Factors Influencing Prediction of Cryptosporidium Removal in Riverbank Filtration Systems: Focus on Filtration

摘要

Quantification of Cryptosporidium removal in the subsurface is critical for utilities consideringimplementation of in situ riverbank filtration systems and/or faced with quantifying contaminant removalsat sites where Ground Water Under Direct Influence (GWUDI) of surface water exists. Subsurfaceattenuation of pathogens occurs through filtration, dilution, and degradation processes. Detailedinvestigations of these processes in the field have not been reported in the literature. Field assessments ofCryptosporidium in riverbank filtrate are either unable to detect oocysts in filtrate or, when oocysts aredetected, report bulk removals without assessing the true filtration of their riverbank by accounting fordilution and degradation effects. Laboratory and column investigations have characterized the effects ofionic strength and pH on particle zeta potential and hydrophobicity, two important factors that influenceparticle depositions in porous media. These effects have not been adequately assessed in the field. Acomparison of contact efficiency between a rapid sand filter and a riverbank filter system in Kitchener,Ontario suggests that the riverbank filter may achieve the same or better particle contact with collectors.Ionic strength of the same riverbank well indicates favorable chemical conditions for particle attachmentto collectors. Although a theoretical riverbank contact efficacy was calculated for a range of particlesizes, predicting Cryptosporidium removal using conventional physico-chemical filtration models maynot be possible because of non-ideal deposition conditions in the subsurface; namely, surface charge andphysical heterogeneity of the collectors, time-dependent deposition processes, and uncertainty in particleattenuation resulting from organic sorption and physical straining. An examination of subsurfacefiltration conditions indicates that in situ experimentation is most appropriate for assessing riverbankfiltration efficacy for pathogen removal and that optimizing RBF design for low velocity may bebeneficial.