Preferential Flow Effects on Chemical Transport and Retardation in Soils

The study’s objective was to investigate the presence of preferential flow and its affect on the transport of a conservative tracer, m-trifluoromethylbenzoic acid (m-TFMBA), and a mildly retarded tracer, 5-bromo-3-sec-butyl-6-methyluracil, (bromacil) through two in situ and two1g repacked soil columns. All columns were of the same bulk density and soil type. Each measured 15-cm in diameter by 32-cm high. Soil in the in situ columns showed no visible cracks or macropores. A comparison of the degree of preferential flow between steady state unsaturated and intermittent ponding flow regimes was investigated.

Curve fitting average pore-water velocities to solute flux-averaged breakthrough curves (BTCs) indicated all columns had some degree of preferential flow for both flow regimes. Preferential flow during the intermittent ponding flow regime was much more pronounced in the in situ soil columns. The repacked soil columns showed close to ideal miscible displacement for both flow regime. Comparing BTCs between the two column types implied that soil structure of the in situ columns promoting preferential flow was destroyed during the sieving and repacking process.

During the steady state-unsaturated flow regime, the appearance of both tracer peaks required less cumulative effluent than that of the intermittent ponding flow regime. It is speculated that molecular diffusion and concentration gradients drove both tracers into the less mobile fluid-filled pores during the quiescent period of the intermittent ponding flow regime. This would make the tracers more resistant to miscible displacement during subsequent ponds.

Study results lead to three deductions. One, that preferential flow is highly possible in a seemingly structureless soil profile during steady state-unsaturated flow and intermittent ponding and therefore solute transport may be underestimated using conventional transport models. Two, due to the presence of preferential flow and the time interval between ponding events, intermittent ponding can be less effective than steady state-unsaturated flow in flushing surface-applied salts out of a homogenous soil profile. Three, retardation factors (RfS), derived from batch isotherm partitioning coefficients (Kds), are good predicators of mildly retarded bromacil movement during the steady state-unsaturated flow regime. For the intermittent ponding flow regime, high RfS underestimated the bromacil’s initial appearance for the in situ columns, yet adequately described the movement of bromacil BTC peaks. Underestimating bromacil’s initial appearance relative to m-TFMBA was a result of preferential flow. Preferential flow caused such rapid solute movement through the soil profile that bromacil adsorption kinetics did not come to equilibrium. Bromacil RfS were higher for the intermittent ponding flow regime than for the steady state-unsaturated flow regime. This was due to all columns having lower moisture contents during the long quiescent period between ponding events.

This report is based largely upon the Master’s thesis of Turney (1991), which may be consulted for futher details.