Search for a command to run...
Abstract. The ploughing of soils in autumn drastically loosens the soil structure and, at the same time, reduces its stability against external stresses. A fragmentation of these artificially produced soil clods during wintertime is often observed in areas with air temperatures fluctuating around the freezing point. From the pore perspective, it is still unclear (i) under which conditions frost action has a measurable effect on soil structure, (ii) what the impact on soil hydraulic properties is, and (iii) how many freeze–thaw cycles (FTCs) are necessary to induce soil structure changes. The aim of this study was to analyse the cumulative effects of multiple FTC on soil structure and soil hydraulic properties for two different textures and two different initial structures. A silt clay with a substantial amount of swelling clay minerals and a silty loam with fewer swell/shrink dynamics were either kept intact in undisturbed soil cores taken from the topsoil from a grassland or repacked with soil clods taken from a ploughed field nearby. FTCs were simulated under controlled conditions and changes in pore structure ≥ 48 µm were regularly recorded using X-ray µCT. After 19 FTCs, the impact on hydraulic properties were measured, and the resolution of structural characteristics were enhanced towards narrow macropores with subsamples scanned at 10 µm. The impact of FTC on soil structure was dependent on the initial structure, soil texture, and the number of FTCs. Frost action induced a consolidation of repacked soil clods, resulting in a systematic reduction in pore sizes and macropore connectivity. In contrast, the macropore systems of the undisturbed soils were only slightly affected. Independent of the initial structure, a fragmentation of soil clods and macro-aggregates larger than 0.8 to 1.2 mm increased the connectivity of pores smaller than 0.5 to 0.8 mm. The fragmentation increased the unsaturated hydraulic conductivity of all treatments by a factor of 3 in by a factor of 3 in a matrix potential range of −100 to −350 hPa, while water retention was only slightly affected for the silt clay soil. Already 2 to 5 FTCs enforced a well-connected pore system of narrow macropores in all treatments, but it was steadily improved by further FTCs. The implications of fewer FTCs during milder winters caused by global warming are twofold. In ploughed soils, the beneficial seedbed consolidation will be less intense. In grassland soils, which have reached a soil structure in dynamic equilibrium that has experienced many FTCs in the making, there is still a beneficial increase in water supply through increasing unsaturated hydraulic conductivity by continued FTCs that might also be less efficient in the future.