Effects of addition of tilapia on the abundance of periphyton in freshwater prawn culture ponds with periphyton substrates

An experiment was conducted to evaluate the effect of addition of tilapia on abundance of periphyton in freshwater prawn, Macrobrachium rosenbergii (de Man) in periphyton based culture system for a period of 120 days at Fisheries Field Laboratory Complex, Bangladesh Agricultural University, Mymensingh. A large pond (83×8.9 m) was drained completely and partitioned by galvanized iron sheet into 18 small ponds of 40 m each; of which 6 ponds were used for this experiment. The experimental ponds were divided into 2 treatments each with 3 randomly selected ponds. The absence and presence (0 and 0.5 individual m) of tilapia (Oreochromis niloticus) were investigated in 40 m ponds stocked with 3 prawn juveniles (5±0.05 g) m with added substrates for periphyton development. A locally formulated and prepared feed containing 30% protein was supplied considering the body weight of prawn only. Addition of periphyton substrates significantly reduced the inorganic N-compounds (TAN, NO2-N, and NO3-N) in water column. Forty six genera of periphyton were identified belonging to the Bacillariophyceae (10), Chlorophyceae (21), Cyanophyceae (7), Euglenophyceae (2), Crustacea (1) and Rotifera (5) with significant difference (P<0.05) of phyto-periphyton except Euglenophyceae and without significant difference (P>0.05) of zoo-periphyton between the treatments. The abundance of periphyton biomass in terms of dry matter, ash, ash free dry matter and chlorophyll-a were significantly higher in tilapia-free ponds comparing to tilapia added ponds. Benthic organisms had no significant difference (p>0.05) between the treatments. Addition of tilapia in periphyton-based system benefited the freshwater prawn culture through (i) reducing toxic inorganic nitrogenous compounds in water (ii) reducing demand for supplemental feed (iii) using periphyton as additional natural feed and, (iv)improving survival and production of prawn and tilapia.


Introduction
Freshwater prawn, Macrobrachium rosenbergii (de Man) is indigenous to the South and South-East Asia, together with Northern Australia and the western Pacific Islands (New 1988).Through successful domestication in late 1960s (Ling, 1969), the culture of freshwater prawn has gained a great popularity worldwide, mostly in the tropical and subtropical regions, with the limited production in temperate regions (D'Abramo et al., 1989).The global production of freshwater prawn has increased gradually (FAO, 1997) with the major production in East and South Asian countries like China, India, Indonesia, Bangladesh, Thailand and the Philippines.
The prawn farming area in Bangladesh is expanding very fast; the expansion rate in last 3 years is about 10% per year.At present approximately 50,000 ha of land and water bodies are used in prawn aquaculture.The freshwater prawn play a vital role in the development of socio-economic conditions of the country through increasing export trade, food production, creation of rural employment and proper utilization of natural resources (Rahman, 2000).In Bangladesh different culture systems are being practiced, such as prawn monoculture, prawn polyculture along with other fishes (mostly carps), prawn aquaculture in paddy fields and prawn aquaculture in paddy fields after paddy harvesting.
Macrobrachium culture presently practiced in Bangladesh may be categorized broadly into two culture methods-Beri (gher) culture and polyculture with carps.The cultivation of freshwater prawn in modified rice fields locally referred to as "gher" has been developed in Bangladesh (BOBP, 1990;Rosenberry, 1992;Rutherford, 1994).Periphyton is very preferable natural food for herbivorous and omnivorous fish species especially for Indian major carp (Azim et al., 2002;Keshavanath et al., 2002), for tilapia (Trewavas, 1982;Ali, 1998;Wahab et al., 1999), and for freshwater giant prawn (Cohen et al., 1983).The use of periphyton substrates in freshwater finfish and prawn production has been found promising (van Dam et al., 2002).Substrates based system can increase freshwater prawn production to a significantly higher level when compared to traditional production system (Tidwell and Bratvold, 2005).Cohen et al. (1983) reported that added substrate in ponds increased prawn production by 14% and average size by 13%.Experiments conducted in Bangladesh showed that vertical substrate addition resulted in prawn survival of increment of 75% and production of 127% in prawn-tilapia polyculture system (Uddin, 2007).Therefore, freshwater prawn productivity can be enhanced through stimulation of suspended and attached bacteria and algae development, and by using them to improve water quality, provide additional food and improve nutrient efficiency.
Introducing substrates for periphyton development (Tidwell et al.;2000, Tidwell andBratvold, 2005;Uddin, 2007) manipulation of C:N ratio (Azim and Little, 2006;Avnimelech, 2007) and combination of both C:N ratio and periphyton substrates in freshwater prawn ponds (Asaduzzaman et al., 2008) have been found promising.These techniques require installation of hard substrates and application of cheap carbohydrate resources which are available within the farmer's traditional agricultural systems.Besides substrate and carbohydrate addition, stocking of tilapia was suggested to reduce underutilization of natural foods (plankton, periphyton and microbial floc) observed in monoculture ponds (Asaduzzaman et al., 2008).Tilapia in such system depends on the availability of natural foods in the form of plankton (Perschbacher and Lorio, 1993), periphyton (Azim et al., 2003a;Uddin, 2007) and microbial floc (Avnimelech, 2007).In addition, tilapia driven movements increases the bottom dissolved O 2 availability leading to better mineralization and stimulating the natural food web (Jimenez-Montealegre et al., 2002).
Tilapias and prawns have different food and feeding habits, but for both species, the addition of substrates resulted in extra growth and production (Tidwell et al., 2000, Uddin et al., 2006).This study monitored the effect of tilapia addition on prawn survival and production, pond ecology, and economic performance in presence and absence of substrates for periphyton development ponds.Thus considering above things in mind, the present study was undertaken to investigate the effects of addition of tilapia on the survival, growth and production of prawn in the presence and absence of substrates with periphyton in ponds.

