Exploring the effects of weather-driven dynamics of desiccation cracks on hydrological process of expansive clay slope: Insights from physical model test

Abstract

Desiccation cracks, serving as preferential pathways for both water evaporation and infiltration, are critical to the hydrological processes governing the stability of expansive clay slopes. However, the desiccation crack dynamics under drying-wetting cycles strongly influence the preferential pathways for water movement. The effects of such weather-driven crack dynamics on slope hydrology remain unclear. To address this gap, a non-failure slope model test was conducted under multiple drying-wetting cycles. Crack metrics (crack ratio, aperture, connectivity, depth), hydrological responses (volumetric water content, matric suction) and slope water balance components (evaporation, runoff, discharge) were measured to investigate the interplay between crack dynamics and slope hydrology. The findings revealed that during drying, evaporation-driven soil shrinkage rapidly expands cracks until the shrinkage limit, followed by temperature-driven “crack breathing,” causing 2 % crack ratio fluctuations within a 10 °C range. Desiccation cracks facilitate fast evaporation within the crack depth but has minimal impact below it. During wetting, deep primary desiccation cracks are the leading pathways for preferential flow even when they are in closing, while the shallow small cracks help retain surface runoff in the surface matrix, slowing deep soil saturation by reducing preferential flow. For expansive clay with dynamic cracks, heavy rainfall after prolonged drought enhances preferential flow, while short-interval rainfall weakens it. Weather-driven crack dynamics have a dual effect on slope water balance: during intense drying, crack development initially promotes slope discharge through preferential evaporation but later limit evaporation. Similarly, during wetting, crack closure initially reduces preferential flow and thus slope water storage, but the decreased aperture and connectivity of crack networks concurrently reduce the slope discharge capacity, potentially increasing slope water storage and impacting slope stability. Our research highlights the need for incorporation of desiccation crack dynamics into hydrological models to quantify such dual effect on slope stability.