Reach specificity in sediment E. coli population turnover and interaction with waterborne populations.

Sediment-borne Escherichia coli can elevate waterborne concentrations through sediment resuspension or hyporheic exchange. This study sought to correlate hydrological, sediment transport, and water quality variables with: (i) the temporal stability of sediment E. coli populations [concentrations, strain richness and similarity (Raup-Crick index)]; and (ii) the contribution of sediment E. coli to the water column as defined through a library-dependent microbial source tracking approach that matched waterborne E. coli isolates to sediment E. coli populations. Three monitoring locations differing in their hydrological characteristics and adjacent upland fecal sources (dairy operation, low-density residential, and tile-drained cultivated field) were investigated. Sediment E. coli population turnover was influenced by sediment transport at upstream, high-energy reaches, but not at the downstream low-energy reach. Sediment contributions to the water column averaged 13% and 18%, and fecal sources averaged 17% and 21% at the upstream sites adjacent to dairy operations and low-density residential areas, respectively. Waterborne E. coli at the downstream site had low matches to E. coli from reach sediments (1%), higher matches to the upstream sediments (27% and 12%), and an average of 14% matches to the tile drained field. The percentage of waterborne E. coli matching sediment-borne E. coli at each stream reach varied in correlations to hydrological and sediment transport variables, suggesting reach-specific differences in the role of sediment resuspension and hyporheic exchange on E. coli transport.

Characterizing spatial structure of sediment E. coli populations to inform sampling design.

Escherichia coli can persist in streambed sediments and influence water quality monitoring programs through their resuspension into overlying waters. This study examined the spatial patterns in E. coli concentration and population structure within streambed morphological features during baseflow and following stormflow to inform sampling strategies for representative characterization of E. coli populations within a stream reach. E. coli concentrations in bed sediments were significantly different (p = 0.002) among monitoring sites during baseflow, and significant interactive effects (p = 0.002) occurred among monitoring sites and morphological features following stormflow. Least absolute shrinkage and selection operator (LASSO) regression revealed that water velocity and effective particle size (D 10) explained E. coli concentration during baseflow, whereas sediment organic carbon, water velocity and median particle diameter (D 50) were important explanatory variables following stormflow. Principle Coordinate Analysis illustrated the site-scale differences in sediment E. coli populations between disconnected stream segments. Also, E. coli populations were similar among depositional features within a reach, but differed in relation to high velocity features (e.g., riffles). Canonical correspondence analysis resolved that E. coli population structure was primarily explained by spatial (26.9­31.7 %) over environmental variables (9.2­13.1 %). Spatial autocorrelation existed among monitoring sites and morphological features for both sampling events, and gradients in mean particle diameter and water velocity influenced E. coli population structure for the baseflow and stormflow sampling events, respectively. Representative characterization of streambed E. coli requires sampling of depositional and high velocity environments to accommodate strain selectivity among these features owing to sediment and water velocity heterogeneity.

Effect of hillslope position and manure application rates on the persistence of fecal source tracking indicators in an agricultural soil.

The influence of liquid dairy manure (LDM) application rates (12.5 and 25 kL ha) and soil type on the decay rates of library-independent fecal source tracking markers (host-associated and mitochondrial DNA) and persistent (>58 d) population structure was examined in a field study. The soils compared were an Aquic Haplorthod and a Typic Haplorthod in Nova Scotia, Canada, that differed according to landscape position and soil moisture regime. Soil type and LDM application rate did not influence decay rates (0.045-0.057 d). population structure, in terms of the occurrence of abundance of strain types, varied according to soil type ( = 0.012) but did not vary by LDM application rate ( = 0.121). Decay of ruminant-specific (BacR), bovine-specific (CowM2), and mitochondrial DNA (AcytB) markers was analyzed for 13 d after LDM application. The decay rates of BacR were greater under high-LDM application rates (0.281-0.358 d) versus low-LDM application rates (0.212-0.236 d) but were unaffected by soil type. No decay rates could be calculated for the CowM2 marker because it was undetectable within 6 d after manure application. Decay rates for AcytB were lower for the Aquic Haplorthod (0.088-0.100 d), with higher moisture status compared with the Typic Haplorthod (0.135 d). Further investigation into the decay of fecal source tracking indicators in agricultural field soils is warranted to assess the influence of soil type and agronomic practice on the differential decay of relevant markers and the likelihood of transport in runoff.

Evaluation of statistical models for predicting Escherichia coli particle attachment in fluvial systems.

Modeling surface water Escherichia coli fate and transport requires partitioning E. coli into particle-attached and unattached fractions. Attachment is often assumed to be a constant fraction or is estimated using simple linear models. The objectives of this study were to: (i) develop statistical models for predicting E. coli attachment and virulence marker presence in fluvial systems, and (ii) relate E. coli attachment to a variety of environmental parameters. Stream water samples (n = 60) were collected at four locations in a rural, mixed-use watershed between June and October 2012, with four storm events (>20 mm rainfall) being captured. The percentage of E. coli attached to particles (>5 µm) and the occurrences of virulence markers were modeled using water quality, particle concentration, particle size distribution, hydrology and land use factors as explanatory variables. Three types of statistical models appropriate for highly collinear, multidimensional data were compared: least angle shrinkage and selection operator (LASSO), classification and regression trees using the general, unbiased, interaction detection and estimation (GUIDE) algorithm, and multivariate adaptive regression splines (MARS). All models showed that E. coli particle attachment and the presence of E. coli virulence markers in the attached and unattached states were influenced by a combination of water quality, hydrology, land-use and particle properties. Model performance statistics indicate that MARS models outperform LASSO and GUIDE models for predicting E. coli particle attachment and virulence marker occurrence. Validating the MARS modeling approach in multiple watersheds may allow for the development of a parameterizing model to be included in watershed simulation models.