Animal versus plant protein sources in marine ingredient-free aquaponic diets: A case study on nutrient release, and retention of African catfish (Clarias gariepinus) reared in RAS

Aquaponic systems have potential to improve water and nutrient use efficiency compared to separately operated recirculating aquaculture systems (RAS) and hydroponic facilities. However, fish feeds are optimized for fish performance and low effluent nutrient discharge instead of integrated production of fish and plants, often leading to deficiencies of certain dissolved plant nutrients in RAS water. This study tested the effect of gradually replacing a blend of animal by-products (poultry by-product meal, catfish by-product meal, poultry blood meal) in a marine ingredient-free and entirely animal protein-based diet with a blend of common plant protein sources in four quasi-realistic isonitrogenous (42% crude protein) and isocaloric (20 MJ/kg) diets for African catfish on growth performance, body composition and nutrient release in dissolved and solid form into RAS water. Reducing the contribution of animal proteins to dietary protein from 100% over 75% and 50% to 25% resulted in the best biomass gain (%), feed conversion ratio (dry matter basis) and protein efficiency ratio for the 50% level, with differences, however, non-significant compared to the 100% and 75% level. At the 25% level, i.e. the highest level of plant protein inclusion, significant impairments regarding these growth indices as well as body ash, P and Ca were recorded compared to all other diets. No differences were observed for total dissolved 4 release per unit of feed between diets. Higher animal protein contribution lead to significantly higher release of dissolved P and Ca into RAS water, whereas increasing plant protein contribution caused significantly higher release of dissolved K, Mg and B. A mass balance of dietary nutrient fate revealed that N, P, K, S and Mg primarily ended up in fish biomass and in dissolved form in RAS water rather than in collected solids, while Ca, Fe, Mn, Zn and Cu predominantly ended up in solids and fish biomass rather than in dissolved form in RAS water. Modeling daily RAS water exchange rates needed to achieve the dissolved N concentration of a standard hydroponic nutrient solution on the basis of dissolved N release per unit of feed revealed that the entirely animal protein-based diet could reach 100%, 86% and 5% of the required dissolved N, P and K concentrations and that the diet with the highest plant protein inclusion could reach 100%, 14% and 15%, both at an exchange rate of 118 L/kg of feed/day.