329-18 Rainfall Intensity-Dependent Release of Indicator Bacteria from Dairy Cattle Manure Applied on Vegetated Soil.
See more from this Division: SSSA Division: Soil PhysicsSee more from this Session: Environmental Soil Physics and Hydrology Student Competition: Lightning Orals with Poster Presentations
Tuesday, November 4, 2014: 3:30 PM
Long Beach Convention Center, Room 102B
Simulating the rainfall-induced release of indicator bacteria from manure is an essential component to microbial fate and transport modeling. The objectives of this work were to determine the effect of rainfall intensity on the release of E. coli, enterococci, and total coliforms from manure and to test the performance of three kinetic-based models in describing the observed release. Dairy cattle manure from a CAFO was applied over Kentucky 31 tall fescue in soil boxes (sandy loam; 100 x 35 x 25 cm). Three levels of rainfall intensity (3, 6, and 9 cm hr-1) were used and runoff and leachate samples were collected at gradually increasing time intervals for one hour of release. Following rainfall, soil was sampled at 0, 2, 5, and 10 cm depths. The one-parametric exponential dependence model and the two-parametric Bradford-Schijven and Vadas-Kleinman-Sharpley models were fitted to the release curves. Concentrations of bacteria in the initial runoff were approximately one order of magnitude below their concentrations in manure. The dependency of cumulative bacteria release on rainfall depth included a several order of magnitude increase (approximately 10-5 to 10-0.5 %) of the released mass during the initial 1 to 2 cm of rainfall, after which time the mass release rate drastically decreased and became closer to zero. Kinetics for release of the three bacteria substantially differed. Although rainfall intensity did not significantly affect most parameters in the model fits, it affected the release of E. coli more than the other bacteria and it significantly affected the content of all bacteria found at different soil depths after rainfall. Model performance, assessed with the root-mean-squared-error and the Akaike information criterion, was organism-dependent. The Vadas-Kleinman-Sharpley model performed better for enterococci release, while the Bradford-Schijven model yielded the best simulation results for E. coli and total coliform release.
See more from this Division: SSSA Division: Soil PhysicsSee more from this Session: Environmental Soil Physics and Hydrology Student Competition: Lightning Orals with Poster Presentations