Sand box experiments with bioclogging of porous media: Hydraulic conductivity reductions

Tracer experiments during clogging and de-clogging experiments in a 2D sand box were via an image analysis used to establish a data set on the relation between changes in hydraulic conductivity (K) and relative porosity (β). Clogging appears to create a finger-like tracer transport, which could be c...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of contaminant hydrology 2012-08, Vol.136-137, p.1-9
Hauptverfasser: Seifert, Dorte, Engesgaard, Peter
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Tracer experiments during clogging and de-clogging experiments in a 2D sand box were via an image analysis used to establish a data set on the relation between changes in hydraulic conductivity (K) and relative porosity (β). Clogging appears to create a finger-like tracer transport, which could be caused by an initial heterogeneous distribution of biomass in the sand box. De-clogging occurs at a slower rate possibly due to the presence of inert biomass that is not affected by the starvation conditions by sudden removal of the substrate source. The tracer front was observed to get disturbed closer and closer to the substrate source during the experiments suggesting that the zone of clogging moved upstream. Three clogging models, K(β), from the literature were tested for their ability to describe the temporal changes in clogging at the scale of the sand box; the model of Clement et al. (1996) that makes no assumption on biomass distribution, the plug formation model of Thullner et al. (2002a), and the biofilm-plug formation model of Vandevivere (1995). The plug formation and biofilm-plug formation models both match the observed changes between the hydraulic conductivity of the sand box and the relative porosity. Unfortunately our experiments did not reach low relative porosities where the two models predict different behaviors. The model by Clement et al. (1996) underestimates clogging. ► 2D Sand box experiments with flow and bioclogging. ► Non-destructive image analysis of bioclogging. ► Bioclogging creates preferential flow paths. ► De-clogging is slower than clogging due to the presence of inert biomass. ► Test of three bioclogging models.
ISSN:0169-7722
1873-6009
DOI:10.1016/j.jconhyd.2012.04.007