Multivariate analysis and biodegradability test to evaluate different organic wastes for biological treatments: Anaerobic co-digestion and co-composting

Abstract

This study proposes the combination of statistical analysis and a biodegradability test to complement the composition of different wastes in order to find the optimal balance of nutrients for their joint bioconversion. Due to the need to determine the adequate balance of nutrients, the use of alternative techniques to experimental procedures could significantly reduce the cost and time of the process. With this aim, fifteen organic wastes (nine solid and six liquid wastes) were selected and different statistical analyses were performed on the physico-chemical characterization and respirometric variables. Liquid and solid wastes were analyzed separately using principal components analysis (PCA) (PC1 + PC2: 67% of total variance explained for solid substrates and PC1 + PC2: 85% of total variance explained for liquid substrates). The analysis provided considerable information about the predominant chemical composition of each substrate as well as their similarities and deficiencies to identify possible mixtures. In addition to PCA, cluster analyses (CA) were performed to group the substrates and identify the most significant differences between them. The joint evaluation of PCA and CA permitted identifying the optimal waste mixtures (i.e., glycerol-strawberry-fish waste) by correlating the loadings and scores plot, the cluster analysis dendograms and the COD/TKN ratio from the physico-chemical characterization. Moreover, multivariate regression was found to be an appropriate tool for predicting microbiological activity, as well as the soluble available biodegradable organic matter of each substrate. Inorganic carbon (CIC) and total organic carbon (CTOC) were found to be the most influential parameters in the prediction correlation of oxygen consumption and oxygen uptake rate.