NGS-enriched activated sludge modelling of industrial wastewater treatment plant

Abstract

ABSTRACTMetagenomics advances with the Next Generation Sequencing (NGS) technology offer detailed insights into the microbial communities and their activities in a Wastewater Treatment Plant (WWTP). Since it has been shown recently that the microbial communities can be related to process data through machine learning, we investigate in this paper the enrichment of Activated Sludge Model 1 (ASM1) using time-series NGS data. We first present a modified ASM1 (mod-ASM1) to describe the industrial wastewater treatment at North Water’s WWTP facility in Delfzijl, the Netherlands. Subsequently, we identify the parameters for the ten weeks (weeks 40-50, 2014) of process data from North Water WWTP with prior parameters from the recommended ones from IWA. We further established a subset of parameters that are correlated to NGS data. Based on this relationship, a parameter-varying mod-ASM1 is obtained where the parameter variation is directly linked to the NGS data. We validate the NGS-enriched mod-ASM1 in the prediction of process data in the subsequent three weeks (weeks 50-53, 2014). While the enriched mod-ASM1 gives a good estimation of the COD effluent data, it cannot capture the production of nitrogen, which is often missed when the static model is deployed.