UV-C integrated bio-mechanical filtration for water quality management in koi (Cyprinus rubrofuscus) culture tank

Introduction

The ornamental fish industry is rapidly recognized as one of the most prominent and fastestgrowing sectors within global aquaculture, contributing substantially to international trade, rural employment and recreational activities. Currently, over 125 countries are actively engaged in the production and global distribution of ornamental fishes, with freshwater species dominating nearly 90% of the entire market. The widespread popularity of aquarium keeping is largely driven by its aesthetic appeal and documented therapeutic benefits, which include reducing stress and lowering blood pressure for hobbyists (Oertli et al., 2023) [17] . Water quality remains the most critical determining factor in ornamental fish culture because these aquatic organisms spend their entire life cycle in constant, direct contact with their surrounding environment (Hemal et al., 2024) [8] . Consequently, any sudden deterioration in water chemistry can rapidly compromise fish physiology, suppress immune responses and severely impact survival rates (Abdel-Tawwab et al., 2019) [1] . A primary challenge inherent to intensive ornamental fish culture is the continuous accumulation of organic wastes, which originate from fish excretion, unconsumed feed residues, decaying plant materials and microbial decomposition (Elmitwalli et al., 2006) [5] . As these organic wastes undergo biochemical transformations, they release highly toxic nitrogenous compounds directly into the water column. Ammonia serves as the principal toxic metabolite in these closed systems, capable of causing severe gill damage, disrupting osmoregulation and inducing mass mortality (Bhatnagar & Devi, 2013; Hussain et al., 2014) [4, 9] . Because ammonia toxicity is strictly exacerbated by elevated pH and temperature levels which dictate the concentration of the highly lethal unionized ammonia (NH 3 ) fraction, controlling its accumulation remains a central objective in aquarium management (Salisu et al., 2023) [19] . The nitrogen cycle serves as the fundamental biological mechanism responsible for mitigating this toxicity (Endut et al., 2010) [6] . Beneficial nitrifying bacteria actively convert toxic ammonia into nitrite and subsequently into less harmful nitrate, through a sequential aerobic process known as nitrification (Endut et al., 2010; Jafari et al., 2024) [6,10] . Establishing and sustaining these stable microbial communities requires highly efficient biological filtration systems that provide adequate surface area for bacterial colonization (Jafari et al., 2024) [10] . Modern recirculating aquaculture operations rely heavily on comprehensive filtration systems that seamlessly combine mechanical, biological and occasionally chemical processes to achieve superior water treatment (Maucieri et al., 2018) [14] . Mechanical filtration plays a vital preliminary role by physically extracting suspended solids, feed residues and environmental debris that would otherwise decompose and elevate the biological oxygen demand (Liltved & Cripps, 1999) [13] . Simultaneously, these systems frequently battle challenges related to severe microbial contamination and nutrient-driven microalgal blooms. The excessive proliferation of phytoplankton results in green water conditions, which drastically reduce visibility, compromise the tank's aesthetic value and interfere with routine fish management practices (Alam et al., 2001; Zhu et al., 2024) [2, 22] . Traditional water exchanges provide only temporary relief and are not economically viable for large-scale operations (Shnel et al., 2002) [20] . Therefore, the industry is increasingly shifting its focus toward alternative, advanced management practices, such as ultraviolet (UV-C) sterilization (Summerfelt et al., 2003) [21] . UV-C radiation effectively damages the nucleic acids of suspended microorganisms, preventing their cellular replication and safely inactivating pathogens without introducing toxic chemical residues into the aquatic environment (Kamel et al., 2022; Zhu et al., 2024) [12,22] . Among freshwater ornamental species, Koi carp (Cyprinus rubrofuscus) stand out as one of the most economically valuable and highly prized fishes globally, renowned for their vibrant color patterns and cultural significance (Jha & Barat, 2005; Hussain et al., 2014) [9,11] . However, because Koi are exceptionally sensitive to sudden environmental fluctuations, inadequate filtration and overcrowding frequently lead to elevated concentrations of toxic ammonia and suspended solids (Jha & Barat, 2005; Fan et al., 2022) [7,11] . This environmental stress suppresses their immunity and adversely affects pigmentation (Abdel-Tawwab et al., 2019; Fan et al., 2022) [1,7] . Furthermore, maintaining optimal dissolved oxygen (DO) is a fundamental requirement for the aerobic metabolism of both the Koi and the beneficial nitrifying bacteria residing in the biofilters (Abdel-Tawwab et al., 2019; Jafari et al., 2024) [1,10] . Despite the significant advancements in Recirculating Aquaculture System (RAS) technology, many small-scale ornamental fish keepers and commercial breeders still face challenges related to the high costs and maintenance complexities of commercial filters. Based on the current literature, there is a distinct necessity to develop affordable, highly efficient filtration systems that utilize locally available natural materials, such as laterite gravel (Nguyen et al., 2021) [15] or cost-effective synthetic options, such as biofoam sponges, integrated with modern UV-C technology. Sufficient scientific evidence evaluating the exact comparative efficacy of these specific media combinations under controlled high-density Koi rearing conditions remains limited. Therefore, the primary aim of the present study is to design and construct a multi-stage filtration system utilizing these accessible media, both with and without UV-C integration, to evaluate their efficacy in maintaining optimal water quality. The findings of this research will contribute a practical, scalable and environmentally sustainable framework for managing physical and biological contaminants in closed aquatic systems. By optimizing these combined biomechanical and UV-C filtration setups, this study provides actionable insights to improve aquarium husbandry, enhance Koi survival and promote the responsible utilization of accessible materials within the broader ornamental aquaculture industry.

