The dynamic nature of transient macropore flow in two soils under conventional and conservation tillage.

摘要

The effect of multiple rain events on macropore flow, a major cause of the rapid appearance of agricultural chemicals in groundwater, was examined in clay and silt loam soils under no-till (NT) and plow-till (PT). Undisturbed samples measuring 0.3 x 0.3 x 0.3 m, collected from untrafficked interrows of fields in corn (Zea mays L.), were installed on grid wick samplers. The spatial distribution of drainage from the 5 x 5 grid of cells drained by 0.2 m wicks was determined over time intervals ranging from 0.3 to 12 h. Rains were produced by a constant-head drop-forming rainfall simulator designed for this experiment. For the silt loam soil, two rain treatments consisting of four 37-mm (High) and three 12-mm (Low) rains were applied to three replicate sets of columns. For the clay soil, four replicate pairs of columns received five rains averaging 40 mm. Concentrations of a KBr tracer applied to all soil columns were determined in drainage from each cell wick in each time interval. Outflow concentrations of atrazine applied to clay soil columns were determined for each time interval in the first event.;Spatial distributions of the transient outflows from both soils and tillage treatments were non-uniform and differed between rain events. Over all rain events, drainage was greater in NT than in PT in the clay soil and in the silt loam Low-Rain. High rainfall rates on the silt loam resulted in pore clogging, eliminating differences between tillage treatments. Tracer concentrations in drainage were generally higher in NT than in PT in both soils. However, cumulative tracer mass leached differed between tillage treatments only in the silt loam Low-Rain. Bypass flow was more prevalent in NT, and mixing was more prevalent in PT. Atrazine and tracer fluxes were correlated in NT but not in PT. Flowpaths changed with each rain event, indicating that the transient flow domains were sensitive to spatially heterogeneous antecedent conditions. Experiments conducted under realistic rainfall and surface conditions can provide information on transient flow domains which is masked under steady state or saturated flow conditions.