Evaluation of local ingredient formulated diet for golden rabbitfish, Siganus guttatus grow-out

. Golden rabbitfish, S. guttatus culture is feasible to be developed in ponds and floating net cages. To support rabbitfish culture, cheap and good quality feed using local ingredients is necessary. The study aimed to evaluate the local ingredients formulated feed for golden rabbitfish S. guttatus grow-out. Feed was formulated based on the least cost feed formulation. The feeding trial was carried out using 15 units of 1x1x2 m 3 floating net cages, with a fish density of 23 fish/cage with an initial weight of 39.88±8.05 g. The study was completely randomized, with four treatments of feed formulation (A, B, C, D) and commercial feed as control (E) in triplicates for each treatment. Parameters observed included growth response, feed utilization, digestibility, carcass composition, and feed cost per kg of fish production. The data obtained were analyzed statistically using analysis of variance (ANOVA). The result showed feed B gave the highest growth rate (1.05±0.09%) and was not significantly different from commercial feed (P>0.05). Still, it had a cheaper cost of production (IDR 15,000/kg of fish production) compared to commercial feed (IDR 22,000/kg of fish production) (P<0.05) with a 31.82% efficiency of production cost.


Introduction
Due to its high economic value, the golden rabbitfish (S. guttatus) has become an emerging species for aquaculture in Indonesia.The development of golden rabbitfish culture necessitates fulfilling feed requirements, as rabbitfish-specific commercial feed is not widely produced in Indonesia.Components and sustained availability of some ingredients are considered when determining feed ingredients [1].The herbivorous golden rabbitfish, S. guttatus, has the potential to be developed due to its relatively simple cultivation method, able to be cultivated at high densities, and can also feed on formulated feed [2][3][4].
Both ponds and floating net cage cultures are viable options for rabbitfish growth.However, one of the most common problems encountered in fish grow-out is the high feed cost, which can reach up to 60-75% of cultivation costs in intensive aquaculture activities [5], such as floating net cages and fish farming.In addition, there was no commercially available rabbitfish feed in Indonesia.Therefore, it is necessary to make an effort to acquire rabbitfish feed that is inexpensive and high quality.Utilizing local potential raw materials that are abundantly available but underutilized and enhancing their quality is one strategy for acquiring inexpensive feed.
The development of rabbitfish farming requires the development of low-cost feed derived from locally available raw materials at the cultivation center, such as fish meal, soybean meal, shrimp head meal, copra coke meal, corn meal, rice bran, seaweed, etc. Due to the various characteristics of cultivation areas, the availability of potential feed raw materials, and the influence of season, some feed formulations are required to obtain inexpensive rabbitfish feed while still promoting optimal growth.Several studies have explored local feed ingredients to produce affordable feed.[6] reported the potential local feed ingredients for ornamental fish, such as coconut copra waste, groundnut cake, maize, and soybean.[7] reported the utilization of locally fed ingredients had a significant effect on juvenile and fingerling catfish growth.
Therefore, it is also necessary to evaluate a variety of feed formulations based on regional raw materials that have the potential to yield low feed prices based on the characteristics of regional raw materials for rabbitfish growth.This study aims to evaluate the local ingredients formulated feed for rabbitfish S. guttatus grow-out.

