Select the search type
  • Site
  • Web

Student Project

The use of Bivalves in Assisting Filtration of Waste Solids within Fish Aquaculture Systems

Mark Scanlan 2016


Aquaculture is increasingly being incorporated into our primary industries in order to meet growing demands within the seafood industry. While aquaculture technology has greatly advanced in previous decades, it is still far from perfect and further innovation is required for maximum efficiency. The filtration and removal of waste solids is a major requirement for all aquaculture facilities in order to maintain clean water and healthy livestock. This paper will explore the ability of Trichomya hirsute, a filter feeding bivalve, to filter waste solids and algae from its environment, with aims to assess the suitability of mussels to be incorporated into aquaculture systems to assist in filtration. 6 individual mussels were placed in 6 different containers, 3 containers were treated as controls, having only algae added, while 3 individuals had treatments introduced to their containers, treatment consisted of a fish food, pellet dilution. Concentration changes were measured over a period of 3 hours at 30 minute intervals, 1ml samples were collected and changes were measured using a spectrophotometer to measure turbidity. A total decline in algal concentration was seen within control containers, while a slight increase in concentration was seen within treatment containers. Results indicated that there was successful filtration of algae within control containers, while within treatment containers, there was no evidence of filtration of waste solids. 


Aquaculture is a fast growing industry, being increasingly relied on to meet global seafood demands. Future estimations have been made that while seafood supplied via the capture of wild stocks will remain similar to those seen today of roughly 90 million tonnes per year, seafood supplied by aquaculture will nearly double by 2030, rising from 52 million tonnes to 93 million tonnes per year (FAO 2014). These predictions create a high demand for further improvements and increased reliability and efficiency in technology used within aquaculture systems. 

Current aquaculture systems have many issues involving the removal of waste solids. Waste solids are produced within the system and include uneaten food, fish faecal matter, and algae. Within sustainable closed aquaculture systems, waste solids need to be removed in order to safely recirculate water within the system. Failing to do so will lead to an increase in nutrients within the system, increasing bacteria density, potentially leading to disease and sickness within fish, but also providing threats to human consumers (Capriulo et. al, 2002). Current systems are in place to filter waste solids out of the system, however, all systems require cleaning and maintenance to continue running efficiently (Ebeling & Timmons 2012). 

Bivalves, more specifically mussels, are effective filter feeders within the ocean, capturing particles, bacteria and small plankton as a food source (Jorgensen 1996). Naturally in the wild, the act of filter feeding assists in particle and nutrient recruitment to the local ecosystem from the water column, but also assists in control of bacteria, algae and plankton densities. Mussels themselves also serve as a food source for many organisms, having both vertebrate and invertebrate predators. The ability for mussels to filter feed so efficiently opens the possibility of using colonies of these organisms within aquaculture systems as a possible waste solids filter. While filtration rate of waste solids may not efficiently reach 100%, the proposal still stands of mussels becoming a filtration buffer, reducing cleaning and maintenance efforts required by waste solid filters. Depending on harvested species, it is also possible that as these bivalves become larger and reach maturity, they could be used as a food source for the fish, minimising waste created by the system.  Mussels could also be sold for human consumption, further increasing potential profits produced by the system.  

This paper will attempt to gain insight into the ability of Trichomya hirsute, also known as the hairy mussel, to be used as an aquaculture system waste solids filtration buffer. The paper will expose individual hairy mussels to dissolved fish feed and filtration rates will be measured as a representation of their ability to filter out waste solids from an aquaculture system. 

Materials and Methods

In order to determine Bivalves as a suitable organism to be used as a filtration buffer within fish aquaculture systems, bivalves were exposed to fish food and rate of filtration was measured. 

Fish pellet Dilution
In order to create a concentrated dilution of fish feed, 1.2 grams (9 pellets) of Hikari cichlid gold, large pellets (40% Crude Protein, 4% Crude Fat, 4% Crude Fiber, 10% Moister, 15% Ash, 1% Phosphorous) were grinded into powder form. This powder was then added to a blender with 150ml of water and adequately mixed to form a homogenous solution. The solution was then separated into 3 individual 50ml conical centrifuge tubes, care was taken as to not add unmixed sediments to the tubes. 