Materials and Methods

Experimental design

The filtration system was designed with locally available HDPE plastic buckets measuring 40 cm × 30 cm, capable of holding 20 liters of water. The experiment utilized five circular HDPE water tanks, each with a 110 L capacity, located at the College of Fisheries, Shirgaon, Ratnagiri. Plumbing consisted of 1/2-inch flexible PVC tubing, ball valves and PVC threaded adaptors to regulate the incoming water flow. A 40 W (180-240 V) mini submersible pump was installed to circulate water from the experimental tanks to the filters. The experimental setups were defined as: (C) (Control with no filtration), T 1 (Laterite gravel filter), T 2 (Laterite gravel + UV light), T 3 (Biofoam sponge filter) and T 4 (Biofoam sponge + UV light). Submersible UV-C lights (9W) were integrated into the T 2 and T 4 setups for sterilization.

Experimental fish

Koi carp (Cyprinus rubrofuscus) with an average initial length of approximately 3 cm were selected for the experiment. Prior to stocking, the fish were acclimatized to minimize thermal stress. Subsequently, 30 fish were stocked in each treatment tank, maintaining a stocking density of 30 fish/tank (100L tank). The fish were fed floating commercial pellets containing 32% protein twice daily, in the morning and evening, at 5% of their body weight. The experimental trial was conducted over a 30-day period.

Filter efficacy parameters Water flow test

The water flow rate per minute was calculated by timing with a stopwatch, measuring the volume of water returning to the ornamental fish tank from the filtration unit. Maintaining an adequate flow rate was closely monitored to evaluate the physical operation of each filter design (Othman et al., 2017) [24] .

Green water estimation

The green water algal density, specifically for unicellular algae, was estimated using a spectrophotometric method (Rodrigues et al., 2011). Absorbance readings of the water samples were taken at 650 nm using a spectrophotometer, as this represents the wavelength of maximum sensitivity for quantifying these samples. For obtaining the cell density of the water samples, the absorbance value was applied to a calibrated mathematical model. The formula according to maximum wavelength absorbance by the sample was used as given below:

Cell Density = 𝑒 ln 𝐴+16.726 1.0219

 Cell Density = Algal concentration (millions of cells/ml)  A= Absorbance at 650 nm Total dissolved solids and total suspended solids Total Dissolved Solids were measured by using a TDS meter to monitor water quality (Budianto et al., 2023) [23] . Total Suspended Solids (TSS) were determined by passing a 100 ml water sample through a pre-weighed glass fiber filter, which was subsequently dried at 103-105 °C for a minimum of one hour, after which its final weight was recorded (APHA, 2005) [3] .