Ingredients and Diets
Four feed formulations were developed from local ingredients that had a protein content of 28.9-31.2%,a lipid content of 6.3-8.6%, and a total energy of 4144-4636 Kcal/kg served as the tested feeds as treatments, along with one commercial feed (MG3) that had a nutrition composition similar to the other tested feeds (Table 1).The feed was formulated based on the least cost feed formulation for rabbitfish and the availability of main ingredients.Local fish meal, soybean meal, shrimp head meal, fermented copra coke meal, corn meal, rice bran, and Sargassum meal fermented by RICA 4+5, a probiotic product of RIBAFE (Research Institute for Brackishwater and Fishery Extension), as well as a vitamin-mineral mix with a range of prices for the South Sulawesi region, particularly Maros, were used as local ingredients (Table 2).  3Based on protein, lipid, fiber, and ash proximate calculation: NFE=100-(protein+lipid+fiber+ash) 4) Calculated from the diet's determined crude protein, crude lipid, and carbohydrate using gross energy conversion coefficients of 5.64, 9.44, and 4.11kcal/g, respectively (NRC, 1993). 5)Calculated based on rounding up the average price per kg of local ingredients in South Sulawesi Commercial feed (E) was used as a control with the following proximate composition (% dry matter): protein (29.2%), lipid (6.5%), fiber (6.5%), ash (13.0%),NFE (44.8%) and gross energy 4622 kcal/kg with a price per kg of IDR 8,600.Mineral mix 8,000-10,000 9,000

Experimental fish
Rabbitfish juveniles with 39.88±8.05g in average weight and 11.68±0.91cm in average length obtained from RIBAFE's hatchery at Barru, South Sulawesi-Indonesia, were used as experimental fish.The fish were reared in 15 units of 1x1x2 m 3 floating net cages with an initial density of 23 fish/cage.For five months, a feeding trial was conducted where feed was given in satiation three times daily in the morning, afternoon, and evening.To reduce feed waste, feeding was done slowly and carefully.Every month, rabbitfish growth is sampled concurrently with changing the cage nets.Additionally, a test of feed digestibility was performed.Using ten units of 200 L fiberglass tanks with a water flow system, aeration, and feces storage, the digestibility of the tested feed was observed.Fifteen juvenile rabbitfish weighing around 40 g in each tank were stocked.
As an indication, 0.6% chromium oxide (Cr2O3) was added to each tested feed.The juvenile rabbitfish were acclimated to the tested chromium feed for a week before the digestibility test.In satiation, feeding took place in the morning and evening.The remaining feed was removed by opening the water faucet at the bottom of the tank, and then the feces were collected every two to three hours.The feces were immediately placed in the freezer until they had enough amount to be dried and examined.

Observation of growth response and feed utilization parameters
Among the variables observed in this study was the specific growth rate (SR) of rabbitfish calculated with the following formula [6]: where Wt is the final weight of the fish, W0 is the initial weight of fish, and t is the rearing period.
Survival rate (SR) was determined using the following formula: Relative feed intake was determined using the following formula: )   100 where Wt is final weight of fish (g), and W0 is the initial weight of fish (g).
Feed efficiency (FE) was determined using formula: The protein efficiency ratio (PER) was determined using the formula: The feed conversion ratio (FCR) was determined using the formula: Where F is the amount of dry feed consumed (g), Wt is final weight of fish (g), and W0 is the initial weight of fish (g).Whereas apparent digestibility (AD) was determined using the following formula [8]: where AD is apparent digestibility (%), MD is chromium concentration of feed (%), MF is chromium concentration of feces (%), AD is the nutrient concentration of feed (%), and AF is the nutrient concentration of feces.Feed cost per fish production was calculated based on local ingredient feed price (Table 2) for 1 kg fish production.

