Survival of manure-borne E. coli in streambed sediment: effects of temperature and sediment properties

Authors: Garzio-Hadzick, Amanda; Shelton, Daniel; HILL, ROBERT - University Of Maryland; Pachepsky, Yakov; Guber, Andrey; Rowland, Randy

Submitted to: Water Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/9/2010
Publication Date: 2/13/2010
Citation: Garzio-Hadzick, A.M., Shelton, D.R., Hill, R., Pachepsky, Y.A., Guber, A.K., Rowland, R.A. 2010. Survival of manure-borne E. coli in streambed sediment: effects of temperature and sediment properties. Water Research. 44:2753-2762

Interpretive Summary: E. coli bacteria are commonly used as water quality indicators for designation of impaired surface waters. Recent studies have shown that stream bottom sediments can harbor large numbers of E. coli. During rainfall events, these bottom sediments are resuspended, resulting in high E. coli concentrations in the water. Therefore, an understanding of the factors affecting the survival of E. coli in sediments is critical to predicting potential water contamination. Experiments were conducted to simulate runoff of dilute manure into sediments and the subsequent growth and survival of E. coli strains. A dilute dairy manure slurry containing fresh E. coli was added to sediments with different textures, organic matter contents, and indigenous E. coli concentrations, and incubated in chambers with flowing creek water at 4ºC, 14ºC, and 24ºC. E. coli were initially observed in the water, but quickly dissipated, due to either die-off or settling. E. coli concentrations initially increased in sediments 2- to 5-fold, followed by an oscillatory stage, and then an exponential die-off phase. A direct correlation between temperature and die-off rate was observed for all sediments. The highest die-off rates were observed in sandy sediments with low organic carbon and low indigenous E. coli concentrations, while the lowest die-off rates were observed in silty sediments with high organic carbon content and high indigenous E. coli concentrations. Additional experiments were conducted to investigate whether differences in die-off rates were due to differential rates of survival of indigenous vs. added manure-borne E. coli, or due to sediment properties. Aged sediments, with virtually no indigenous E. coli, were inoculated with a culture of non-pathogenic E. coli, and incubated at 14ºC and 24ºC. Results were comparable with the previous experiments---die-off rates for the sandy, low organic matter sediment were substantially higher than for the silty, high organic matter sediment at both temperatures----indicating that sediment properties were the primary factor responsible for E. coli survival. These results are relevant to research scientists investigating the dynamics of water-borne bacteria and to regulatory agencies tasked with minimizing surface water contamination.

Technical Abstract: E. coli bacteria are commonly used as water quality indicators for designation of impaired surface waters and to guide the design of management practices to prevent fecal contamination of water. Recent studies have shown that stream sediments can serve as a reservoir and potential source of fecal bacteria (E. coli) for stream water. The objectives of this work were to determine (a) if manure-borne E. coli survival in stream sediments is affected by sediment particle size distribution and organic matter content, and (b) if temperature affects E. coli survival in sediments similar to other environmental media. Laboratory experiments were conducted at three temperatures (4ºC, 14ºC, and 24ºC) using sediment from three locations at the Beaverdam Creek Tributary in Beltsville, Maryland mixed with a dairy manure slurry in the proportion of 1000:1. Indigenous E. coli populations in sediments ranged from ca. 101 to 103 MPN per gram while approx 103 manure-borne E. coli MPN per gram were added. Stream conditions were simulated using flow-through chambers. E. coli survived in sediments much longer than in the overlaying water. An initial 2- to 5-fold increase in E. coli sediment concentrations was observed during the first 2 days of incubation, followed by an oscillatory stage, and then an exponential die-off or inactivation stage. A direct correlation between temperature and inactivation rate was observed for all sediments. Inactivation rates were similar at 4ºC, but increasingly diverged at 14ºC and 24ºC. The highest inactivation rates were observed in sandy sediments with low organic carbon while the lowest inactivation rates were observed in silty sediments with high organic carbon content. The increase in sediment clay and organic carbon also decreased the sensitivity of inactivation rates to temperature, evaluated as the ratio of inactivation rates at 24 ºC and 14 ºC. Higher inactivation rates for sediments with low indigenous E. coli populations was necessarily due to die-off of added strains; however, lower inactivation rates for sediments with high indigenous populations could be due to differential rates of die-off between indigenous and added strains. Consequently, additional experiments were conducted with the sandy, low organic matter sediment and the silty, high organic matter sediment at 14ºC and 24ºC. Sediments were aged, allowing for inactivation of essentially all indigenous E. coli (<101 MPN per gram) and inoculated with ca. 104 MPN per gram of a pure culture of E. coli O157:H12. Results were comparable with the previous experiments---inactivation rates for the sandy, low organic matter sediment were substantially higher than for the silty, high organic matter sediment at both temperatures----indicating that sediment properties were the primary factor responsible for survival.