Effects of field storage method on E. coli concentrations measured in storm water runoff

Abstract

Storm water runoff is increasingly assessed for fecal indicator organisms (e.g., Escherichia coli, E. coli) and its impact on contact recreation. Concurrently, use of autosamplers along with logistic, economic, technical, and personnel barriers is challenging conventional protocols for sample holding times and storage conditions in the field. A common holding time limit for E. coli is 8 h with a 10 °C storage temperature, but several research studies support longer hold time thresholds. The use of autosamplers to collect E. coli water samples has received little field research attention; thus, this study was implemented to compare refrigerated and unrefrigerated autosamplers and evaluate potential E. coli concentration differences due to field storage temperature (storms with holding times ≤24 h) and due to field storage time and temperature (storms >24 h). Data from 85 runoff events on four diverse watersheds showed that field storage times and temperatures had minor effects on mean and median E. coli concentrations. Graphs and error values did, however, indicate a weak tendency for higher concentrations in the refrigerated samplers, but it is unknown to what extent differing die-off and/or regrowth rates, heterogeneity in concentrations within samples, and laboratory analysis uncertainty contributed to the results. The minimal differences in measured E. coli concentrations cast doubt on the need for utilizing the rigid conventional protocols for field holding time and storage temperature. This is not to say that proper quality assurance and quality control is not important but to emphasize the need to consider the balance between data quality and practical constraints related to logistics, funding, travel time, and autosampler use in storm water studies.

Escherichia coli Antimicrobial Resistance Variability in Water Runoff and Soil from a Remnant Native Prairie, an Improved Pasture, and a Cultivated Agricultural Watershed

“…Within GSWRL, the experimental watersheds representing improved hay pasture, native prairie and cultivated cropland were used for this study. The native prairie (SW12) is a 1.2 ha remnant native prairie plot with a 3.8% slope that has been consistently managed since 1948 [39,40]. Management practices for the sites include mowing or haying interspersed with intermittent herbicide treatments and prescribed burns.…”

“…The cultivated cropland field (Y6) is a 6.6 ha, terraced, conventionally cultivated cropland site with 3.2% slope [39,40] that has been continuously cropped since 1943. During the study, corn and wheat were grown.…”

Elucidating the Effects of Land Cover and Usage on Background Escherichia coli Sources in Edge‐of‐Field Runoff

“…This confounds E. coli fate and transport modeling (Harmel et al, 2006a, 2010). Large measured E. coli concentration variations in edge‐of‐field and small watershed runoff (Donnison et al, 2004; Goto and Yan, 2011; Harmel et al, 2013, 2016; Wagner et al, 2012; Wittman et al, 2013) and soil as a potential E. coli source in upland watersheds are two of these factors. Soil E. coli concentrations up to 20,800 colony forming units (cfu) g −1 have been measured in tropical and temperate soils (Byappanahalli et al, 2006, 2012a, 2012b; Goto and Yan, 2011; Ishii et al, 2006; Pandey and Soupir, 2013; Soupir and Mostaghimi, 2011), and streambank soil may be a significant source contributing to streams (Byappanahalli et al, 2003; Desmarais et al, 2002; Solo‐Gabriele et al, 2000).…”

Isotopic (δ18O and δ2H) Integrity of Water Samples Collected and Stored by Automatic Samplers

“…1998; Jarvie et al, 2002;Harmel et al, 2006;2009;2016a;2016b), uncertainty in stable water isotope signatures has not been quantified. During the time between sample collection and sample retrieval from the autosampler, water samples are left open to the atmosphere and may be at risk for evaporative losses.…”

“…Many monitoring programs, however, use automatic water samplers to collect and store samples from hours to weeks, depending on research objectives and methodologies. While uncertainty associated with automated sampling and analysis of sediment, nutrients, and bacteria has been estimated (Kotlash and Chessman, 1998; Jarvie et al, 2002; Harmel et al, 2006; 2009; 2016a; 2016b), uncertainty in stable water isotope signatures has not been quantified.…”