Usage of wastewater from RAS for microbial protein production.

The aim of this diploma thesis was to analyse waste water quality from RAS, verification of its possible cleaning in a model of wastewater treatment plant using heterotrophic bacteria, experimentally compare the influence of used carbon source in the initial phase of bacteria cultivation and evaluation of biofloc usage in fish feed. Analysed water was obtained from rearing tanks, sewage water from mechanical filtration and tap water for filling the system. Wastewater treatment plant (built according to AS VARIOcomp K model) was divided into 3 parts (primary settling, activated part, secondary settling) and then filled with water from running BFT system. For carbon source comparing experiment 4 groups were used: flour, glucose, acetate and glycerol, each in 3 replications. Cultivation took 28 days and at the end the nutritional composition of each BFT system was evaluated. Water analyses showed that sewage water from mechanical filtration is nutrient rich and contain big amount of undissolved solids. This water is not suitable for purification in conventional system without pre-treatment. The model of water treatment plant purified approximately 50 l of water from vortex per day which was pumped back into RAS. Probably it would be capable of purifying bigger volume. Bacteria cultivation using different carbon sources is very similar to start-up of nitrifying bacteria in biofilter at the beginning. It was proved that biofloc is also capable of removing nitrates from water. The biggest measured concentration among groups was 1695 +- 438 mg.l-1 NO3- and in 6 days it decreased to 493 +- 409 mg.l-1. The average FVI for flour, glucose, acetate and glycerol were 102 +- 57; 267 +- 59; 219 +- 26 and 293 +- 9 ml.l-1 at the end of cultivation. Experiment comparing carbon sources also proved that nutritional composition of biofloc is similar to composition of commercial feed. Heterotrophic bacteria are also capable to cumulate heavy metals and mineral matter over time. Experiment with added biofloc to feed in amount 0 %, 25 % and 50 % did not show any statistically significant differences (p < 0,05) on survival rate, FCR and SGR between the groups of juvenile grass carp (Ctenopharyngodon idella)