Study area and pond facilities
The experiment was conducted at the Fisheries Field Complex of the Faculty of Fisheries, Bangladesh Agricultural University, Mymensingh, for a period of 120 days from 20 February to 20 June 2008 in six ponds having an area of 40 m 2 each.The ponds were rectangular in shape, well exposed to sun light, independent, and having water supply facilities.The water depth was maintained to a maximum of 01 m over the study period.The surrounding of all ponds was covered by 01 m height nylon net to prevent the entry of predators like birds, snakes, frogs and others.

Experimental design
The experiment was conducted under completely randomized design (CRD).Two treatments were tested each of which was replicated.The experimental design has been shown in Table 1

Pond preparation
All undesirable fish were completely eradicated by applying rotenone at a rate of 2.5 g/m 3 .Aquatic weeds were removed manually.The grasses of the pond dikes were also cut into small size by using scythe.
After one week of rotenone application, lime (CaO) was applied at a rate of 250 kg ha -1 .Three days after liming, ponds were fertilized with urea and triple super phosphate (TSP) each at a rate of 25 kg ha -1 and cow dung at a rate of 1,000 kg ha -1 .
The shelter was built by bamboo branch (locally known as kanchi) with date tree leaves and was installed in each pond before stocking with prawn juveniles to provide shelter for prawn.About 436 bamboo with a mean diameter of 0.05 m were posted vertically into the bottom mud of each pond, excluding a one meter wide perimeter water surface from the dike.This resulted in an additional area for periphyton development equaling about 60% (i.e.24 m 2 ) of the pond surface area.Ponds were not fertilized during the grow-out period.After the first fertilization and before prawn stocking, the ponds were left 10 days to allow plankton development in water column and periphyton growth on bamboo kanchi.

Stocking
Juveniles of Macrobrachium rosenbergii (5± 0.05g) procured from a nearby commercial hatchery were stocked at 3 juveniles m -2 in the ponds and juveniles of Oreochromis niloticus (24.3± 0.24g) from the Bangladesh Fisheries Research Institute (BFRI) were stocked according to the experimental design.

