Process dominance analysis for fate modeling of flubendazole and fenbendazole in liquid manure and manured soil

Abstract

Fate monitoring data on anaerobic transformation of the benzimidazole anthelmintics flubendazole (FLU) and fenbendazole (FEN) in liquid pig manure and aerobic transformation and sorption in soil and manured soil under laboratory conditions were used for corresponding fate modeling. Processes considered were reversible and irreversible sequestration, mineralization, and metabolization, from which a set of up to 50 different models, both nested and concurrent, was assembled. Five selection criteria served for model selection after parameter fitting: the coefficient of determination, modeling efficiency, a likelihood ratio test, an information criterion, and a determinability measure. From the set of models selected, processes were classified as essential or sufficient. This strategy to identify process dominance was corroborated through application to data from analogous experiments for sulfadiazine and a comparison with established fate models for this substance. For both, FLU and FEN, model selection performance was fine, including indication of weak data support where observed. For FLU reversible and irreversible sequestration in a nonextractable fraction was determined. In particular, both the extractable and the nonextractable fraction were equally sufficient sources for irreversible sequestration. For FEN generally reversible formation of the extractable sulfoxide metabolite and reversible sequestration of both the parent and the metabolite were dominant. Similar to FLU, irreversible sequestration in the nonextractable fraction was determined for which both the extractable or the nonextractable fraction were equally sufficient sources. Formation of the sulfone metabolite was determined as irreversible, originating from the first metabolite.