Temperature (°C)

Water temperature was recorded daily by immersing a calibrated digital thermometer (MEXTECH DT-9) directly into the water tank until a completely stable reading was visually obtained (Nuwansi et al., 2016) [16] .

pH

The pH was measured daily using a calibrated digital pH meter, where a water sample was collected in a beaker and the electrode was submerged until the reading stabilized (Nuwansi et al., 2016) [16] .

Dissolved Oxygen (DO) (mg/L) DO was determined daily via the Winkler titrimetric method (APHA, 2005) [3] . Water was carefully collected in DO bottles to avoid air bubbles. The oxygen was fixed utilizing 1 ml of Winkler A (manganese sulphate) and 1 ml of Winkler B (alkaline potassium iodide). After a 5-minute floc settling period, the floc was dissolved with 1 ml of H₂SO₄ and a 20 ml aliquot was titrated against an N/40 standard sodium thiosulphate solution using a starch indicator.

Alkalinity was estimated weekly via the titrimetric method (APHA, 2005) [3] . A 100 ml sample received 0.5 ml of phenolphthalein and was titrated against N/50 H₂SO₄ until the pink color vanished. Subsequently, two drops of methyl orange were added and titration continued until the solution transitioned from yellow to orange.

Hardness was measured weekly using the titrimetric method (APHA, 2005) [3] . A 100 ml sample was mixed with 1 ml of ammonia buffer and 3 drops of Eriochrome Black-T indicator. The resulting wine-red solution was titrated against a standard EDTA solution until the color shifted to blue.

Ammonia

Ammonia concentration was determined using the phenate method (APHA, 2005) [3] . In 25 ml of the water sample, 1 ml of phenol solution, 1 ml of sodium nitroprusside solution and 2.5 ml of oxidizing solution (alkaline citrate and sodium hypochlorite) were added and mixed thoroughly. The mixture was covered, kept in the dark for 1 hour and absorbance was measured at 640 nm on a spectrophotometer. Sample concentrations were derived by comparison with a constructed standard curve.

Nitrite concentration was determined weekly utilizing the colorimetric method (APHA, 2005) [3] . A 2 ml portion of a coloring reagent (containing phosphoric acid, sulfanilamide and N-(1-naphthyl)-ethylenediamine dihydrochloride) was added to a 50 ml water sample. Absorbance was then measured at 543 nm using a Shimadzu UV-1780 spectrophotometer and concentrations were calculated via a standard calibration curve.

Nitrate was analyzed weekly using the ultraviolet spectrophotometric screening method (APHA, 2005) [3] . In a 50 ml sample, 1 ml of 1M HCl was added and absorbance was read at 220 nm and 275 nm using the spectrophotometer. The final absorbance attributable to nitrate was calculated by subtracting twice the 275 nm reading from the 220 nm reading.

Growth parameters of the fish

Fish growth parameters were monitored and recorded weekly throughout the study.

Total length (from mouth tip to caudal fin) was measured using a standard scale. Gain (cm) was calculated by subtracting the initial average length from the final average length (Jha and Barat, 2005) [11] .

Individual fish mass was recorded using a digital analytical balance (iScale DM-3). Gain (g) was calculated by subtracting the initial average weight from the final average weight (Jha and Barat, 2005) [11] .

The percentage of surviving fish was determined by dividing the final number of fish by the initial number stocked, then multiplying by 100 (Jha and Barat, 2005) [11] .

Statistical analysis

All collected data were statistically analyzed using One-way ANOVA to determine whether significant differences existed among the independent treatment groups. Data analysis was carried out using SPSS-16 software. Significance of differences was defined at (p < 0.05).

