This website has changed. We hope you can find what you need easily, but items have moved around. If you have trouble finding what you are looking for please let us know.

Contact us

Filtration and transport of Bacillus subtilis


Filtration of Bacillus subtilis spores and the F-RNA phage MS2 (MS2) on a field scale in a coarse alluvial gravel aquifer was evaluated from the authors' previously published data. An advection-dispersion model that is coupled with first-order attachment kinetics was used in this study to interpret microbial concentration vs. time breakthrough curves (BTC) at sampling wells. Based on attachment rates (katt) that were determined by applying the model to the breakthrough data, filter factors (f) were calculated and compared with f values estimated from the slopes of log (cmax/co) vs. distance plots. These two independent approaches resulted in nearly identical filter factors, suggesting that both approaches are useful in determining reductions in microbial concentrations over transport distance. Applying the graphic approach to analyse spatial data, we have also estimated the f values for different aquifers using information provided by some other published field studies. The results show that values of f, in units of log (cmax/co) m(-1), are consistently in the order of 10(-2) for clean coarse gravel aquifers, 10(-3) for contaminated coarse gravel aquifers, and generally 10(-1) for sandy fine gravel aquifers and river and coastal sand aquifers. For each aquifer category, the f values for bacteriophages and bacteria are in the same order-of-magnitude. The f values estimated in this study indicate that for every one-log reduction in microbial concentration in groundwater, it requires a few tens of meters of travel in clean coarse gravel aquifers, but a few hundreds of meters in contaminated coarse gravel aquifers. In contrast, a one-log reduction generally only requires a few meters of travel in sandy fine gravel aquifers and sand aquifers. Considering the highest concentration in human effluent is in the order of 10(4) pfu/l for enteroviruses and 10(6) cfu/100 ml for faecal coliform bacteria, a 7-log reduction in microbial concentration would comply with the drinking water standards for the downgradient wells under natural gradient conditions. Based on the results of this study, a 7-log reduction would require 125-280 m travel in clean coarse gravel aquifers, 1.7-3.9 km travel in contaminated coarse gravel aquifers, 33-61 m travel in clean sandy fine gravel aquifers, 33-129 m travel in contaminated sandy fine gravel aquifers, and 37-44 m travel in contaminated river and coastal sand aquifers. These recommended setback distances are for a worst-case scenario, assuming direct discharge of raw effluent into the saturated zone of an aquifer. Filtration theory was applied to calculate collision efficiency (alpha) from model-derived attachment rates (katt), and the results are compared with those reported in the literature. The calculated alpha values vary by two orders-of-magnitude, depending on whether collision efficiency is estimated from the effective particle size (d10) or the mean particle size (d50). Collision efficiency values for MS-2 are similar to those previously reported in the literature (e.g. ) [DeBorde, D.C., Woessner, W.W., Kiley, QT., Ball, P., 1999. Rapid transport of viruses in a floodplain aquifer. Water Res. 33 (10), 2229-2238]. However, the collision efficiency values calculated for Bacillus subtilis spores were unrealistic, suggesting that filtration theory is not appropriate for theoretically estimating filtration capacity for poorly sorted coarse gravel aquifer media. This is not surprising, as filtration theory was developed for uniform sand filters and does not consider particle size distribution. Thus, we do not recommend the use of filtration theory to estimate the filter factor or setback distances. Either of the methods applied in this work (BTC or concentration vs. distance analyses), which takes into account aquifer heterogeneities and site-specific conditions, appear to be most useful in determining filter factors and setback distances.

view journal