Experimental Procedure
The experiment was setup using six individual Trichomya hirsuta submerged in small, shallow white containers. Three containers were setup as controls, and three containers as treatments. Each control container had 450ml of seawater added while each treatment container had only 400ml (450ml total with dilution) added. One individual hairy mussel was then also added to each container. Individuals within control containers were exposed to 1ml of Pavlova 1800 algae sized between 4-7 microns, which was added to each container and stirred for 10 seconds. As algae is a known food source for bivalves, they were used as a control to prove feeding had occurred. Within treatment containers, 50ml of fish pellet dilution was added to each container and water was stirred until all sediments had been suspended. 1ml samples were taken from each container at 30 minute intervals from time 0 for a total of 3 hours, these samples were marked with both time and treatment type and stored for later examination. Containers were mixed thoroughly at each sample interval until all sediments resuspended. 

In order to measure changes in substance concentrations within each container, turbidity measurements were taken. A Spectrophotometer was used to test each 1ml sample at a wavelength of 550nm. Wavelength readings were recorded along with time the sample was taken. Reduction in readings over time would indicate a decline in concentration and therefore provide evidence of the bivalves consuming the algae and fish pellet dilution. 

Changes in concentration at each time interval were averaged across each treatment type. This allowed a more precise representation of the true value which would be filtered within the given timeframe for each treatment type. Regression analysis was performed using R studio software for both Treatment types against time. Total change in concentration from time 0 to 3 hours was also compared to the weight of individuals using linear regression in order to test if size influenced filtration rates.  


Results for the Control group, containing algae showed a decreasing trend in wavelength absorbance (figure 1). A mean initial wavelength absorbance reading of 0.417nm with a SE (standard error) of ±0.048, was recorded, with a final average wavelength absorbance of 0.341nm, SE±0.033. A linear regression analysis was performed on wavelength absorbance reading against time for Control groups, which showed a non-significant p-value of 0.0796 and R² of 0.49 (F statistic=4.818, DF=4). 

The treatment group showed a horizontal trend, with a slight increase in wavelength absorbance (figure 1). Initial averaged wavelength absorbance recordings of 0.141nm, SE±0.002 were obtained at time 0 hours, with a slight increase at the 3 hour mark with averaged absorbed wavelength readings of 0.159nm, SE±0.007. Analysis of this data also gave a non-significant p-value of 0.514, with an R² of 0.08 (F statistic=0.49, DF=5).

When weight was compared to total concentration change in figure 2, no trend was evident. Given the results shown in figure 2, treatment was ignored when concentration change was compared to weight, as within the treatment group, all individuals had negative wavelength absorbance changes from time 0 to 3 hours (concentration increased). Within the control group, individual 1 with a weight of 6.8 grams saw a change in wavelength absorbance of 0.164nm, while individual 2, weighing 10.9 grams saw a wavelength absorbance change of 0.06nm. Individual 3 weighing 9.78g, contributed to a total wavelength absorbance change of 0.002nm. Results for relationship between size and concentration change within the control group showed a non-significant relationship with a p-value of 0.4 (F statistic=1.89, DF=1).
Figure 1
Figure 2


To provide evidence of the usability of bivalves in assisting filtration of waste solids within fish aquaculture systems, this report assessed the ability for Trichomya hirsute to filter out a diluted fish food solution. Results were conclusive of the inability of the hairy mussel to filter any waste solids found throughout the diluted fish food solution. This indicates that hairy mussels would not be suitable as a buffer for waste solid filtration. Individuals exposed to algal dilutions saw a negative trend showing that mussels were in fact feeding and successfully reduced overall algal concentration within the control group. 

The lack of results indicating an inability of mussels to filter waste solids in this experiment may not be reliably conclusive. As seen depicted within figure 1, there was little variation in concentration seen within treatment containers, and overall, an increase in concentration was seen from time 0 to the 3 hour mark. An increase in concentration indicates that there was an input of sediments during the experiment, possibly originating from the individual hairy mussel itself as waste product or faeces. Overall, there was no decrease in concentration seen within the fish pellet dilution treatment group and therefor evident that Trichomya hirsute was unable or disinterested in filtering out waste solids from its environment during the experiment. 

