Select the search type
  • Site
  • Web

Student Project

The use of Katelysia scalarina to assist the filtration of waste solids such as fish food in fish recirculating aquaculture systems

Loup Paitard 2017


The world fisheries have been drastically exploited in recent decades and many populations are harvested at their carrying capacity in order to meet growing demand. Aquaculture, which has become increasingly sought-after in the past three decades, is a promising solution for the wellbeing of both the oceans and human needs (Eliasson, 2015). However, this relatively new technology is far from perfect and requires great innovation to reach its maximum efficiency. One of these required innovations is a solution to the waste of fish feed in recirculating aquaculture systems (RAS). On one hand, the uneaten fish pellet causes a big economic loss for aquaculture farms. On the other hand, this highly nutritious feed dissolves into the water and endangers the fish in the farm (Li, 2017). This paper will study the use of K. scalarina, a filter feeder bivalve, as a filter for fish food wastes. We aim to acquire test the suitability of this clam to assist in the filtration of RAS.  Twenty-two individual clams were placed in 22 containers; 6 containers were set up as control with sea water and clams. Another only contained fish dilution and sea water. The remaining 16 containers housed a clam in a sea water and fish feed solution. Ten containers were composed of 25% fish feed solution and 6 containers held a 50% fish feed solution. Concentration changes were sampled every 30 minutes for a period of 3h, 1ml samples were collected from each container, and changes in optical density was measured using a spectrophotometer. The 25% treatment showed a strong decrease in turbidity. Compared to the 50% treatment, which did not significantly decrease in turbidity. The control, composed of sea water and a clam, slightly decreased in turbidity. However, the fish feed and sea water increased significantly over time. Results indicated that K. scalarina, could be used to assist filtration of fish feed waste from a RAS.


Fisheries and aquaculture are an essential source of nutrition for hundreds of millions of people throughout the world. As the world’s population increases so will the consumption, and demand, of fish – as has occurred in the past three decades. As the limits of sustainable fishing are being reached – and even exceeded – the demand is ever-increasing and seafood derived from aquaculture is becoming a necessity if we are to maintain our ocean biodiversity (Eliasson, 2015). To date, half of all fish consumption worldwide comes from aquaculture farms (FAO, 2016). However, as the aquaculture industry grows, the water quality of these farms is becoming more important as farmers are attempting to grow more fish in less space (Lekang, 2013). Recirculating aquaculture systems (RAS) are a type of aquaculture farming that has been used for three decades. This fish farming technique reduces the space and water requirements but also lowers the water quality very quickly (Thorarensen, 2007). This water has to be well managed to prevent disease outbreak and other water quality stressors (Lekang, 2013).

The intensification of aquaculture industries has drawn greater attention to water quality. Higher fish production causes the water quality to deteriorate at an expedited rate (Lekang, 2013). A wide variety of methods have been invented to remove impurities from this system. These methods are used to manage two different types of wastes, particulate or dissolved. (Lekang, 2013). Waste solids are composed of uneaten fish pellets, fecal matter, and algae. Recirculating aquaculture systems (RAS), due to their enclosed water system, need to be filtered and carefully managed. Otherwise, the waste solids would increase the concentration of nutrients, leading to an increase in bacteria. The outcomes from this issue could lead to a massive die-off of fish in the farm. As discussed in a paper by Dr. Li Dawei (2017), fish food leftovers are a serious problem for aquaculture. This issue encompasses two main problems, the first one being the economic loss of this fish food because it accounts for the largest aquaculture expenses. The second one being the pollution of the water, caused by this highly nutritious pellet, that lower the water quality of the environment these aquatic organisms reside in (Li, 2017).

Bivalves such as K. scalarina feed through a set of siphon and fin-mesh gills. The water passes through their inhalant siphon, is filtered by the gills, and leaves through the exhalant siphon (Cranford, 2011). The mollusks’ main food source is phytoplankton and organic particles found in the water column. The feeding, and accompanying water filtration, could be a possible solution to the problem of waste solids in RAS. The clams’ influence on the water quality could be a tremendous benefit for the system. Furthermore, once these species of clam reach maturity, they could be reused as fish feed or sold to seafood stores at about AUD$32/Kg (Frank's Seafood).

This study aims to measure the ability of K. scalarina, also known as Venus shells, to be used as a waste solid filter in an aquaculture system. K. scalarina will be exposed to fish pellets and the turbidity over time will be measured and used as a determinant of their ability to filter fish food from an aquaculture system.

