Remoção biológica de matéria orgânica e nitrogênio de água residuária da suinocultura em sistema A2/O

Abstract

Pig farming generates effluents rich in organic matter and nutrients, such as nitrogen (N), which must be treated, since, if incorrectly discharged into the environment, they become environmental liabilities. Therefore, it is essential to search for alternatives for removing N from swine wastewater, with the determination of optimal operational parameters for biological reactors such as the A2/O type. The objective of the present work was to evaluate the operational control parameters for the removal of organic matter and nitrogen from swine wastewater by biological process in a reactor with a three-stage Phoredox A2/O configuration. The study effluent came from the UFLA pig farming sector. The anaerobic reactors - Rana, anoxic - Rano had a useful volume height of 0.30 m and 18 L, the aerobic reactor - Raer with a useful height and volume of 0.40 m and 49 L and the secondary decanter - DS had a useful height and volume of 0.25 m and 4.0 L. The system supply (0.049 m3d-1) and the recirculation of the Raer to Rano (0.025 m3 d-1) was carried out by peristaltic pumps, which resulted in hydraulic detention time of 8.8 h (R1), 6.0 h (R2), 24.0 h (R3) and 2.0 h (DS). Different organic loads were applied in three phases of operation (F1 - 90 days, F2 - 110 days and F3 - 45 days). The experimental conditions were modified based on the COD present in the raw ARS. The average aeration rate provided in R3 was 3.55 mg L-1. The characterization of the influent and effluent of each reactor was carried out in weekly analyzes of pH, DO, COD, TS, VS, BA, TVA, NTK, nitrate and TP. The microbiological characterization of the sludge generated was carried out at the end of phases 1 and 2, using the next generation sequencing technique. The monitoring results showed a significant reduction in the system's COD, around 75% during the first phase of operation, and around 84% for nitrogen (NTK + N-NO3-), highlighting the efficiency of the A2O system. Finally, some genera of microorganisms, such as Clostridium sensu stricto 1, one of the abundant genera in the process, Caldisericum and Methanobacterium, which were present among the most abundant genera in the treatment system, presented themselves as some of the main genera during the process and whose families showed importance of action for the variation in pH and organic matter between phases 1 and 2, according to the machine learning method, Random Forest.