Feeding
Juveniles of freshwater prawn were fed with processed palleted feed containing 30.03% crude protein daily at a rate of 10% of the body weight for the 1 st month, 7% for 2 nd month and 3% for the rest of the culture period.Half of the required feed for a day was supplied in the morning and rest half in the evening.Feed requirement was calculated and adjusted after monthly sampling of prawn.Locally purchased tapioca starch was used as carbohydrate source for manipulating the C/N ratio.In order to raise the C/N ratio to 20 in all the ponds, 0.9kg tapioca starch was applied for each kg of formulated feed.The preweighed tapioca starch was mixed in a beaker with pond water and uniformly distributed over the pond surface directly after the feed application at 7.00 am.

Phyto and zoo-periphyton enumeration
Plankton samples were collected monthly from each pond.A bucket contained two litres of water was used to collect 10 litres of water from five different places and depth of the pond and passed through a fine mesh (25 µm) plankton net.The concentrated samples were transferred to a measuring cylinder and carefully made up to a standard volume of 50 ml with distilled water.Then the collected plankton samples were preserved in 10% buffered formalin in small plastic bottles each for subsequent studies.
From each of the 50 ml preserved sample, 1 ml sub-sample was examined using an S-R cell (Sedge Wick-Rafter cell S50, Microlitre) under a binocular microscope (Olympus, M-4000D, Japan) with phase contrast facilities.
One ml sub-sample from each sample was transferred to the cell and then all planktonic organisms present in 10 squares of the cell were identified and counted.Identification of plankton to the genus level was carried out using the Keys from Ward and Whipple (1959), Prescott (1962) and Belinger (1992).For each pond, mean number of plankton was recorded and expressed numerically per litre of water.Plankton density was calculated using the formula of Azim et al., 2001b.

N= (P X C X 100)/L
Where, N= The number of plankton cells or units per liter of original water P= The number of plankton counted in ten fields C= The volume of final concentrate of the sample L= The volume (liters) of pond water sample.

Harvesting
Adult freshwater prawn were harvested and counted for total number separately from each plot to evaluate the survival rate.After direct counting, weight and length of each individual prawn was also taken.

Analysis of growth data
Experimental data collected during the growth trial were used to determine the growth parameters as follows: Weight

Statistical analysis
For the statistical analysis of the data, a one-way ANOVA and DMRT were done by using the SPSS (Statistical Package for Social Science) version-11.5.Significance was assigned at the 0.05% level.Duncan's test was used to test the results of multiple ranges for comparisons of averages.

Water quality parameters
Physical parameters like transparency, temperature and chemical parameters such as pH, dissolved oxygen (DO), nitrate-nitrogen (NO 3 -N), phosphate-phosphorous (PO 4 -P), nitrite-nitrogen (NO 2 -N), total ammonia-nitrogen (TAN) and chlorophyll-a were measured throughout the study period.All parameters were more or less within the acceptable range for freshwater prawn culture.Water quality parameters in different treatments have been presented in Table 2. Temperature and pH of water were similar among the treatments.The addition of tilapia increased the bottom dissolved O 2 by 6.9% compared to the treatments without tilapia.The addition of tilapia increased the transparency and decreased the chlorophyll-a concentration of water.The chlorophyll-a concentration was always lower in tilapia added ponds (PT 0.5 ) compared to tilapia free ponds (PT 0 ) during the culture period.The mean values of TAN, NO 2 -N, NO 3 -N and PO 4 -P decreased by the addition of tilapia.All of the inorganic nitrogenous compounds (TAN, NO 2 -N, and NO 3 -N) decreased continuously during the culture periods in all treatments except for NO 3 -N in treatment without tilapia addition.The rate of decreasing of all inorganic nitrogenous compounds was higher and lower with and without addition of tilapia respectively.