Materials and Methods

FishPellet Dilution:

The fish feed for the clams was prepared with specialized tilapia fish feed. To create this solution 1.2 grams of fish pellet was ground into powder and diluted to 50ml of seawater, sourced from the marine aquarium at the University of Queensland. Eighteen conical tubes of 50ml of seawater and ground fish pellets were stored at 4°C for one week to dissolve the fish feed. One-hour post-experiment all the fish feed solution was filtered out using a coffee filter, which allows particles of approximately 10-15 microns to pass.  All the fish feed solutions were then mixed into a 1L Erlenmeyer flask.
The dilution of 25ml of fish feed with 75ml (the 25% solution) of seawater gave in average and optical density for an absorbance wavelength of 600nm, 0.163nm.
The dilution of 50ml of fish feed with 50ml (the 50% solution) of seawater gave in average and optical density for absorbance wavelength of 600nm, 0.329nm. The particles size and shape were not measured and therefore not taken into account in the table below.

Nutritional Declaration: Tilapia fish pellet 

Crude Protein (%)


Crude Fat (%)


Nitrogen Free Extract (%)


Fibre (%)


Ash (%)


Phosphorus (%)


Gross Energy (MJ/kg)


Digestible Energy (MJ/kg)



Experimental design:

The experiment used 22 K. scalarina submerged in 100ml containers. Twelve containers were set up as controls. One measured the change in turbidity with just fish feed (25%) and sea water. The other measured the turbidity of clams in seawater. The two treatments were composed of 16 individuals, each separated into its own 100ml container. One treatment was composed of 25ml of fish feed added to 75ml of seawater. The second treatment was composed of 50ml of fish feed added to 50ml of seawater. Every 30 minutes 1ml was sampled from each container during a period of three hours. No containers and their solution were mixed as this could stress the clams during their feeding process.

The change in turbidity during the experiment was analyzed using absorbance, measured with a spectrophotometer. Each 1ml sample was tested at a wavelength of 600nm. The blank for the spectrophotometer was taken from the water of the aquarium where the clams came from. The decrease of absorbance measured over time was used as an indication of fish pellet consumption by K. scalarina. This should ultimately be able to determine the ability of this type of clam to be a filter for waste solids in fish aquaculture.

Number of replicates per treatments: 

  • Sea water (75ml) + Fish pellet(25ml) + One clam  (10)
  • Sea water (50ml) + Fish pellet(50ml) + One clam   (6)
  • FSW (75ml) + Pellet (25ml)                                   (6)
  • FSW (100ml) + Clam                                            (6)


Concentration change for each treatment was averaged. This gave a better representation of the actual change in turbidity for each treatment over time. If the clam did not open during the 3 hour time period then the result from that replicate was removed from the analysis. The statistical analysis was executed using the software “R”. A linear regression test was used to analyze each treatment over time.

Figure 1
Figure 2


Sea water(75ml) + Fish pellet (25ml) + clam

Results from the treatment (SW + Pellet(25%) + clam) showed a strong decreasing trend in average wavelength absorbance (figure 3). At time T=0 the average optical density measured was 0.1636nm, SD± 0.0021, and at time T=180, 0.0158nm SD±0.0080. The outcomes from the linear regression analysis showed a significant p-value of 0.009145 and an R-squared of 0.7727, (F-statistic: 17, DF= 5).

Seawater (50ml) + Fish pellet (50ml) + clam

The treatment (SW + Pellet(50%) + clam) show a horizontal trend, with a minor decrease in average wavelength absorbance (figure 3). The mean of optical density measured at time T=0 was 0.3291nm with a SD (standard deviation) of ± 0.0051 and at the final time, T=180, was 0.3026nm, SD± 0.0055. The analysis of these measured showed a non-significant p-value=0.2015 and an R-squared of 0.3018. (F-statistic= 2.162, DF= 5)

FSW(75ml) + Pellet (25ml)

Results from the treatment (SW + Pellet(25%) ) showed an increasing trend in average wavelength absorbance (figure 4). At time T=0 the average wavelength absorbance measured was of 0.1648nm, SD± 0.0026, and at time T=180 it was 0.2038nm SD± 0.0039. The outcomes from the linear regression analysis showed a significant p-value of 0.00066 and an R-squared of 0.9022, (F-statistic=56.33, DF= 5). 

FSW(100ml) + clam

The outcomes of the treatment (clam + Sea Water) show a horizontal trend, with a very slight decrease in wavelength absorbance (figure 5). The mean of optical density measured at time T=0 was 0.022nm with a SD (standard deviation) of ±0.0013 and at the final time, T=180, was -0.0002nm, SD±0.0004. The analysis of this resulted in non-significant p-value of 0.1146; with an R-squared of 0.4216. (F-statistic= 3.644, DF= 5)

Figure 3
Figure 4
Figure 5


K. scalarina was exposed to diluted fish feed and the turbidity over time was recorded as a measure of their ability to filter waste solids from a recirculating aquaculture system. The results show that K. scalarina, also known as Venus shell, could filter waste solids, such as a diluted solution of fish feed (25%). This indicates that K. scalarina would be suitable as a buffer for waste solid filtration in RAS. The diluted solution of fish feed (50%), by contrast, showed no significant decrease in turbidity. This may be because this concentration of fish feed is toxic, causing the clam's gills to get blocked. (Jorgensen, 1996). The seawater and fish pellet (25%) treatment showed a significant increase in optical density over time. This increase could be due to increased bacterial proliferation, as the extremely nutritious fish feed would provide more than adequate sustenance (Monod, 1949). The seawater and clam control also resulted in a significant decrease in turbidity, denoting the efficiency of the filtration ability of K. scalarina.