Results

The bio-mechanical filtration systems demonstrated varying degrees of efficacy in maintaining comprehensive water parameters over the 30-day period. UV-integrated systems were remarkably successful at maintaining water clarity by suppressing algal blooms. Water flow test results showed the UV-integrated laterite gravel filter (T 2 ) sustained the highest flow rate at 3.75 ± 0.03 L/min, whereas the biofoam filter (T 3 ) exhibited the lowest flow at 3.17 ± 0.02 L/min. Green water estimation in the control (C) spiked to 2.70 ± 0.01 millions of cells/ml, whereas the UV-integrated systems remained exceptionally clear, with T 2 at 0.04 ± 0.01 and T 4 at 0.08 ± 0.01 millions of cells/ml. Furthermore, T 4 showed the lowest accumulation of Total Dissolved Solids (165.47 ± 0.46 ppm) and T 2 demonstrated the lowest Total Suspended Solids (98 ± 1.00 ppm) among the filtered tanks. Regarding physicochemical parameters, water temperature remained highly stable across all setups, fluctuating safely between a minimum of 28.84 ± 0.12 °C in T 1 and a maximum of 31.97 ± 0.16 °C in (C). The pH exhibited a continuous rise across all setups; however, the unfiltered (C) reached a highly alkaline 8.46 ± 0.06, while the UVintegrated T 2 and T 4 maintained more stable final values of 7.45 ± 0.09 and 7.67 ± 0.06, respectively. Dissolved oxygen (DO) increased across all tanks, peaking at 8.17 ± 0.06 mg/L in C and 7.77 ± 0.06 mg/L in T 3 due to higher photosynthetic algal activity, whereas T 2 and T 4 ended at a healthy 6.97 ± 0.06 mg/L and 7.08 ± 0.06 mg/L. Total alkalinity and hardness also showed gradual upward trends over the trial, with alkalinity reaching 201.24 ± 0.06 mg CaCO 3 /L in T 2 and hardness peaking at 161 ± 1.00 mg CaCO 3 /L in T 3 by day 29. Nitrogenous compounds were effectively managed by the active filtration systems. Ammonia levels surged in the non-filtered control, reaching 1.84 ± 0.02 mg/L. While standalone physical filters (T 1 and T 3 ) experienced mid-trial ammonia spikes (peaking at 1.60 ± 0.01 mg/L and 1.53 ± 0.01 mg/L, respectively) before stabilizing, the UV-integrated biofoam filter (T 4 ) maintained the most stable and lowest ammonia profile throughout the study, peaking at only 0.22 ± 0.01 mg/L. Similarly, nitrite peaked on day 8 in the T 1 and T 3 setups (0.45 ± 0.01 mg/L and 0.55 ± 0.01 mg/L) but was successfully reduced and maintained at a safe baseline of 0.02 ± 0.01 mg/L across all filtered treatments by the end of the experiment. Nitrate concentrations progressively decreased in all tanks, with the UV-integrated T 2 and T 4 systems achieving the lowest final levels of 0.48 ± 0.01 mg/L and 0.55 ± 0.01 mg/L. Absolute growth parameters, including length and weight gain, displayed positive trajectories across all tanks. Final absolute lengths ranged from 4.07 ± 0.20 cm to 4.40 ± 0.35 cm and final weights reached between 1.05 ± 0.13 g and 1.40 ± 0.16 g, indicating that the filtration conditions supported normal physiological development. Furthermore, the survival rate remained at a perfect 100% for all treatments during the entire 30-day period.

Discussion

The 30-day experiment showed that UV-integrated filtration systems (T 2 and T 4 ) significantly outperformed standalone physical media. Initial flow rates of approximately 4 L/min decreased by day 30; however, the gravel systems (T 1 and T 2 ) sustained higher flows (3.39 and 3.75 L/min) than the biofoam setups due to reduced clogging, supporting Maucieri et al. (2018) [14] , who noted that 0.8-8.0 L/min optimizes system performance. (2022) [25] . Notably, UV-C integration in T 2 and T 4 nearly eliminated microalgae (0.04 and 0.08 million cells/mL), while (C) suffered severe blooms (2.70 million cells/mL) that standalone mechanical filters (T 1 , T 3 ) could not arrest. This validates findings by Shnel et al. (2002) [20] and Summerfelt (2003) [21] , regarding UV's efficacy in disrupting microalgal reproductive cycles. Physicochemical parameters remained favorable: temperatures were stable (28.8 °C-31.9 °C) and dissolved oxygen stayed safely above 6.5 mg/L across all setups. Unchecked algal photosynthesis drove alkaline pH spikes in (C) and T 3 (8.46 and 8.07), whereas the UV systems maintained optimal pH levels. While (C) ammonia reached toxic concentrations (1.84 mg/L) and standalone filters exhibited mid-cycle spikes, T 4 maintained exceptional stability, with ammonia peaking at 0.22 mg/L and nitrite stabilizing at a safe 0.02 mg/L baseline. This confirms that optimized biofiltration securely manages nitrogenous wastes within strict limits, as demonstrated by Hussain et al. (2014) [9] and Endut et al. (2010) [6] . Ultimately, the Koi exhibited 100% survival, absolute length gains of 1.0-1.33 cm and weight gains of 0.50-0.92 g. Statistical uniformity in these growth parameters indicates that the integrated filtration setups provided stable, stress-free environments highly conducive to normal physiological development.

