E. coli Transport through Surface-Connected Biopores Identified from Smoke Injection Tests

Abstract

Macropores are the primary mechanism by which fecal bacteria from surface-applied manure can be transported into subsurface drains or shallow groundwater bypassing the soil matrix. Limited research has been performed investigating fecal bacteria transport through specific macropores identified in the field. The objective of this research was to better understand how fecal bacteria, using Escherichia coli (E. coli) as an indicator organism, are transported through naturally occurring macropores and potential interactions between the macropore and soil matrix domains in the field under controlled experimental conditions. Direct injection/infiltration tests were conducted in two naturally occurring, surface-connected macropores (biopores) that penetrated to the subsurface drain depth, as identified by smoke tests. Data included total drain flow rate (baseflow rate and biopore flow rate), biopore inflow rate, and Rhodamine WT and E. coli concentrations in the drains. Analysis techniques included determining increases in subsurface drain flow rates due to the infiltration tests and percentage of the injected concentration reaching the subsurface drains after dilution with the drain baseflow. In the absence of data for mechanistic models, empirically based rational polynomial models were compared to the more commonly utilized lognormal distribution for modeling the load rate breakthrough curves. Load estimates were derived from integrated forms of these empirical functions, and percent reductions were calculated for Rhodamine WT and E. coli. Peak total drain flow rates increased nearly two-fold due to direct injection into the biopores. Less than 25% of the initial concentrations injected into the biopores reached the drain after dilution with the baseflow in the drain. Lognormal distributions best fit the Rhodamine WT load rate breakthrough curves (R2 = 0.99 for both biopores) and the E. coli load rate data for one of the biopores (R2 = 0.98). A rational fractional polynomial model that tailed off more slowly best fit the E. coli load rate data for the other biopore (R2 = 0.98). Approximately one log reduction was estimated for E. coli loads due to interaction with the soil profile as water flowed through the tortuous path of the biopores; in other words, the soil surrounding the biopore filtered approximately 90% of the E. coli load that entered the biopore compared to approximately 75% for Rhodamine WT. Considering that applied animal manure can contain millions of bacteria per mL, high concentrations and loads are still possible in the subsurface drain flow if macropores are present.