Field‐Scale Study of Chlortoluron Movement in a Sandy Soil over Winter: II. Modeling

Abstract This study was conducted to evaluate different modeling approaches for their ability to predict chlortoluron movement observed in a 4‐mo field study. The data base includes the results of a Br − transport experiment performed at the same site and of laboratory sorption‐desorption (SD) experiments. The approaches tested are based on the one‐dimensional convection dispersion equation (CDE) with two‐stage or equilibrium sorption. The CDE was applied either at the field scale or, to account for parameter variability, within a parallel soil columns (PSC) approach. Parallel soil column simulations were realized by the Monte Carlo (MC) method. Transport and sorption parameters and their distributions were estimated from Br − concentration profiles and SD isotherms, respectively. Chlortoluron degradation had been investigated in a joint project. The CDE with two‐stage sorption represented the movement of the chlortoluron peak well, but not its spreading. Similar results were obtained assuming equilibrium sorption, provided the same sorption parameters were used. If the sorption parameters were derived from the 1‐d sorption isotherm instead, displacement was overpredicted. Accounting for the variability of sorption and transport parameters through the PSC approach affected the simulation results only slightly. All simulations substantially underpredicted chlortoluron spreading, probably because the time scales of the laboratory and the field experiments were too different. To convert the estimated rate parameter to the appropriate time scale, we derived a simple formula that is based on the hypothesis of intrasorbent diffusion. Application of the CDE with the rescaled rate parameter distinctly improved the agreement between modeled and measured chlortoluron concentration profiles.