In September 2004, the project leader A/ Professor Peter Mather and Queensland University of Technology (QUT) research scientists Dr Alex Anderson and Dr Neil Richardson travelled to Nha Trang, Vietnam to meet with ACIAR partners from Research Institute for Aquaculture No.3 (RIA3) to initiate the project. Discussions were held with RIA3 Deputy Director Dr Thu and senior Research Scientists Mr Thach and Mr Le Vin regarding research priorities and methods. Agreements were reached on the experimental protocols required;
To establish protein and lipid requirements for juvenile mud crabs during grow out
To assess the digestibilities of local feed ingredients with the potential to replace fishmeal in formulated diets.

During this visit Drs Anderson and Richardson also travelled to a commercial mud crab hatchery in Hue to observe crablet production of the indigenous species in Vietnam, Scylla paramamosain.

Unfortunately, due to DFAT alerts about security threats in Indonesia it was not possible to visit the Indonesian ACIAR partners in 2004. Nevertheless, regular e-mail correspondence was maintained with Project Leader in Indonesia, Mr Ketut Suwirya at the Gondol Research Institute for Mariculture (GRIM), Bali and growth trials were initiated aimed at determining optimum protein and lipid requirements for juvenile mud crabs (S. paramamosian. In May 2005, A/ Prof. Peter Mather and Drs Anderson and Richardson travelled to GRIM to meet with the Indonesian ACIAR partners. These meetings confirmed that excellent progress was being made with the growth studies at GRIM and that the necessary infrastructure and technical expertise were available to facilitate the conduct of other nutritional studies outlined in the ACIAR project.

Progress to Date
Growth trials aimed at determining optimum protein and lipid levels in mud crab diets have been completed at all three ACIAR partner institutions (RIA3 GRIM and BIARC). Overall, the general trends to emerge from the trials were very similar. Specifically, optimum growth rates and feeding efficiencies were achieved with diets containing between 42% - 47% crude protein and 7 - 12% lipid. There were, however, significant differences in the maximum rates of growth exhibited by crabs at the different institutions. While the basis of these differences is unclear, we suggest that they may arise from variations in the quality of local ingredients used to formulate diets or differences in the average weight of crabs used at the commencement of the trials at each locality.

An unexpected finding to emerge from the mud crab growth trial conducted at Bribie Island Aquaculture Research Centre (BIARC) in Australia was that high rates of growth could be obtained with the indigenous mud crab species S. serrata using a commercial tiger prawn diet. This tiger prawn diet (Turbo; Thailand) promoted rates of growth equal to those achieved with a kuruma prawn diet (Ebistar, Higashimaru; Japan) previously used as the standard mud crab diet during grow out. Moreover, the Turbo tiger prawn diet contains 15% less crude protein than the Ebistar kuruma prawn diet and is 80% lower in price.

The project was conducted at Queensland University of Technology (QUT, Australia), Bribie Island Aquaculture Research Centre (BIARC, Qld, Australia), Gondol Research Institute for Mariculture (GRIM, Bali, Indonesia) and Research Institute for Aquaculture No. 3 (RIA3, Nha Trang, Vietnam).

Research conducted at BIARC utilised the local mud crab species, S. serrata and was conducted in three separate studies. In the first study, the optimum dietary protein and lipid requirements for juvenile crabs in intensive culture were investigated. The second study assessed the digestibility of high protein feed meals from a range of plant (soybean, cotton seed, canola, lupin), single cell (yeast) or terrestrial animal (meat, poultry) sources. A third study focused on the potential of poultry meal and soybean meal to replace fishmeal in diets formulated for juvenile mud crabs. Research components conducted at GRIM (Indonesia) and RIA3 (Vietnam) utilised the local mud crab species, S. paramamosian. Dietary protein and lipid requirements, apparent digestibilities of selected feed ingredients and potential of soybean and corn gluten as fishmeal replacements in diets for S. paramamosian were investigated at GRIM. At RIA3 research was conducted to determine the optimum dietary protein and lipid requirements, and the apparent digestibilities of selected feed meals from plant (soybean, corn, rice bran) and animal (krill) based ingredients with potential as fishmeal replacements in diets for mud crabs. In addition, pilot pond trials were conducted at GRIM and RIA3 to determine if diets optimised using intensive culture systems could be applied to semi intensive pond environments as typically used in commercial crab culture in South East Asia.

