Soil-water content dependency of water repellency in soils: effect of crop type, soil management, and physical-chemical parameters.

摘要

Water repellency (WR) of soils is a global phenomenon. It affects hydrological processes such as infiltration, preferential flow, and surface erosion. Although the soil WR varies nonlinearly with soil-water content (w), WR is often determined at one fixed soil-water content. For a coarse sandy soil, we provide a comparison of the whole WR-w curve under a range of varying factors: (i) crop type, (ii) straw incorporation, removal, or burning, (iii) levels of nitrogen addition, (iv) pH, and (v) soil organic carbon content. The curves were determined on soil samples ranging from oven-dried at 105 degrees C to soil-water contents above which the soils become permanently wettable using the molarity of an ethanol droplet test. The trapezoidal integrated area under the WR-w curves was used as an index for characterizing WR. Soil under perennial grass followed by forest showed the highest WR, whereas the cereal crops wheat, oat, and rye had the lowest WR over the whole range of soil-water contents. Postharvest straw incorporation gave a higher WR compared with straw removal or burning. The WR tended to decrease with increasing levels of nitrogen addition. For low-phosphorus input soils, WR tended to decrease with increasing pH; for normal phosphorus inputs, we found no clear effect of increasing lime application (pH). The trapezoidal integrated area under the WR-w curves, the soil-water content at the highest value of WR, and the soil-water content above which the soil becomes wettable were significantly positively correlated with the soil organic carbon content throughout. Plotting the soil moisture-dependent WR behavior for this and comparison soils from literature in the form of WR versus pF (log[-soil-water matric potential, in cm H< sub>2O]) showed a normalizing effect on WR behavior with an initial increase (from drainage) in WR situated around pF 2.5-3. Thus, it is crucial to consider the whole WR-w curve and not just a single point when evaluating soil WR.