Field and laboratory studies of Escherichia coli decay rate in subtropical coastal water

• Laboratory study of Escherichia coli decay rate in subtropical coastal water is carried out.
• E. coli decay rate is 13.2–106.5 d−1 under light-exposure.
• A decay formula is derived and validated against field study at a coastal beach.
• Parallel tracer method is used with E. coli and freshwater as natural tracers.
• Provide more realistic estimates of decay rate in subtropical coastal water.

Escherichia coli is a commonly used bacterial indicator of water quality. Despite the significance of E. coli decay rate in many engineering applications, there has been scant field studies in subtropical coastal waters. In this study, laboratory experiments and field studies for E. coli decay rates are carried out for the first time in Hong Kong coastal waters. E. coli decay rates under typical ranges of light intensity, water temperature and salinity in the subtropical coastal waters are determined in the laboratory experiments. It is found that light intensity is the most significant factor affecting E. coli decay rate: 14.7–107 d−1 under light-exposure as compared to 0.85–1.50 d−1 in darkness. A decay formula is derived based on the laboratory results and validated against field studies of in-situ E. coli decay rate carried out at a coastal beach in Hong Kong. A parallel tracer method with reference to storm events is used, with the stream flow and the E. coli loading from the storm water runoff as the two sources of tracers. The freshwater concentration of a beach sample measures the physical dilution due to the mixing of the stream flow with the marine water. The E. coli concentration measures both the physical dilution and the bacterial decay; the biological decay of E. coli can then be determined from the measured freshwater and E. coli concentrations. The in-situ E. coli decay rates range from 1.3 to 5.1 d−1 for the four episodic storm events; the corresponding time required to have a 90% reduction of the initial E. coli level ranges from 10.8 to 42.3 h. The decay rates can be reasonably predicted by the formula derived from laboratory experiments. The present findings provide more realistic estimates of bacterial decay rate than values inferred from indirect model calibration against sparsely sampled data, and enable accurate predictions of water quality accounting for spatial and temporal variations of environmental factors.