Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Disturbance of the bacterial communities in drinking water produced from established and newly-built slow sand filters

Paul, Catherine J. LU orcid ; Chan, Sandy LU ; Rådström, Peter LU ; Pullerits, Kristjan LU and Persson, Kenneth M LU (2016) MEWE 2016
Abstract
Introduction
Slow sand filtration is an established method for drinking water treatment, however, the knowledge about their microbial ecology remains limited. Maintenance of slow sand filters (SSF) includes washing to remove biofilm growth in the sand and restore flow through the filter. Understanding how this impacts produced water is important as the SSF may influence the bacterial flora in the finished drinking water and distribution system biofilms (El-Chakhtoura et al. 2015, Lührig et al. 2016, manuscript in preparation).
Routine SSF monitoring after washing includes heterotrophic plate count (HPC) however this only describes a small fraction of the bacteria in the water (Allen, Edberg et al. 2004). Flow cytometry (FC) has... (More)
Introduction
Slow sand filtration is an established method for drinking water treatment, however, the knowledge about their microbial ecology remains limited. Maintenance of slow sand filters (SSF) includes washing to remove biofilm growth in the sand and restore flow through the filter. Understanding how this impacts produced water is important as the SSF may influence the bacterial flora in the finished drinking water and distribution system biofilms (El-Chakhtoura et al. 2015, Lührig et al. 2016, manuscript in preparation).
Routine SSF monitoring after washing includes heterotrophic plate count (HPC) however this only describes a small fraction of the bacteria in the water (Allen, Edberg et al. 2004). Flow cytometry (FC) has been proposed for routine monitoring of bacteria in drinking water. FC counts the cells present in water and generates a fingerprint based on describing differential DNA staining to describe the population of bacteria (Arnoldini et al. 2013).
Two full-scale new SSFs were constructed using different types of sand as starting material. Traditional microbial parameters (HPC, coliforms and E. coli) and FC data suggested these SSFs differed from each other and were not performing as well as established SSFs (Chan et al. 2016, manuscript in preparation). This study explored: using FC for routine monitoring of SSFs during washing; and, if the response to washing differed between each of the newly built SSFs when compared to an established SSF.
Materials and Methods
Influent and effluent water were characterized using HPC, the Colilert method (E.coli and coliforms), and FC according to established protocols (Prest et al. 2013). FC data was analyzed with Cytometric Histogram Image Comparison (CHIC) (Koch et al. 2013).
Results and Conclusions
Two new SSFs were built using only new sand (SSF-new), or a mix of new sand with sand previously removed during washing from existing sand filters (SSF-mix). After 6 months, the impact on water quality of disturbance from SSF washing was determined by comparing the bacterial populations of the influent and effluent water from SSF-new, SSF-mix, and an established SSF (SSF-est). No significant log reduction of cells was observed in any effluent water. The fluorescence distribution of effluent differed from that of the influent with each SSF showing a unique fingerprint (Fig 1.1).

Figure 1.1 Comparison of fluorescence fingerprints for (left to right) SSF-est, SSF-mix, and SSF-new. Profiles show the combined plot of cell concentration and fluorescence distribution of nucleic acid stained with SYBR Green I in effluent water before wash (blue), and 4 days after washing (red).
Fingerprints of effluent water were stable throughout the sampling period for SSF-est, regardless of washing, while for SSF-new and SSF-mix, fingerprints suggested a destabilization of the bacterial population related to washing. This suggested that SSF-new, SSF-mix and SSF-est differed in the population of cells being seeded into their respective effluent waters and this was supported by results from traditional methods and qPCR, showing lowest number of coliforms in the effluent from SSF-est.
Analysis by CHIC, showed that the cell populations seeded by SSF-mix were more similar to those from SSF-est, than SSF-new. FC fingerprints from SSF-new were very similar to those of the influent water, suggesting that the biofilm in this SSF had little impact on the water passing through SSF-new. Since SSF-mix included washed sand from established SSFs, this appears to have inoculated the SSF-mix biofilm in a way that was not observed for SSF-new. The fingerprints for SSF-est effluent remained stable, consistent with a biofilm that is resilient to disturbance, and coliform counts demonstrating uninterrupted SSF function.
