Using Inundation Extents to Predict Microbial Contamination in Private Wells after Flooding Events.

Disaster recovery poses unique challenges for residents reliant on private wells. Flooding events are drivers of microbial contamination in well water, but the relationship observed between flooding and contamination varies substantially. Here, we investigate the performance of different flood boundaries─the FEMA 100 year flood hazard boundary, height above nearest drainage-derived inundation extents, and satellite-derived extents from the Dartmouth Flood Observatory─in their ability to identify well water contamination following Hurricane Florence. Using these flood boundaries, we estimated about 2600 wells to 108,400 private wells may have been inundated─over 2 orders of magnitude difference based on boundary used. Using state-generated routine and post-Florence testing data, we observed that microbial contamination rates were 7.1-10.5 times higher within the three flood boundaries compared to routine conditions. However, the ability of the flood boundaries to identify contaminated samples varied spatially depending on the type of flooding (e.g., riverine, overbank, coastal). While participation in testing increased after Florence, rates were overall still low. With <1% of wells tested, there is a critical need for enhanced well water testing efforts. This work provides an understanding of the strengths and limitations of inundation mapping techniques, which are critical for guiding postdisaster well water response and recovery.

Comparison of the Biolog OmniLog Identification System and 16S ribosomal RNA gene sequencing for accuracy in identification of atypical bacteria of clinical origin.

The Biolog OmniLog Identification System (Biolog) and the 16S ribosomal RNA (rRNA) gene sequencing methods were compared to conventional microbiological methods and evaluated for accuracy of bacterial identification. These methods were evaluated using 159 clinical isolates. Each isolate was initially identified by conventional biochemical tests and morphological characteristics and subsequently placed into one of seven categories: aerobic Actinomycetes, Bacillus, Coryneforms, fastidious Gram-negative rods (GNR), non-fermenting GNR, miscellaneous Gram-positive rods (GPR), and Vibrio/Aeromonas. After comparison to the conventional identification, the Biolog system and 16S rRNA gene sequence identifications were classified as follows: a) correct to the genus and species levels; b) correct to the genus level only; or c) neither (unacceptable) identification. Overall, 16S rRNA gene sequencing had the highest percent accuracy with 90.6% correct identifications, while the Biolog system identified 68.3% of the isolates correctly. For each category, 16S rRNA gene sequencing had a substantially higher percent accuracy compared to the conventional methods. It was determined that the Biolog system is deficient when identifying organisms in the fastidious GNR category (20.0%). The observed data suggest that 16S rRNA gene sequencing provides a more accurate identification of atypical bacteria than the Biolog system.