Chemical and statistical analysis
Based on the AOAC International method [9], the proximate analysis of feed, fish, and feces was conducted.By drying the typical sample in an oven set at 105°C for 16 hours, dry matter (DM) was calculated.The sample was heated before being rinsed repeatedly with bases and acids to determine crude fiber.The sample was burned for 24 hours at 550°C in a muffle furnace to determine how much ash was present.Micro-Kjeldahl was used to assess crude protein, and petroleum benzene extraction was used to determine the sample's lipid content gravimetrically.
Nitrogen Free Extract (NFE) was determined by calculating the following results: NFE = (100 -crude protein, crude lipid content, ash content, and crude fiber content).Based on the protein conversion value of 5.64 Kcal/kg, fat of 9.44 Kcal/kg, and carbs of 4.11 Kcal/kg, the total energy of the diet was estimated [10].The chromium levels in the feed and feces samples were also examined using the [8] method.Dry fish feces weighed as much as 0.15 g, then added 5 mL of concentrated nitric acid (HNO3).The sample was heated at below 100°C temperature until the solution was white.The solution was cooled, and 3 mL of perchloric acid (HClO4) was then reheated until the green color turned orange.The sample was reheated for approximately 20 minutes.It was cooled at room temperature, and 50 mL of distilled water continued to be added to distilled water until the volume reached 100 mL.Evaporation was performed by letting the solution at room temperature and then reading the chromium concentration using a spectrophotometer at a wavelength of 400 nm.
At the end of the experiment, three fish of each treatment were composited to determine the carcass composition.The carcass composition, such as protein, lipid, fiber, and ash, was determined using the AOAC International method.
Using analysis of variance, growth response parameters, and feed utilization were statistically examined using analysis of variance (ANOVA).If there is a significant difference between the treatments, Tukey post hoc analysis was performed with a 95% confidence level.The data on the carcass composition of rabbitfish was analyzed descriptively.

Result and discussion
Table 3 shows the proximate feed composition made from local and commercial ingredients.Based on chemical analysis in the laboratory, the protein of the test feed was in the 30.17-33.02%range.Lipid content in the 6.37-7.98% in range.The fiber content of the test feed was in the 4.73-9.44range.Meanwhile, ash content was in the 11.55-20.61%range.The two feed formulations containing fermented Sargassum had high fiber and ash content due to the seaweed's high fiber and mineral content generally.NFE of all test feeds was in the 32.37-50.36range.To develop and sustain golden rabbitfish culture in Indonesia, the feed preferably produced locally utilizing locally accessible materials cheaply.The availability of an ingredient, its accessibility, and its nutrient content are of essential relevance [5].Four feed formulations with comparable protein and fat content to commercial feed were produced.Other proximate content varies relatively, such as NFE, crude fiber, and ash.The protein percentage of the four feed formulations made with local ingredients ranged from 30.95 to 33.02%, making them appropriate for rabbitfish.Rabbitfish is a herbivorous fish in coral reefs that consumes various aquatic plants.Herbivorous fish have a lower protein demand than carnivorous fish [11].Protein requirements of most marine fish are generally high, but species from warm seas, such as rabbitfish, do not require high protein [12].Rabbitfish eat a lot of algal protein in the wild and need high protein.A protein requirement of 31% for rabbitfish S. canaliculatus and protein requirement for it grow-out was 31% [13], [14].
Analysis of variance (ANOVA) indicated experimental feed did not influence the specific growth rate of golden rabbitfish (P>0.05),weight gain (P>0.05),relative feed consumption (P>0.05),feed efficiency (P>0.05), and survival rate of rabbitfish (P>0.05).In contrast, the experimental feed significantly affected the protein efficiency ratio (P0<05).Tukey's post hoc test revealed a significant difference between the protein efficiency ratio of commercial feed (E) and treatments A and C (P<0.05), while feed A and C were not significantly different (P>0.05) and treatments B, D, and E were not very different (P>0.05)(Table 4).The growth parameters of rabbitfish fed by four different feed formulations were not significantly different compared to commercial feed, as shown in Table 4.A considerable difference was detected in protein efficiency ratio (PER), with commercial feed having the greatest PER, which was not significantly different from formulated feed B and D. It demonstrated that protein in commercial feed, formula B and D, effectively utilized by rabbitfish.