Initially, all experimental work was conducted utilising intensive culture systems where crabs were held in individual containers and fed experimental diets at fixed rates (4% - 10% body weight / day). At BIARC, research into the optimum dietary protein and lipid requirements for juvenile S. serrata, was performed using juvenile crabs (0.6g) and a range of experimental diets containing 25%, 35%, 45% or 55% crude protein (fishmeal the main protein source) with 5%, 10% or 15% lipid, over a ten week growth trial. Results indicated that maximum weight gain was achieved using diets containing 55% crude protein and 15% lipid. Importantly, average weight gain demonstrated by crabs fed this diet over the culture period was not significantly different to that achieved using a commercial diet (Turbo, Thailand) for tiger prawns (P.monodon).

Subsequently, the digestibility of high protein feed meals was assessed using sub adult mud crabs (approx. 90g) which were fed diets based on the Turbo feed and also contained 30% of the test ingredient. All meals tested, except meat meal, demonstrated apparent dry matter digestibility (ADMD) values above 78% and had digestibility coefficients for crude protein and gross energy that were not significantly less than fishmeal. In addition, it was demonstrated that up to 40% of fishmeal in experimental diets could be replaced by either poultry meal or soybean meal without significantly reducing weight gain over a ten week culture period.

Findings from experiments conducted at GRIM indicated best growth was achieved using diets containing 45% crude protein with 9% or 12% dietary lipid. A digestibility study demonstrated that animal-based feeds (shrimp head, tiny shrimp and squid liver meal) had higher AMDM values than fishmeal. Of the plant-based feedstuffs, soybean meal had highest AMDM value and corn gluten lowest. Soybean meal also had significantly higher protein and energy digestibilities than corn gluten. Subsequently, it was demonstrated that 20% or 40% of fishmeal could be replaced by soybean meal or corn gluten, respectively, without significantly reducing weight gain over the culture period.

Results from experiments conducted at RIA3 showed that after four months of culture, the highest final weights were obtained with crabs fed a diet containing 43% crude protein and 15% lipid. A digestibility study indicated that the ADMD and ACPD values for krill meal and soybean meal were not significantly less than those for fishmeal. As a consequence, the two ingredients were tested for their capacity to replace fishmeal in formulated diets for juvenile mud crabs. Specifically, a reference diet based on fishmeal was formulated which contained 43% crude protein and 15% lipid. Four experimental diets based on the reference formulation were also formulated where fishmeal was progressively replaced by soybean meal or krill meal. Juvenile crabs were then fed the different diets at a feeding rate of 4% body weight per day. After four months of culture it was shown that up to 25% of fishmeal in formulated diets could be replaced by either krill meal or soybean meal without significantly reducing growth performance.

Recently, pilot pond trials have been conducted at GRIM and RIA3 to determine if diets developed using the intensive culture systems can be applied to semi intensive pond culture systems typically used in commercial crab culture in South East Asia. At RIA3, juvenile mud crabs (0.9g) were stocked in earth ponds at a density of one crab per m2 and fed formulated diets containing 43% crude protein with 10% or 15% lipid for three months. Feeding rate was set at 4% body weight per day and trash fish was used as a reference diet. Preliminary findings of these studies have shown that growth equivalent to that achieved using trash fish can be obtained with the formulated diets. Significantly, however, the final weights of crabs grown in pond environments were much higher (8 to 10 times) than those grown for an equivalent period in the intensive culture systems. Early indications from pond trials conducted at GRIM have also indicated significantly higher growth rates are possible in pond environments than in intensive culture systems. We suggest that additional natural nutrition available in pond environments promotes significantly higher weight gain than can be attained using the current intensive culture protocols. Clearly, further work is required to optimise the growth performance of mud crabs in intensive culture systems, particularly where formulated diets are the only source of nutrition. We strongly suggest that additional studies, such as those which seek to investigate different daily feeding rates (ie. 20%, 10% or 5% BW/day), should be undertaken to establish if growth performance equivalent to that demonstrated in pond environments can be attained in intensive culture.