Phyto-periphyton population
Mean abundance of phyto-periphyton along with their different groups are shown in Table 3. Phytoperiphyton population of the fish ponds was composed of four major groups: Bacillariophyceae (Diatom), Chlorophyceae (Green algae), Cyanophyceae (Blue green algae) and Euglenophyceae (Euglenophytes).Bacillariophyceae (Diatom): Bacillariophyceae comprised of 10 genera were observed.The mean abundance (×10 3 cells cm -2 ) of Bacillariophyceae was found to vary from 4.73 to 15.568 and 14.04 to 27.52 (×10 3 cells cm -2 ) with the mean values of 8.99±0.97 and 19.90±1.09(×103 cells cm -2 ) in case of treatments PT 0 and PT 0.5 , respectively.The abundance of treatment PT 0 was significantly higher (P< 0.05) than that of treatment PT 0.5 .Bacillariophyceae were dominated by Navicula, Nitzchia, Fragillaria, Cyclotella, Diatoma, Pinnularia, Surrirella and Synedra.Monthly variations in the abundance of Bacillariophyceae between the treatments are shown in Table 4.
Chlorophyceae: Chlorophyceae was found to be dominat over all groups of phyto-periphyton.Among the 21 genera observed Actinastrum, Ankistrodesmus, Chlorella, Closterium, Coelastrum, Oocystis, Pediastrum, Scenedesmus, Sphaerocystis and Ulothrix were the dominant ones.Mean abundance (×10 3 cells cm -2 ) of Chlorophyceae were 15.51±1.79 and 36.90 ±2.18 in treatments PT 0.5 and PT 0 respectively (Table 5).The mean abundance (×10 3 cells cm -2 ) of Chlorophyceae was found to vary from 8.62 to 33.92 and 19.04 to 49.90 in case of treatments PT 0.5 and PT 0 respectively.The variation in abundance of PT 0 was significantly higher (P< 0.05) than that of the treatment of PT 0.5 .Cyanophyceae: Cyanophyceae comprised of 7 genera and ranked second in respect of abundance.Among 7 genera, Anabaena, Microcystis, Gomphosphaeria and Oscillatoria were dominated.Average abundance (×10 3 cells cm -2 ) of Cyanophyceae was found to range from 5.98 to 17.51 and 10.29 to 22.80 with mean values of 9.75±0.68 and 15.75±0.86(×10 3 cells cm -2 ) in case of treatments PT 0.5 and PT 0 respectively.However, the variations in abundance of PT 0 were significantly higher (P< 0.05) than that of PT 0.5 .Monthly variations in the abundance of Cyanophyceae between the treatments are shown in Table 6.Euglena was the dominant.The mean abundance (×10 3 cells cm -2 ) of Euglenophyceae was 0.44±0.06and 0.60±0.10 in treatments PT 0.5 and PT 0 , respectively.The abundance (×10 3 cells cm -2 ) of Euglenophyceae was found to range from 0 to 0.83 in treatment PT 0.5 and 0 to 1.39 in treatment PT 0. .There was no significant difference (P>0.05) between treatment PT 0.5 and PT 0 when ANOVA was performed.Monthly variations in the abundance of Euglenophyceae among the treatments are shown in Table 7.Total phyto-periphyton: Among the phyto-periphyton group, Chlorophyceae was the most dominant group and Euglenophyceae was the least abundant group.In treatment PT 0.5 the average abundance (×10 3 cells cm -2 ) of phyto-periphyton was found to range from 23.21 to 66.99 with a mean value of 3.69±3.28,while in treatment PT 0 it varied from 45.04 to 97.99 with a mean value of 73.11±3.83.The variation in abundance of phyto-periphyton in treatment PT 0 was significantly higher (P< 0.05) than that of the treatment PT 0.5 .Monthly variations in the abundance of total phyto-periphyton among the treatments have been shown in Fig. 1. )

Zoo-periphyton Population
Two major groups i.e.Crustacea and Rotifera represented zoo-periphyton population of the experimental ponds.The mean abundance of zoo-periphyton in two treatments has been shown in Table 8.Rotifera: Rotifera was the most dominant zoo-periphytonic group, comprised of 5 genera namely Asplanchna, Brachionus, Filinia, Lecane and Trichocerca.The mean abundance (×10 3 cells cm -2 ) was found to range from 0.41 to 1.25 and 0.27 to 1.11 with the mean values of 0.72 ± 0.05 and 0.70 ± 0.06 in PT 0.5 and PT 0 respectively.No significant differences (P> 0.05) were recognized PT 0.5 and PT 0 when ANOVA was performed.Monthly variations in the abundance of Rotifers in the treatments are shown in Fig. 2.

