Characterisation of flow paths and saturated conductivity in a soil block in relation to chloride breakthrough

The nature of flow paths, determined by soil structural features, may greatly influence solute breakthrough in a soil profile. We compared two contrasting methods of characterising flow within an upland soil with distinct organic and mineral horizons: saturated conductivity measured on cores sampled destructively; and breakthrough measurements for chloride in 2.4 x 3.4 x 1 m in situ lysimeter. Chloride tracer was applied to the soil surface, and soil solution samples collected at 20 known locations using suction cup samplers. Breakthrough curves were classified into statistically distinct pathway types using Principal Coordinate Analysis, according to peak concentration and the time to peak concentration. Of the 20 locations, two exhibited rapid macropore flow, two intermediate mesopore flow, and seven slower micropore flow. The remaining nine samplers did not register breakthrough within the 16-day duration of the experiment. Destructive core (6 cm diameter by 6 cm deep) sampling was used to characterise saturated hydraulic conductivity at 116 locations within the soil block. Solute breakthrough speed was linearly related to kriged values of saturated conductivity for the meso and micropore flow paths (r = 0.88 for initial breakthrough; r = 0.85 for peak concentration breakthrough), but not for macropore flow. This indicates that kriged saturated conductivities provided good prediction of the speed of the meso and micro flow paths. Frequency distributions of saturated hydraulic conductivities and breakthrough speeds did not differ significantly, although the breakthrough speed distribution was truncated by the limited duration of the lysimeter experiment. Lysimeter breakthrough for macropore pathways was faster than predicted from the core conductivities, indicating that the locations of these fast flow paths were not predicted accurately by the kriged saturated conductivities. Longer duration lysimeter experiments would be required to characterise breakthrough through the slowest pathways, and so these pathways were better characterised using the destructive saturated conductivity technique. (c) 2007 Elsevier B.V. All rights reserved.