This species of clam filtered the 25% fish feed solution efficiently on average. However, only six out ten clams were considered in the analysis as the other clams didn’t open (Figure 2). The absorbance wavelength readings showed a decrease in turbidity from 0.163nm, SD± 0.002 at time T=0 to 0.015nm SD±0.008 at time T2=180. This large decrease would be advantageous for aquaculture industries. In fish aquaculture farms protein is the most expensive component, as well as being the main source of nitrogenous pollutant. The amount of protein found in the fish pellets is 38%, crude. If the fish pellets are not completely consumed by the fish grown in the farm the excess may be eaten by the clams (Shpigel, 2007). Resulting in an additional profit or savings, as the clam can be then sold on the market or recycled into fish feed. As mentioned in shpigel, under laboratory conditions clams tend to have an optimal food concentration at which their pumping mechanism is most effective. With a 25% of fish feed the clam were highly active, which wasn’t the case for the 50% fish feed treatment.

The treatment composed of 50% fish feed shows a horizontal trend with two small decreases and two small increases in wavelength absorbance (figure 3). The linear regression showed no significant difference in absorbance wavelengths over time though. The high concentration of fish feed in these containers exposed the clams to high concentrations of nutrients and the high turbidity of these containers did not allow us to determine if the organism were open or not. High concentration of particles in a solution causes bivalves to secrete mucus. As the particles enter the inhalant siphon high concentrations of particles are trapped in the mucus and rejected by the clam, resulting in them being forced back into the medium (Jorgensen, 1996). The two small decreases and increases in absorbance wavelength could be explained by the clam trapping some of the particles and rejecting them afterwards. Moreover, when concentration of suspended particles is very high, a reduction of the valves’ gap occurs, along with the retraction of mantle edge and siphon. This is correlated with a decrease in pumping rate (Jorgensen, 1996).

The sea water and 25% fish feed treatment showed a significant increase in absorbance over time. The absorbance wavelength passed from 0.1648nm, SD± 0.0026 at time T=0 to 0.2038nm SD± 0.0039 at time T=180. The increase is slow but has a strong R2 (0.90) and a significant p-value of 0.00066 (figure 4). This fish feed in this solution did not deposit at the bottom of each container. The increase in absorbance is most likely due to bacterial growth. Bacteria, when in a highly nutritious medium, can duplicate at an exponential rate. The bacterial population in the medium could explain the constant increasing wavelength absorbance over time, even though no change was made.  (Navarro, 2010).

The clams in the seawater control, in which all opened except one, showed a horizontal trend. The experiment showed that the clams have filtered the small number of particles found in their containers (figure 5). However, the result from the linear regression was non-significant. The wavelength absorbance readings at time t=0 were of 0.022nm SD±0.0013 and at time t=180 of -0.0002nm, SD±0.0004. The results were negative at the end of the experiment showing that the clams have filtered the small amount of seawater from the aquarium. This control shows that no particles were added when the clams were set in their containers.

This experiment has limitations; the time of the experiment was three hours. The large decrease in this short amount of time is a very good result. However, in tilapia farming, the fish takes 140 days to mature(Santos, 2013). It would, therefore, be of interest to measure the effect of K. scalarina on the filtration of fish feed over a longer period of time (140 days). Additionally, some replicates for the treatment and control did not open. These results could be due to individuals having obtained enough nutrients before the experiment. The organisms were taken away from the aquarium an hour before the experiment started which isn’t enough to starve bivalves (Jorgensen, 1996). A longer time of starvation could therefore be set up to see the full potential of fish food filtration by bivalves.

This experiment could lead to further research on clams and their efficiency as a biofilter for recirculating aquaculture farms. The ability of K. scalarina to filter different types and densities of fish feed particles could be of interest. In a study on Ruditapes philippinarum, a species of clam, the size of particles was tested. These particles had the same shape, density and chemical composition, but clams preferred particles of a certain size. The outcomes showed that particles larger than 22.5μm were rejected as pseudofaeces (Defossez, 1997).