Crustacea:
The abundance (x 10 2 cells cm -2 ) of Crustacea ranged from 0 to 2.78 and 0.139 to 5.56 under treatments PT 0.5 and PT 0 respectively.Crustacean group comprised of nauplius, was dominant.The mean abundance was not significantly different (P>0.05) between the treatments.Monthly fluctuation of Crustacean in different treatments is presented in Fig. 3. Total zoo-periphyton: Monthly variations of total zoo-periphyton in different treatments have been shown in Fig. 4. The mean zoo-periphyton density was 0.87±0.06and 0.95±0.06(× 10 2 cells cm -2 ) in treatments PT 0.5 and PT 0 respectively.Hence the values did not differ.Total zoo-periphyton abundance was found to range from 0.42 to 1.39 and 0.56 to 1.67 (× 10 2 cells cm -2 ) in treatments PT 0.5 and PT 0 respectively.

Growth and yield performance of freshwater prawn
The growth of freshwater prawn in different treatments was different.The different growth performance namely length (cm) and weight (g) gain, survival rate (%), percent weight gain and specific growth rate (% per day) and survival rate are shown in Table 9.

Water quality parameters
Water quality in lentic natural water bodies is strongly dependent on the autotrophic and heterotrophic organisms developed within the systems.In periphyton-based system, the close linkage between autotrophic and heterotrophic processes in periphyton mats speed up nutrient cycling and positively influences water quality (Milstein et al., 2003).The DO concentrations were generally suitable for prawn culture, although exceptionally low bottom DO values were recorded on a few occasions in tilapia free ponds.The addition of tilapia brings some oxygen to the bottom layers by their movements (Jimenez- Mean abundance (10 2 cells cm cm -2 ) Montealegre et al., 2002), thus increasing the bottom dissolved oxygen.Periphyton lowered the PO 4 -P of the overlying water which was also reported by the Hansson (1990) and Bratvold and Browdy (2001).Langis et al. (1988) and Ramesh et al. (1999) reported that the bacterial biofilm (periphyton) reduced toxic nitrogenous compounds through promotion of nitrification.In substrate-based ponds, nitrifying bacteria develop on the substrates which are located in the water column where more oxygen is available than at the sediment-water interface.In addition, periphytic algal community contributes to the processing of the nitrogenous wastes in ponds (Shilo and Rimon, 1982;Diab and Shilo, 1988).Thompson et al. (2002) reported that the attached diatoms and periphytic filamentous Cyanobacteria were responsible for the largest uptake of ammonium from the water in intensive shrimp culture ponds.The very low nitrogenous compounds in all treatments compared to other studies of freshwater prawn farming (Wahab et al., 2008;Kunda et al., 2008) and decreasing trends over time were due to the high C:N ratio of 20 and the addition of substrates for periphyton development.In the present study, tapioca starch was used as carbohydrate source for maintaining C:N ratio at 20. Increasing nutrient inputs caused lower NO 2 -N concentration in the water column and over the time, which can be attributed to low availability of TAN as substrate for nitrification (Avnimelech, 1999;Hari et al., 2004).Thus the lower level of nitrogenous compounds (NH 3 -N; NO 2 -N; NO 3 -N) over time could be attributed due to the addition of carbonaceous substrates that lead to increased microbial biomass, which immobilized TAN (Asaduzzaman et al., 2006;Asaduzzaman et al, 2008;Hari et al., 2004) and uptake of the nitrogenous compounds by periphyton.Addition of tilapia decreased the total nitrogen in the sediment possibly due to increased denitrification in response to fish driven oxygenation events (Torres-Berristain et al., 2006).In addition, another cause might be due to the re-suspension of pond bottom which release nutrients to the water column and tilapia harvested more phytoplankton in the water column keeping the algae more young stage which needed more nutrients as well.

