Frequentist and Bayesian Approaches to Understanding Changes in Redox Potential Due to Hydrology and Vegetation in Agricultural Drainage Ditches

Abstract

Excess nutrient loading from agricultural landscapes contributes to downstream water quality degradation. To mitigate these issues, agricultural drainage ditches have recently gained attention as potential sites for nutrient reduction. We examined the effects of vegetation and hydrology on oxidation-reduction (redox) potential. Testing occurred in an agricultural drainage ditch fitted with weirs in the summer of 2012. Redox potential was recorded using continuous automated data loggers and analyzed using both frequentist and Bayesian methods. Significant difference was found when analyzing redox potential response to vegetation (t = −1.75, P = 0.08, df = 9754) and hydrology treatments (t = 7.51, P < 0.001, df = 9754) in a frequentist manner. Vegetation and hydrologic treatments were significant when analyzed using Bayesian methods; however, the interactions of the two terms had the greatest posterior weights. This study suggests that the innovative use of vegetation and controlled drainage can affect Eh, in particular Eh heterogeneity within ditch systems. Results obtained by analyzing these data in both a frequentist and Bayesian methods were similar in terms significance and magnitude. However, Bayesian methods, and their ability to incorporate prior information into a management framework, may be better suited for systems where previous information can be incorporated into analyses.

from #AlexandrosSfakianakis via Alexandros G.Sfakianakis on Inoreader http://ift.tt/2o0kjOF
via IFTTT