Control treatments showed a negative trend in concentration change, with absorbed wavelength declining over time. Slight increases were seen throughout interval testing, which may have indicated a similar situation as seen within treatment containers, where increase of turbidity was attributed to waste product provided by the individual being studied. Overall, from time 0 to the 3 hour mark, a decrease in algal concentration within control containers was seen, from an averaged absorbed wavelength of 0.417nm at time 0, to 0.341nm seen from an average of final samples taken at the 3 hour mark. This provided evidence that mussels were alive and actively feeding throughout the experiment. An experiment conducted by Soto & Mena found that when mussels were placed within a simulated laboratory salmon farm for algal filtration, the mussels were able to reduce total eutrophication levels from hypereutrophic to oligotrophic within 18 days, with mussels reducing overall chlorophyll-a, phosphorus and ammonium concentrations via filtration (Soto & Mena, 1999). Within the study done by Soto & Mena, chlorophyll-a concentrations represented algal concentrations, and therefore the study also saw reductions in algal concentrations due to filtration by the bivalves. 

When weight was compared to concentration change, treatment was ignored due to the increase in concentration change seen across all treatment containers. When looking at individual weight compared to concentration change in control containers, there was no correlation. 

While efforts were made to ensure samples were taken from a constant position to reduce variation in concentration created by collection location, it is possible that this played a role in distorting the results. Prior to sample collection, mixing of the container would take place in an attempt to create a homogenous concentration across the container, however, it is possible that just by random chance, some samples may have been more concentrated than others. This would ultimately lead to distorted results in wavelength absorbance changes. Evidence of this error can be seen within the fish pellet dilution treatment, where final concentrations exceeded initial concentrations, indicating that either there was an input during the experiment, or most likely mixing of solution failed to create a homogenous concentration across the container. Concentration inconsistency due to collection location can also be seen within control results, where fluctuations of increasing and decreasing concentrations are seen across time intervals. 

Many factors could attribute to the results obtained, for example, individuals may have gained sufficient nutrients prior to experimental procedure, reducing the amount required by the organism during the experiment timeframe. A study conducted by Riisgard, Kittner & Seerup on filtration rates within bivalves found that when algal concentrations fell below a certain level, bivalves would cease filtration activity (Riisgard, Kittner & Seerup, 2003). 

With regards to concentration change according to weight, it is evident that perhaps the use of one individual to each container could create a high amount of variability just by random chance. To further increase reliability when testing this variable, it would be beneficial to have multiple individuals (4-6) of similar weight within each container, this would allow an increased sample size of individuals, allowing a closer result to that of the actual change in concentration attributed to weight. 

Larger sample sizes of both containers, and individuals within each container would also allow a greater population size, and upon statistical analysis, a more accurate description of the actual averages would be obtained. It is there for recommended that any further studies include larger sample sizes of both containers, and individuals within each container. 

Alternative phylums may also be more efficient as aquaculture waste solid filters, such as Porifera, which are also effective filter feeders, feeding on plankton, bacteria and dissolved organic particles within the water column (Hill & Hill, 2009). 

Concluding this report, it was decided that results supplied by this experiment may not be considered reliable enough to conclude the suitability of bivalves as waste solid filtration buffers within aquaculture. It is, however, evident that within this study, bivalves were proven to consume algae and reduce overall turbidity within control containers. This could still be regarded as a useful ability which could potentially be used within aquaculture systems for reduction in algal concentrations. 



Capriulo, G. Smith, G. Troy, R. Wikfors, G. Pellet, J. Yarish, C 2002. The Planktonic food web structure of a temperate zone estuary, and its alteration due to eutrophication. Hydrobiologia. Vol. 475, Issue 1, pp 263-333.

Ebeling, J. Timmons, M 2012. ‘Recirculating Aquaculture Systems’, in Tidwell J, Aquaculture Production Systems. Pp 249-250. 

FAO (2014). State of the World Fisheries and Aquaculture. FAO, Rome.

Hill, M. Hill, A 2009. ‘Porfera’, in Likens G, Encyclopedia of Inland Waters. Pp 423-432.

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

Queensland Museum 2016, Hairy Mussel, Viewed 28 May 2016, <

Rissgard, U. Kittner, C. Seerup, D 2003. Regulation of opening state and filtration rate in filter-feeding bivalves (cardium edule, Mytilus edulis, Mya arenaria) in response to low algal concentration. Journal of Experimental Marine Biology and Ecology. Vol. 284, Issue 1-2, pp 105-127.

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.