Periphyton production
The major natural food types in ponds are phytoplankton, zooplankton, microbial floc, periphyton and benthic macroinvertebrate.The amounts of these natural food in ponds are influenced by management factors such as species combination, stocking density and ratio, and nutrient input quality and quantity (Milstein, 1993;Diana et al., 1997).
The periphyton community constitutes a major component of aquatic biological systems (Biggs, 1987).Perithyton includes both the phyto-perithyton and zoo-periphyton and sometime aquatic insects (Wetzel, 1983;Biggs, 1987).In the present study, phyto-periphyton and zoo-periphyton were only recorded as periphyton.The periphyton community was composed of Bacillariophyceae, Chlorophyceae, Cyanophyceae, Euglenophyceae, Crustacea and Rotifera.A total of 40 genera of periphyton were identified in the treatments PT 0.5 and PT 0 .The most dominant genera were Cyclotella, Navicula, Synedra, Chaetophora, Chlorella, Pediastrum, Scenedesmus, Oscillatoria, Ulothrix, Tabellaria, Gopmhosphaeria, and Microcystis colonizing the bamboo substrates in large numbers.Islam (1996), Haque (1996), Kawser (1998) and Ali (1998) observed similar pattern of findings on natural substrates in BAU campus ponds.Wahab et al. (1999) reported 53 genera of periphyton collected from scrap of bamboo in fishponds in Bangladesh among which 12 genera rarely occurred.Huchette et al. (2000) identified about 32 species of diatom as periphyton along with other microorganisms of both animal and plant kingdoms growing on artificial substrates in tilapia cages.
The mean abundance of periphyton in treatment PT 0 was very high, indicating that bamboo is a good substrate for periphyton growth.Eminson and Cattaneo et al. (1978) stated that the hard substrates such as bamboo poles are the most suitable substrate for periphyton growth.
The addition of tilapia decreased the phyto-periphyton and biomass per unit surface area, indicating the preference of tilapia towards periphyton as food.Tilapias are omnivores capable of feeding on benthic and attached algal and detrital aggregates (i.e.periphyton) (Dempster et al., 1993;Azim et al., 2003a).
There is also evidence that Nile tilapia grows better grazing on periphyton than filtering suspended algae from water column (Hem and Avit, 1994;Guiral et al., 1995;Huchette et al., 2000;Azim et al., 2003b).The abundance of periphytic zooplankton was similar in all treatments, indicating that the zooplankton communities were less preferable for the tilapias or escaping predation during grazing.

Growth and yield performance of prawn
The growth and net yield of freshwater prawn were significantly higher with no tilapia than with tilapia, indicating that inter-specific competition between tilapia and prawn decreased the net yield of prawn.
Although Uddin (2007) reported that tilapia addition might affect prawn survival during molting but the similar survival revealed that addition of substrates might have minimized the territoriality effect of tilapia on prawn.The FCR calculated based on prawn biomass increased significantly with the addition of tilapia because part of the feed was eaten by the tilapia whereas, substrates decreased FCR value by 13%.Uddin (2007) reported that FCR was 13% lower in fed-periphyton-based ponds compared to substrate free fed ponds.In case of tilapia, substrate addition increased the gross and net yield, indicating that substrates provide additional natural food for tilapia (Uddin, 2007).

Conclusion
The addition of periphyton substrates increased survival rate of prawn.Final weight and weight gain were significantly higher (P<0.05) in treatment PT 0 than those in treatment PT 0.5 .Specific growth rate was significantly higher in PT 0 than in PT 0.5 .The addition of tilapia decreased the gross and net yield of prawn.Tilapia had significant effects on FCR of freshwater prawn, substrates decreased FCR by 13.5% whereas addition of tilapia increased 16% gross yield.
Initial live prawn body weight (g) at time T 1 (day) W 2 = Final live prawn body weight (g) at time T 2 (day) T 2 -T 1 = Duration of the experiment (day).

Fig. 1 .
Fig. 1.Monthly variation of abundance of total phyto-periphyton in pond water under different treatments Mean abundance (10 2 cells cm -2

Fig. 2 .
Fig. 2. Monthly variation of abundance of periphytic rotifers in pond water under different treatments

Table 3 . Mean abundance ± SE (×10 3 cells cm -2 ) of phyto-periphyton of the ponds under two treatments each having three replicates. Values are means of 3 replicates and 5 sampling dates (N=15)
* Mean values with the different superscripts in rows are significantly different (P<0.05)

Table 6 . Effects of addition of tilapia on the abundance of periphyton (blue green algae) (biomass scraped from bamboo kanchi in different treatments)
Euglenophyceae: Euglenophyceae consisted of Euglena and Phacus species and between them