|Guber, Andrey - ARS VISITING SCIENTIST|
|Nemes, Attila - ARS VISITING SCIENTIST|
Submitted to: American Society of Agronomy
Publication Type: Abstract Only
Publication Acceptance Date: June 10, 2004
Publication Date: July 11, 2004
Citation: Weltz, L., Pachepsky, Y.A., Shelton, D.R., Guber, A.K., Nemes, A., Starr, J.L. 2004. Distribution of manure-borne fecal coliforms in soil on grass buffer strip [Abstract]. American Society of Agronomy, Northeastern Branch Annual Meeting, July 11-14, 2004, Bordentown, New Jersey. p.4 . Technical Abstract: Bacteria from animal manure waste are one of the causes of water pollution. Vegetated buffer strips (VFS) are used to reduce manure-borne bacteria transport to waterways. The objective of this study was to determine what abiotic and biotic variables were the best predictors of manure-borne bacteria retention in soils at vegetative buffer strips. Manure was applied as a band of slurry at the top of a 6-meter-wide VFS on 20% slopes. A 25-year storm event was simulated in triplicate at four sites with silt loam and sandy loam soils. Fecal coliform (FC) distributions were measured in six points per replication. The VFS consisted of tall fescue Bermuda grass and white clover. The variables included soil saturated hydraulic conductivity (Ksat), plan curvature, slope, local rain intensity, vegetative biomass by life forms, and canopy and ground cover. The regression trees analysis was used to determine variables that provided the tightest grouping of data and thus explained the most of variation. The tree model split the dataset into two main groups or branches based on its best predicting variable, Ksat. The litter density was the second most influential factor where Ksat values were low. When Ksat was high. The total canopy cover was the second most influential variable where Ksat values were high. The slope was the third influential variable. Distance from the manure application band had no significant effect on FC contents in soil. The regression tree explained about 84% of variation in FC contents in soil. The results indicate that VFS designs must consider the site's soil permeability and vegetation density/cover.