvested (individuals); N is total number stocked (individuals).
2. Results
2.1 Water Exchange
During the trial, total water exchange was 1,000 m³, with an average daily exchange rate of 1.8%.
2.2 Ammonia and Nitrite
Ammonia concentration in the tanks remained below 1.3 mg/L (except day 5), and nitrite concentration remained below 1.6 mg/L, both at relatively stable levels (Figure 5).
In the early stage (first 15 days), tank ammonia decreased rapidly while nitrite increased rapidly, indicating biofilm establishment in the biofilter and conversion of ammonia to nitrite. In the mid-stage (15–50 days), with increased feeding, ammonia and nitrite concentrations remained stable, indicating synchronized ammonia and nitrite oxidation in the biofilter and stable system operation. After day 50, both ammonia and nitrite showed a downward trend, possibly indicating enhanced nitrification capacity and a more mature system. This could not be confirmed further as the trial ended.
Figure 6 shows that ammonia trends in the biofilter inlet and outlet were similar, but the gap between the curves gradually widened, indicating improving ammonia removal. The nitrite curves for inlet and outlet nearly overlapped and did not show an overall increasing trend, suggesting the system maintained nitrite oxidation capacity until the end.
2.3 Dissolved Oxygen and Total Alkalinity
As shown in Figure 7, despite increasing system load, the combined aeration methods maintained tank DO above 6 mg/L. Furthermore, by adding NaHCO₃, total alkalinity was maintained between 175–260 mg/L.
2.4 Total Suspended Solids
Trends in TSS concentration at key system points are shown in Figure 8. The TSS in the inflow to the vertical flow sediment collector and siphon side box (part of the three-way drainage) reflected TSS trends in the tanks. Overall TSS increased gradually, stabilizing during mid-late stages (after day 35), and showed a decreasing trend through successive treatment stages.
2.5 Farming Results
Total stocking was 241,800 post-larvae at an average size of 0.52 g, across 6 tanks at an average density of 403 individuals/m³. After 74 days, total harvest was 3,012.2 kg, average size 15.82 g, average survival 78.75%, average yield 5.02 kg/m³. Total feed input was 3,386.51 kg, FCR 1.18. Calculated costs (seed, feed, health products, electricity, artificial seawater, disinfection) totaled 155,870.6 CNY. Revenue from shrimp sales was 192,780.8 CNY, resulting in a profit of 36,910.2 CNY for the cycle.
3. Discussion
In recent years, RAS has become a highly promising direction for L. vannamei farming. This trial constructed an RAS including culture tanks, composite shell/particulate collection, brush filter, biofilter, and aeration equipment, and successfully conducted one cycle of inland indoor farming.
Compared to traditional RAS, this system is simpler. Structurally, it omitted equipment like drum filters and protein skimmers, which have relatively higher fixed and maintenance costs. Instead, it used simpler water treatment devices to create a multi-level composite treatment for particles and dissolved pollutants, achieving good water quality control with simpler processes and lower cost.
By employing various water quality management methods tailored to different growth stages and system loads, the system maintained ammonia and nitrite below 1.3 and 1.6 mg/L, respectively, and DO above 6 mg/L, ultimately achieving a yield of 5.02 kg/m³. This is close to results from Yang Jing et al. Furthermore, the water treatment system controlled the average daily exchange rate to 1.8%, fully utilizing its treatment capacity and significantly reducing costs.
RAS offers environmental benefits, product safety, and fewer diseases. Due to transportation limitations, L. vannamei holds great market potential inland. Conducting RAS for L. vannamei inland aligns with industry trends. Current inland shrimp farming is primarily freshwater, with yield and quality lagging behind marine farming. Using artificial seawater in this trial partly addressed this gap. However, the current high cost of artificial seawater necessitates optimizing RAS processes for nitrogen and phosphorus removal to enable water reuse, which is an effective way to reduce costs and should be a key research focus for inland L. vannamei RAS.
FCR is an important indicator for RAS performance. The final FCR of 1.18 in this trial is comparable to traditional intensive farming. As a closed system, RAS's advantage lies in input reuse. Based on enhancing water treatment capacity, formulating precise feeding strategies to lower FCR should be the next optimization focus.