Algae can become also a problem in aquaculture, K. scalarina can filter this as well, increasing its potential. An experiment on a salmon farm showed that bivalves such as mussels were able to change a hypereutrophic situation to an oligotrophic in 18 days (Soto, 1999). Studying clam filtration of algae could increase our knowledge on the efficiency of K. scalarina as a biofilter in aquaculture.

As discussed by Cranford, bivalves have a strong relationship between size and pumping rate. A small increase in body size results in a large increase in pumping rate (Cranford, 2011). A study could aim to measure the efficiency of different body size clams to find the most efficient biofilters for RSA.

Due to its high importance, the issue of waste water and filtration in aquaculture has been studied by many research groups. One of these has studied the use of polychaete-assisted sand filters as a highly effective waste filter (Palmer, 2010). Sponges (Porifera) could also be of interest due to their high filtration rate, which can reach 75 L/Hr. (NOAA, 2016).

To conclude, this report has shown that K. scalarina has the ability to filter fish feed and would be suitable in assisting waste solid filtration in recirculating aquaculture systems. However, at high concentrations of fish feed, the cut-off being somewhere between 50% and 25%, the clam is ineffectiveFinally, K. scalarina, needs to be tested at a larger scale (more replicates) and for a longer period of time (140days) in order to be validated as an efficient aquaculture filter.


Firstly I would like to thank Bernard and Sandie Degnan, for guiding me through various problems and questions I encountered. Secondly, I would like to thank the students of our marine invertebrates class for taking turns to use the spectrophotometer.


Cranford, P.J., Ward, J.E. and Shumway, S.E., 2011. Bivalve filter feeding: variability and limits of the aquaculture biofilter. Shellfish aquaculture and the environment, pp.81-124.

Defossez, J.M. and Hawkins, A.J.S., 1997. Selective feeding in shellfish: size-dependent rejection of large particles within pseudofaeces from Mytilus edulis, Ruditapes philippinarum and Tapes decussatus. Marine Biology, 129(1), pp.139-147.

Eliasson, Thomas. "Dairy Waste – Feed for Fish?" Swedish University of Agricultural Sciences (2015): n. pag.Print.

Ebeling, James M., and Michael B. Timmons. "Recirculating aquaculture systems." Aquaculture Production Systems (2012): 245-277.

FAO. 2016. The State of World Fisheries and Aquaculture 2016. Contributing to food security and nutrition for all. Rome.200 pp. 

Frank's Seafood, 296, Montague Rd. West End QLD 4101, Web, 2017.

Jorgensen, C 1996. Bivalve Filter Feeding Revisited. Marine Ecology Progress Series. Vol. 142, pp 287- 302. 

Lekang, Odd-Ivar. "Water Quality and Water Treatment." Lekang/Aquaculture Engineering Aquaculture Engineering (2013): 32-36.

Li, D., Xu, L. and Liu, H., 2017. Detection of Uneaten Fish Food Pellets in Underwater Images for Aquaculture. Aquacultural Engineering.

Monod, J., 1949. The growth of bacterial cultures. Annual Reviews in Microbiology, 3(1), pp.371-394.

Navarro Llorens, J.M., Tormo, A. and Martínez-García, E., 2010. Stationary phase in gram-negative bacteria. FEMS microbiology reviews, 34(4), pp.476-495.

Naylor, R.L., Goldburg, R.J., Primavera, J.H., Kautsky, N., Beveridge, M.C., Clay, J., Folke, C., Lubchenco, J., Mooney, H. and Troell, M., 2000. Effect of aquaculture on world fish supplies. Nature, 405(6790), pp.1017-1024.

Palmer, P.J., 2010. Polychaete-assisted sand filters. Aquaculture, 306(1), pp.369-377.

Santos, V.B.D., Mareco, E.A. and Dal Pai Silva, M., 2013. Growth curves of Nile tilapia (Oreochromis niloticus) strains cultivated at different temperatures. Acta Scientiarum. Animal Sciences, 35(3), pp.235-242.

Shpigel, M. and Neori, A., 2007. Microalgae, macroalgae, and bivalves as biofilters in land-based mariculture in Israel. In Ecological and genetic implications of aquaculture activities (pp. 433-446). Springer Netherlands.

 Soto, D. Mena, G 1999. Filter Feeding by the freshwater mussel, Diplodon chilensis, as a biocontrol of salmon farming eutrophication. Aquaculture. Vol. 171, Issue 1-2, pp 65-81.

Thorarensen, Helgi. "WATER QUALITY IN RECIRCULATING AQUACULTURE SYSTEMS FOR ARCTIC CHARR (Salvelinus Alpinus L.) CULTURE." The United Nations University (2007): n. pag. Print.

"Filter-Feeding in Reef Sponges." National Marine Sanctuaries. NOAA, 2016. Web.