Conclusion

The present study successfully designed and evaluated a cost-effective, multi-stage filtration system for ornamental aquaculture. Based on the comprehensive water quality and efficacy assessments: Filtration setups integrating ultraviolet (UV) irradiation (T 2 and T 4 ) exhibited unparalleled efficiency in mitigating hazardous parameters, specifically Total Suspended Solids (TSS), ammonia and microalgal green water blooms. The synergistic combination of highly porous biofoam sponge with continuous UV sterilization (T 4 ) proved to be the most optimal configuration for stabilizing Total Dissolved Solids (TDS) while preventing toxic nitrogenous spikes. Standalone mechanical filtration is insufficient for managing high-density phytoplankton. UV sterilization is absolutely essential for disrupting algal reproductive cycles and preventing the severe pH fluctuations associated with green water outbreaks. These findings hold significant practical implications for the aquaculture industry. Utilizing affordable, locally sourced materials in tandem with UV technology offers a scalable, sustainable and highly effective water management strategy. This model encourages future entrepreneurial innovation in the design of efficient bio-mechanical filters for intensive ornamental fish cultivation.

Acknowledgement

AcknowledgementsThe author also extends their deep sense of gratitude to the Professor, Department of Fisheries Engineering, College of Fisheries, Ratnagiri, for their valuable guidance. The first author wishes to express his heartfelt thanks to the Associate Dean and Major Advisors, College of Fisheries, Ratnagiri, for their constant support, encouragement throughout the course of the study.

