Linking Bacterial Populations and Nutrient Removal in the Granular Sludge Ecosystem

DynamicsDynamics in microbial selectionMicrobial selection were investigated in aerobic granular sludgeAerobic granular sludge over five months under steady-state operationsOperation of two anaerobicAnaerobic-aerobicAerobic sequencing batch reactorsSequencing batch reactor at 20 (SBRSBR-20) and 25 °C (SBRSBR-25). Numerical ecologyNumerical ecology was used to link clustersCluster of bacterial populationsPopulation with fluctuations in operationOperation variables and biological nutrient removalBiological nutrient removal (BNRBNR). OperationOperation by controlling the sludgeSludge age (20 days) and biomassBiomass concentrationConcentration (12 gVSS L−1) with purgePurge of excess granular sludgeGranular sludge not only resulted in the effective removalRemoval of phosphorusPhosphorus but also in the stabilization of the communityCommunity compositionComposition. An efficient simultaneous BNRBNR (> 98% CODCOD, > 95% P, > 70% N) was obtained with influentInfluent concentrationsConcentration of acetateAcetate > 500 mgCOD L−1 and nutrientsNutrient ratiosRatio > 30 gCOD gP−1 and > 10 gCOD gN−1. AmmoniumAmmonium was fully nitrified in SBRSBR-20 with 60% O2 sat. In SBRSBR-25, nitrificationNitrification deteriorated already at 80% O2 sat: the bigger granulesGranule and volume of granular sludgeGranular sludge in this reactorReactor limited the mass transferMass transfer of oxygen from the airAir to the mixed liquorMixed liquor and to the biomassBiomass particlesParticle. In addition to nutrientNutrient level controlControl, oxygenationOxygenation strategies of granular sludgeGranular sludge reactorsReactor should be adapted to the physical properties of the biomassBiomass. Multivariate numerical analysesMultivariate numerical analyses highlighted different behaviors of both SBRsSBR. Significant correlationsCorrelation between the fine-scale population dynamicsPopulation dynamics, operationOperation variables, and BNRBNR were only obtained for SBRSBR-20. The microbiomeMicrobiome was composed of two main opponent clustersCluster: (i) “Ca. AccumulibacterAccumulibacter”, NitrospiraNitrospira, XanthomonadaceaeXanthomonadaceae, and AminobacterAminobacter were selected under efficient BNRBNR and (ii) “Ca. CompetibacterCompetibacter”, SphingobacterialesSphingobacteriales, CytophagaCytophaga, and IntrasporangiaceaeIntrasporangiaceae under BNRBNR deteriorationDeterioration. Other flanking populationsFlanking population frequently detected in activated sludgeActivated sludge were present as well. An ecosystem modelEcosystem model of the BNR granular sludgeBNR granular sludge microbiomeGranular sludge microbiome was built to guide functional analysesFunctional analyses.