Redistribution of Escherichia Coli in Intact Soil Macro-Aggregates Using Multivariate Analysis

• Activity Number: 254
• Type: Contributed
• Date/Time: Monday, July 30, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Wei Wang*+ and Alexandra Kravchenko and Timothy Johnson and Sangeetha Srinivasan and Alvin Smucker and Mark Rivers
• Companies: Vanderbilt University and Michigan State University and Michigan State University and Michigan State University and Michigan State University and Argonne National Lab
• Keywords: Redistribution ; pore characteristics ; soil aggregate ; multivariate analysis ; principal component analysis ; partial least square regression

Abstract

Past research had demonstrated the ability of E. coli to regrow in the soil environment, to survive and finally to be released during saturated conditions. Pore configurations within soil aggregates is imperative for understanding soil micro-organism movement and functioning. The objective of this study is to explore the use of multivariate analysis techniques to understand the redistribution pattern of E.coli within soil aggregates and relate E. coli movement within aggregates to the aggregates' pore structures in the aggregates of three studied soil treatments - conventionally tilled (CT) and no-till (NT) corn/soybean/wheat rotation and native succession vegetation (NS) at Long-Term Ecological Research site, southwest Michigan. Univariate analysis using each pore characteristics individually was suffering small sample size issues, thus inappropriate for this study. In contrast, categorical principal component analysis(PCA) showed clear distinction among aggregates from different treatments. Using partial least square regression (PLSR), E.coli spatial distribution was somehow related to pore characteristics (large pores and max pore-network flow).