References

1. Abdel-Tawwab M, Monier MN, Hoseinifar SH, Faggio C. Fish response to hypoxia stress: growth, physiological and immunological biomarkers. Fish Physiol Biochem. 2019;45(3):997-1013.
2. Alam MZB, Otaki M, Furumai H, Ohgaki S. Direct and indirect inactivation of Microcystis aeruginosa by UVradiation. Water Res. 2001;35(4):1008-1014.
3. APHA. Standard methods for the examination of water and wastewater. 20th ed. Arlington, VA: American Public Health Association; 2005. p. 1-1134.
4. Bhatnagar A, Devi P. Water quality guidelines for the management of pond fish culture. Int J Environ Sci. 2013;3(6):1980-2009.
5. Elmitwalli TA, Van Leeuwen M, Kujawa-Roeleveld K, Sanders W, Zeeman G. Anaerobic biodegradability and digestion in accumulation systems for concentrated black water and kitchen organic-wastes. Water Sci Technol. 2006;53(8):167-175.
6. Endut A, Jusoh A, Ali N, Nik WW, Hassan A. A study on the optimal hydraulic loading rate and plant ratios in recirculation aquaponic system. Bioresour Technol. 2010;101(5):1511-1517.
7. Fan Z, Wu D, Li J, Li C, Zheng X, Wang L. Phosphorus nutrition in Songpu mirror carp (Cyprinus carpio Songpu) during chronic carbonate alkalinity stress: effects on growth, intestinal immunity, physical barrier function and intestinal microflora. Front Immunol. 2022;13:900793.
8. Hemal MM, Rahman A, Nurjahan, Islam F, Ahmed S, Kaiser MS, et al. An integrated smart pond water quality monitoring and fish farming recommendation aquabot system. Sensors. 2024;24(11):3682.
9. Hussain T, Verma AK, Tiwari VK, Prakash C, Rathore G, Shete AP, et al. Optimizing koi carp, Cyprinus carpio var. koi (Linnaeus, 1758), stocking density and nutrient recycling with spinach in an aquaponic system. J World Aquac Soc. 2014;45(6):652-661.
10. Jafari L, Montjourides MA, Hosfeld CD, Attramadal K, Fivelstad S, Dahle H. Biofilter and degasser performance at different alkalinity levels in a brackish water pilot scale recirculating aquaculture system (RAS) for post-smolt Atlantic salmon. Aquac Eng. 2024;106:102407.
11. Jha P, Barat S. The effect of stocking density on growth, survival rate and number of marketable fish produced of koi carps, Cyprinus carpio var. koi in concrete tanks. J Appl Aquac. 2005;17(3):89-102.
12. Kamel A, Fuentes M, Palacios AM, Rodrigo MJ, Vivar M. Deactivating environmental strains of Escherichia coli, Enterococcus faecalis and Clostridium perfringens from a real wastewater effluent using UV-LEDs. Heliyon. 2022;8(12).
13. Liltved H, Cripps SJ. Removal of particle-associated bacteria by prefiltration and ultraviolet irradiation. Aquac Res. 1999;30(6):445-450.
14. Maucieri C, Nicoletto C, Junge R, Schmautz Z, Sambo P, Borin M. Hydroponic systems and water management in aquaponics: a review. Ital J Agron. 2018;13(1):1012.
15. Nguyen TH, Nguyen AT, Loganathan P, Nguyen TV, Vigneswaran S, Nguyen THH, et al. Low-cost lateriteladen household filters for removing arsenic from groundwater in Vietnam and waste management. Process Saf Environ Prot. 2021;152:154-163.
16. Nuwansi KKT, Verma AK, Prakash C, Tiwari VK, Chandrakant MH, Shete AP, et al. Effect of water flow rate on polyculture of koi carp (Cyprinus carpio var. koi) and goldfish (Carassius auratus) with water spinach (Ipomoea aquatica) in recirculating aquaponic system. Aquac Int. 2016;24(1):385-393.
17. Oertli B, Decrey M, Demierre E, Fahy JC, Gallinelli P, Vasco F, et al. Ornamental ponds as nature-based solutions to implement in cities. Sci Total Environ. 2023;888:164300.
18. Rodrigues LHR, Arenzon A, Raya-Rodrigues MT, Fontoura NF. Algal density assessed by spectrophotometry: a calibration curve for the unicellular algae Pseudokirchneriella subcapitata. J Environ Chem Ecotoxicol. 2011.
19. Salisu A, Umar BI, Zhong Z, Yu C, Wang Y. Charcoal stabilizes pH and improves nutrients removal of bacteria-microalgae interaction: a potential for improving water quality in aquaculture. Environ Eng Res. 2023;28(6).
20. Shnel N, Barak Y, Ezer T, Dafni Z, van Rijn J. Design and performance of a zero-discharge tilapia recirculating system. Aquac Eng. 2002;26(3):191-203.
21. Summerfelt ST, Davidson J, Waldrop T. Evaluation of full-scale carbon dioxide stripping columns in a coldwater recirculating system. Aquac Eng. 2003;28(34):155-169.
22. Zhu Y, Ding J, Wang X, Wang X, Cao H, Teng F, et al. Optimizing UVA and UVC synergy for effective control of harmful cyanobacterial blooms. Environ Sci Ecotechnol. 2024;22:100455.
23. Budianto. Bibliometric and literature review of financing risk in Islamic banking. JPS (Jurnal Perbankan Syariah). 2023 Jan 1.
24. Othman, Ariff AB, Rios-Solis L, Halim M. Extractive fermentation of lactic acid in lactic acid bacteria cultivation: A review. Frontiers in Microbiology. 2017 Nov 20;8:2285.
25. Ezhilmathi, Samy SS. Evaluating performance metrics in classifying bitcoin mixing services using decision tree algorithm. In2022 international conference on electronic systems and intelligent computing (ICESIC) 2022 Apr 22 (pp. 304-310). IEEE.