Variability in Sorption Coefficients of Metolachlor on a Captina Silt Loam

Abstract

AbstractSpatial variability of soil properties is important in determining pesticide behavior in soils. A study was conducted on a 4‐ha field to measure the spatial variability in sorption coefficients (K) for the herbicide metolachlor [2‐chloro‐N‐(2‐ethyl‐6‐methylphenyl)‐N‐(2‐methoxy‐1‐methylethyl)acetamide] and to correlate these K values to other soil properties. The soils in this field were Captina and Johnsburg (Typic Fragiudults). The rectangular field was sampled on a grid with 135 nodes. At each node, samples of the Ap, Bt, and Btx horizons were taken for determinations of water content at 10 and 1500 kPa, pH, particle size analysis, sorption coefficients (K), and organic matter content. Means, standard deviations, minimum values, maximum values, and coefficients of variation (CV) were determined on all properties. In addition, multiple linear regression analyses were performed to determine the maximum correlation between K and the soil properties. Soil organic matter content was the soil property most highly correlated with the adsorption of metolachlor in the surface soils (Ap). Adsorption was also significantly correlated with clay content, water content at 1500 kPa, and pH. Semivariograms were constructed for each horizon for the organic matter content and K. Semivariograms for organic matter indicated a range of spatial dependency of 110 m in the Ap and no spatial structure for the Bt and Btx horizons. The range for K was 110 m in the Ap and 42 m in the Bt with no spatial structure evident in the Btx horizon. The data for K and organic matter were normalized by dividing each semivariance by the mean squared for each parameter, which resulted in similar semivariograms in the Ap horizon. Contour plots of kriged values for normalized organic matter and normalized K in the Ap horizon were also similar.