Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: June 2, 2006
Publication Date: November 4, 2006
Citation: Guber, A.K., Pachepsky, Y.A., Shelton, D.R., Van Genuchten, M.T. 2006. Preferential E. coli transport in saturated soil columns. Preferential Flow and Transport Processes in Soil Conference, November 4-9,2006, Monte Verita, Ascona, Switzerland. Technical Abstract: By the frequency of being the cause of water quality impairment, pathogenic microorganisms rank first and second among five leading pollutants in estuaries and rivers, respectively, in the United States. Animal manures contain large numbers of pathogenic bacteria including pathogenic strains of Escherichia coli. Vegetated buffer strips are a common management practice to prevent bacteria transport to watercourses. The efficiency of the vegetated filter strips depends on the ability of bacteria to move in soil with the infiltrating water. Soil columns are often used to evaluate the infiltration ability and transport parameters of bacteria. The objective of this work was to compare sieved soil columns, soil aggregate columns, and undisturbed soil columns in terms of the information they give about the infiltration of E. coli bacteria that are relatively large “living colloids.” Twenty-cm long and 7.5 cm ID columns were used in the study. The air-dry clay loam soil was passed through the 2-mm sieve and used in one group of columns; another group of columns was packed with the soil aggregates remaining on the sieve. Soil porosity was about 0.43 cm3 cm-3 in all columns. The columns were capillary-saturated from the bottom. Pulses of E. coli suspension with and without manure suspensions of about 1.2 pore volume were passed through the columns. Chloride ion from KCl was used as a tracer. E. coli concentrations and turbidity were measured in influent and effluent. Columns were cut into 2-cm layers after the experiments to measure viable bacteria concentrations and soil water content. The experiments were carried out at 9±1 ºC. Chloride recovery in effluent was 49-55% after one pore volume passed through sieved soil, and 92-98% in the aggregated soil columns. Effluent E. coli concentrations were negligible in sieved clay loam soil columns. Distinct peaks of viable E.coli were observed in effluents from the soil aggregate columns. The peak of the effluent E. coli concentration was observed much earlier in the experiments with manure-bacteria suspensions than in pure E. coli suspensions applied. Peaks of E. coli concentrations coincided with turbidity peaks, reflecting similarity in manure colloid and E. coli transport. From 86% to 95% of all viable E. coli in soil were concentrated in the upper 4 cm layer of sieved soil columns. E. coli were distributed relatively uniformly along columns with aggregated clay loam soil. The macropore flow was critical for bacteria transport in disturbed soil. Undisturbed cores of Tyler silt loam soil were taken with the 30-cm spacing at the flood plain grassland. Particles larger than 2 mm dominated the aggregate size distribution. The saturated hydraulic conductivity ranged from 2.3 to 9.3 cm day-1, and increased with soil porosity that ranged from 0.42 to 0.45 cm3 cm-3. A pulse of the manure E.coli suspension with the KCl solution was passed through the columns. The E.coli transport was similar to the chloride transport in slow-flow columns before half of the pore solution was replaced with the influent, but was retarded after that. The Cl breakthrough was delayed compared to E. coli breakthrough in the fast flow columns. The bacteria leaching was observed much longer compared with that of chloride in those columns. The non-equilibrium, two-region convective-dispersive transport model was fitted to the breakthrough curves using CXFITIM. The estimated proportion of pore space available for Cl and E.coli transport ranged from 54 to 89 %, and from 3% to 23 %, respectively. The pore space available for E. coli transport was larger in columns with faster flow. Transport velocities in mobile zone were from 2 to 31 times higher for E. coli than for Cl. Overall, using undisturbed columns appears to be the preferable for manure-borne bacteria transport studies compared with sieved-soil and soil aggregate columns. Undisturbed soil columns demonstrate a substantial ability of manure-borne E. coli to infiltrate and be retained in well-structured soils where the preferential bacterial transport can occur in relatively small fraction of pore space and can be enhanced by the presence of manure colloids.