The digestibility of the feed is one criterion for determining its quality.The experimental feed significantly affected the dry matter and protein digestibility coefficients (P<0.05) but had no significant impact on the lipid digestibility coefficient (P>0.05).According to Tukey's post hoc analysis, feed formulation of A, B, and E had higher and substantially different dry matter digestibility coefficients than feeds C and D (P<0.05).In contrast, feed C had a higher dry matter digestibility coefficient and significantly differed from feed D (P<0.05).The protein digestibility of feeds A, B, C, and E was more excellent and statistically different from the protein digestibility values of D (P<0.05) but not significantly different (P>0.05)(Table 5).The dry matter digestibility coefficient of feed A, B, and E were not significantly different.Dry matter digestibility is one of the measurements to indicate the feed quality.The greater the digestibility of dry matter, the greater the possibility for nutrients to be absorbed by fish for growth.The protein digestibility coefficients of feed A, B, C, and E are also nearly the same, indicating that rabbitfish can digest the protein in the feed properly.[15] reported that local ingredients such as freshwater shrimp, marine shrimp, and moringa leaves had acceptable digestibility values for tilapia feed formulation.Data on the nutritional value of some local ingredients can be used for good quality and low-cost feeding development [16].An essential factor in formulating fish diets is to ascertain their digestive capability.Understanding the accessibility of nutrients to the species facilitates the choice of suitable ingredients and the development of economically efficient diets [17].
The quality of fish meat is influenced by the diet they consume.Table 6 shows the approximate body carcass composition of fish raised for four months on the test meal.The results of the current study suggest that feed for rabbitfish manufactured from local ingredients can be developed.Locally sourced feed formulations support growth and feed utilization parameters and efficiencies comparable to commercial feed.
From an economic point of view, feed made from local raw materials is more economical.In contrast, feed made from local raw materials has a relatively lower price (price range IDR 5,350-6,850/kg) compared to commercial feed (IDR 8,600/kg).The results of calculating the cost of feed/kg of fish production are presented in Table 7.In general, the feed cost made from local raw materials for producing 1 kg of fish is lower than commercial feed.Note: Values are expressed as means±SD (n=3).Values with the same superscript at the same column mean not significantly different (P>0.05) in the Tukey post hoc test at a 95% confidence level .
*) The average feed cost was in rounded-up value Feed B had the cheapest cost production, meanwhile commercial feed is the most expensive one.The cost production of commercial feed was significantly different compared to feed B (P<0.05), but it was not significantly different compared to feed A and C (P>0.05).Feed expenses for producing 1 kg of fish was lower in the four feed formulations compared to commercial feeds.Based on these findings, it is concluded that local ingredient feed B provided the same growth, survival rate, feed consumption level, and feed efficiency as commercial feed but had a relatively lower cost production than commercial feed with 31.8% efficiency of fish production cost.Locally ingredient formula feed offers a cost-effective alternative to commercial feed.Moreover, farmers have the ability to produce this feed themselves, ensuring a consistent supply even in areas sometimes commercial feed may not be accessible [18].
The Ministry of Marine Affairs and Fisheries, Republic of Indonesia, funded this study through the national budget of RIBAFE 2021.We thank all RIBAFE technicians and analysts for their assistance during this study.We also thank Dr. Usman (in memoriam) for his leadership during the project.

Table 1 .
Formulation and proximate estimation (% dry matter) of tested feed

Table 2 .
The price of local feed ingredient in South Sulawesi

Table 3 .
Proximate analysis of several local ingredient-formulated feed and commercial feed (% dry matter)

Table 4 .
Growth performance and feed utilization on rabbitfish, S. guttatus grow-out

Table 5 .
Apparent digestibility of local ingredients and commercial feed in golden rabbitfish growoutNote: Values are expressed as means±SD (n=3).The value at the same row with the same superscript is not significantly different (P>0.05) in the Tukey post hoc test at a 95% confidence level.

Table 6 .
The composition (% dry matter) of golden rabbitfish carcass fed by different feed formulations Calculated from a composite of three samples of each treatment Treatment D produced fish carcass with the highest protein level, whereas treatment A produced the lowest protein level of fish carcass.Table6demonstrated that treatment E had a somewhat similar protein level to treatments B, C, and D. In comparison to treatment E, treatments A and B had the highest relative lipid content, whereas treatment C had the lowest.The ash and fiber contents varied slightly among the various treatments.

Table 7 .
Comparison of feed costs/kg of rabbitfish production in various feed formulations