This study thus suggests that washed sand is preferable to new sand for SSF construction and that more than 6 months is required for a resilient SSF biofilm. While total cell numbers did not change significantly across any SSF, SSF-est selectively reduced HPC and coliforms suggesting that this biofilm function is selective compared to newer SSF biofilms. In this study FC was not only useful for monitoring total cell counts to follow changes during SSF washing, but showed a consistent effluent water fingerprint associated with a resilient and functional SSF biofilm. As this was not observed for the newly built SSFs, FC can follow both the character and establishment of SSF biofilms and should be helpful for analyzing the impacts of any potential events on SSF biofilm integrity.
References
Allen, M.J., Edberg, S.C. and Reasoner, D.J., (2004). Heterotrophic plate count bacteria—what is their significance in drinking water? International journal of food microbiology, 92(3), pp.265-274.
Arnoldini, M., Heck, T., Blanco-Fernández, A. and Hammes, F., (2013). Monitoring of dynamic microbiological processes using real-time flow cytometry. PloS one, 8(11), p.e80117.
El-Chakhtoura, J., Prest, E., Saikaly, P., van Loosdrecht, M., Hammes, F. and Vrouwenvelder, H., 2015. Dynamics of bacterial communities before and after distribution in a full-scale drinking water network. Water Research, 74, pp.180-190.
Koch, C., Fetzer, I., Harms, H. and Müller, S., (2013). CHIC—an automated approach for the detection of dynamic variations in complex microbial communities. Cytometry Part A, 83(6), pp.561-567.
Prest, E.I., Hammes, F., Kötzsch, S., Van Loosdrecht, M.C.M. and Vrouwenvelder, J.S., (2013). Monitoring microbiological changes in drinking water systems using a fast and reproducible flow cytometric method. Water research, 47(19), pp.7131-7142.
(Less)
Please use this url to cite or link to this publication:
author
; ; ; and
organization
publishing date
type
Contribution to conference
publication status
published
subject
conference name
MEWE 2016
conference location
Copenhagen, Denmark
conference dates
2016-09-04 - 2016-09-07
language
English
LU publication?
yes
id
812fd1f2-4fc1-4d1c-82d0-4459e173a4d7
date added to LUP
2019-06-19 16:22:09
date last changed
2019-07-23 11:45:16
@misc{812fd1f2-4fc1-4d1c-82d0-4459e173a4d7,
  abstract     = {{Introduction<br/>Slow sand filtration is an established method for drinking water treatment, however, the knowledge about their microbial ecology remains limited. Maintenance of slow sand filters (SSF) includes washing to remove biofilm growth in the sand and restore flow through the filter. Understanding how this impacts produced water is important as the SSF may influence the bacterial flora in the finished drinking water and distribution system biofilms (El-Chakhtoura et al. 2015, Lührig et al. 2016, manuscript in preparation).   <br/>Routine SSF monitoring after washing includes heterotrophic plate count (HPC) however this only describes a small fraction of the bacteria in the water (Allen, Edberg et al. 2004). Flow cytometry (FC) has been proposed for routine monitoring of bacteria in drinking water. FC counts the cells present in water and generates a fingerprint based on describing differential DNA staining to describe the population of bacteria (Arnoldini et al. 2013). <br/>Two full-scale new SSFs were constructed using different types of sand as starting material. Traditional microbial parameters (HPC, coliforms and E. coli) and FC data suggested these SSFs differed from each other and were not performing as well as established SSFs (Chan et al. 2016, manuscript in preparation). This study explored: using FC for routine monitoring of SSFs during washing; and, if the response to washing differed between each of the newly built SSFs when compared to an established SSF. <br/>Materials and Methods<br/>Influent and effluent water were characterized using HPC, the Colilert method (E.coli and coliforms), and FC according to established protocols (Prest et al. 2013). FC data was analyzed with Cytometric Histogram Image Comparison (CHIC) (Koch et al. 2013). <br/>Results and Conclusions<br/>Two new SSFs were built using only new sand (SSF-new), or a mix of new sand with sand previously removed during washing from existing sand filters (SSF-mix). After 6 months, the impact on water quality of disturbance from SSF washing was determined by comparing the bacterial populations of the influent and effluent water from SSF-new, SSF-mix, and an established SSF (SSF-est). No significant log reduction of cells was observed in any effluent water. The fluorescence distribution of effluent differed from that of the influent with each SSF showing a unique fingerprint (Fig 1.1). <br/><br/>Figure 1.1 Comparison of fluorescence fingerprints for (left to right) SSF-est, SSF-mix, and SSF-new. Profiles show the combined plot of cell concentration and fluorescence distribution of nucleic acid stained with SYBR Green I in effluent water before wash (blue), and 4 days after washing (red).<br/>Fingerprints of effluent water were stable throughout the sampling period for SSF-est, regardless of washing, while for SSF-new and SSF-mix, fingerprints suggested a destabilization of the bacterial population related to washing. This suggested that SSF-new, SSF-mix and SSF-est differed in the population of cells being seeded into their respective effluent waters and this was supported by results from traditional methods and qPCR, showing lowest number of coliforms in the effluent from SSF-est.  <br/>Analysis by CHIC, showed that the cell populations seeded by SSF-mix were more similar to those from SSF-est, than SSF-new.  FC fingerprints from SSF-new were very similar to those of the influent water, suggesting that the biofilm in this SSF had little impact on the water passing through SSF-new. Since SSF-mix included washed sand from established SSFs, this appears to have inoculated the SSF-mix biofilm in a way that was not observed for SSF-new. The fingerprints for SSF-est effluent remained stable, consistent with a biofilm that is resilient to disturbance, and coliform counts demonstrating uninterrupted SSF function. <br/>This study thus suggests that washed sand is preferable to new sand for SSF construction and that more than 6 months is required for a resilient SSF biofilm. While total cell numbers did not change significantly across any SSF, SSF-est selectively reduced HPC and coliforms suggesting that this biofilm function is selective compared to newer SSF biofilms. In this study FC was not only useful for monitoring total cell counts to follow changes during SSF washing, but showed a consistent effluent water fingerprint associated with a resilient and functional SSF biofilm. As this was not observed for the newly built SSFs, FC can follow both the character and establishment of SSF biofilms and should be helpful for analyzing the impacts of any potential events on SSF biofilm integrity.<br/>References<br/>Allen, M.J., Edberg, S.C. and Reasoner, D.J., (2004). Heterotrophic plate count bacteria—what is their significance in drinking water? International journal of food microbiology, 92(3), pp.265-274.<br/>Arnoldini, M., Heck, T., Blanco-Fernández, A. and Hammes, F., (2013). Monitoring of dynamic microbiological processes using real-time flow cytometry. PloS one, 8(11), p.e80117.<br/>El-Chakhtoura, J., Prest, E., Saikaly, P., van Loosdrecht, M., Hammes, F. and Vrouwenvelder, H., 2015. Dynamics of bacterial communities before and after distribution in a full-scale drinking water network. Water Research, 74, pp.180-190.<br/>Koch, C., Fetzer, I., Harms, H. and Müller, S., (2013). CHIC—an automated approach for the detection of dynamic variations in complex microbial communities. Cytometry Part A, 83(6), pp.561-567.<br/>Prest, E.I., Hammes, F., Kötzsch, S., Van Loosdrecht, M.C.M. and Vrouwenvelder, J.S., (2013). Monitoring microbiological changes in drinking water systems using a fast and reproducible flow cytometric method. Water research, 47(19), pp.7131-7142.<br/>}},
  author       = {{Paul, Catherine J. and Chan, Sandy and Rådström, Peter and Pullerits, Kristjan and Persson, Kenneth M}},
  language     = {{eng}},
  month        = {{09}},
  title        = {{Disturbance of the bacterial communities in drinking water produced from established and newly-built slow sand filters}},
  year         = {{2016}},
}