Summary
The major findings to emerge from this study were;
In intensive culture, maximum growth of S. serrata was achieved using a formulated diet containing 55% crude protein and 15% lipid. Growth performance obtained using this diet was equivalent to that obtained using a commercial shrimp diet.
In intensive culture, maximum growth of S. paramamosian was achieved using formulated diets containing 43% - 45% crude protein and 9% -15% lipid.
Digestibility coefficients were obtained for a variety of plant and animal-based feed meals which were equivalent or superior to those obtained for fishmeal.
In some instances, up to 40% of fishmeal in formulated diets could be replaced by high protein animal or plant-based meals without significant reduction in growth performance.
Substantially higher levels of growth were achieved in pond environments than in intensive culture systems. We recommend that further research be conducted to optimise feeding protocols in intensive culture systems.

The current project represents an extension to ACIAR project FIS/2000/065 which aimed to identify the potential of terrestrial animal and plant-based feed meals for use in diets formulated for mud crabs cultured in Australia, Vietnam and Indonesia. The project was initiated in response to recognition that a major constraint to mud crab aquaculture is the lack of diets specifically formulated for cultured crabs. Although artificial diets have been used for crab aquaculture, such diets have primarily been designed for prawn species and contain high levels of relatively expensive animal-based feed ingredients, such as fishmeal. A key aim of the original FIS/2000/065 project was therefore to identify low cost animal and plant-based feed ingredients with potential to replace marine animal-based protein in diets used for mud crab production in Indonesia, Vietnam and Australia.

The original project was conducted at Queensland University of Technology (QUT, Australia), Bribie Island Aquaculture Research Centre (BIARC, Qld, Australia), Gondol Research Institute for Mariculture (GRIM, Bali, Indonesia) and Research Institute for Aquaculture No. 3 (RIA3, Nha Trang, Vietnam). Research conducted at BIARC utilised the local mud crab species, S. serrata and was conducted in three separate studies. In the first study, the optimum dietary protein and lipid requirements for juvenile crabs in intensive culture were investigated. The second study assessed the digestibility of high protein feed meals from a range of plant, single cell or terrestrial animal sources. A third study focused on the potential of poultry meal and soybean meal to replace fishmeal in diets formulated for juvenile mud crabs. Research components conducted at GRIM (Indonesia) and RIA3 (Vietnam) replicated the studies conducted in Australia but utilised locally available feed ingredients and selected the local mud crab species, S. paramamosian for culture.

In June 2006, the major findings of the original FIS/2000/065 project were presented at a two day workshop at the the Nha Trang Lodge in Nha Trang, Vietnam. The key findings presented at the workshop were;

1. In intensive culture, maximum growth of S. serrata was achieved using a formulated diet containing 55% crude protein and 15% lipid. Growth performance obtained using this diet was equivalent to that obtained using a commercial shrimp diet.

2. In intensive culture, maximum growth of S. paramamosian was achieved using formulated diets containing 43% - 45% crude protein and 9% -15% lipid.

3. A broad range of plant and animal-based feed meals were identified that were readily digested by mud crabs.

4. In some instances, up to 40% of fishmeal in formulated diets could be replaced by high protein animal or plant-based meals without significantly reducing growth performance.

During an ACIAR commissioned review of this project, (Heasman, 2006) it was stated that;

"Results in the main were encouraging and supported earlier findings that mud crabs can efficiently digest a wide range of relatively cheap and abundant plant, animal and single cell based ingredients."

Significant problems matching experimental systems to "real world" conditions, however, were encountered. In particular, the results of growth trials using individually reared crablets were confounded by relatively low growth rates (around 50% of those achieved in pond culture), raising concerns that some outcomes might differ when crabs are raised in ponds. In particular, it was observed that access to "ad libitum" feeding in preliminary pond trials in Vietnam promoted higher growth rates than were achieved in experimental systems. Since growth rates in the laboratory may be influenced by feeding practices, additional studies were recommended to optimise factors such as optimal feeding rates and frequencies.

The current extension to project FIS/2000/065 represents an attempt to address the concerns raised by the review of the original two year project and was initiated in late 2006 when ACIAR invited the Australian project partners (QUT and BIARC) to apply for a twelve month extension to the project. The project aims to investigate why growth rates previously recorded in intensive cellular culture are significantly lower than those seen in pond systems. This aim will be addressed through the use of an improved culture system at BIARC with an automated feeding mechanism which permits a much broader range of feeding protocols to be tested than was possible in the original project. Once feeding protocols have been optimised, the next major aim will be to determine appropriate inclusion levels of protein and lipid in formulated diets

Funding for the extension to project FIS/2000/065 was released to the Australian project partners in May-June 2007 and growth trials associated with the first study have recently commenced at BIARC.