Field-scale solute transport is extremely complex due to the variability of soil properties. In this paper, a mechanistic-stochastic procedure for predicting field-scale solute transport was developed and evaluated. The mechanistic-stochastic method was based on a stream-tube model, which makes predictions using field estimates of shallow soil mobile water contents (θm) determined with a multiport permeameter. This procedure was evaluated for its ability to predict solute transport in a field soil. A field leaching experiment was conducted using fluorinated benzoic acid tracers that were applied to the surface at 12 plots in a field under pivot irrigation. Soil cores were taken to 1.5-m depth at 1, 8, 18, and 36 days after initial tracer application to determine tracer redistribution. The mechanistic-stochastic method was compared with the observed data and against another stochastic method, a transfer function model (TFM), to evaluate its ability to predict solute redistribution for each sampling date. The proposed method had satisfactory predictions of the mean solute concentration redistributions from the leaching experiment and compared well against the predictions made by the TFM. The coefficients of determination ranged from 0.57 to 0.92 for the mechanistic-stochastic method and 0.29 to 0.99 for the TFM. The mechanistic-stochastic method can be greatly simplified, making it practical and desirable over the TFM, which required extensive subsoil leaching data for its calibration.