SOIL ELECTRICAL CONDUCTIVITY AS A COVARIATE TO IMPROVE THE EFFICIENCY OF FIELD EXPERIMENTS

Abstract

An accurate comparison of treatment means over a range of conditions is a primary objective of most fieldexperiments. The spatial variability of soil properties, including soil texture and moisture, can have adverse effects on theaccuracy and efficiency of field trials. Apparent soil profile electrical conductivity (ECa) has been shown to be a usefulindicator of soil variability. Compared to conventional soil sampling, measurements of ECa are rapid and inexpensive. In thisstudy, soil ECa was used as a covariate in evaluating the effect of manure application on soil phosphorus (P) concentration.Compared to a standard analysis of variance (ANOVA), an analysis of covariance (ANCOVA) with soil ECa as a covariatereduced standard errors for treatment means and thereby improved the accuracy of estimates of P concentrations for bothmanure application treatment means and aeration tool tine disturbance position means within treatments. For example, thestandard error for tine disturbance position means from ANOVA was equal to 11.0, while the standard error with ECa as acovariate was equal to 6.6. Different conclusions were drawn regarding treatment effects with and without ECa as a covariate.For example, an ANOVA-based conclusion was that the control (no manure) treatment was not different from thesurface-applied manure treatment. The ANCOVA-based conclusion was that only in soils with low ECa values were the twotreatments not different. In soils with medium and high ECa values, the control treatment had a significantly lower Pconcentration.