Predicting the Fate of Preferentially Moving Herbicides

Abstract

Core Ideas A simple multiregion model predicts herbicide transport well using tracer data. A chloride tracer is used to characterize the preferential flow paths. A surface layer in the model distributes water and solutes to the preferential pathways. Simulation of preferential flow remains a challenge despite being a recognized phenomenon. With short‐interval data, we adapted and tested the preferential flow model (PFM) to simulate the vertical transport of herbicides to lower soil layers. The PFM divides the soil profile into a top distribution zone and a conveyance zone below. The distribution zone acts as a reservoir, with an exponential loss of solutes to the conveyance zone. In the conveyance zone, water and solutes move as convective–dispersive flow through multiple flow paths—preferential and matrix—to shallow groundwater. Our field experiment was performed on a structured Hudson silty clay loam soil (a fine, illitic, mesic Glossaquic Hapludalf) that exhibits preferential flow. The site was instrumented with a variety of soil water samplers placed at depths of 60 cm to monitor the volume and quality of the leachate. Agronomic application of atrazine [6‐chloro‐N‐ethyl‐N′‐(1‐methylethyl)‐1,3,5‐triazine‐2,4‐diamine] and 2,4‐D [2‐(2,4‐dichlorophenoxy)acetic acid] was used, followed by 75 cm of controlled and natural rainfall over 100 d. In addition, Cl− was applied as a conservative tracer. All samplers monitored during this period showed a fast breakthrough of solutes consistent with the occurrence of preferential flow, with two groups of breakthrough curves observed. By fitting the Cl− breakthrough curve for each group, PFM input parameters were estimated including water velocity in preferential flow paths and the fraction of water moving through each flow path. With two additional parameters for herbicide adsorption and degradation rates, the model successfully simulated the extent of preferential flow of herbicides.