Piston and preferential water flow are viewed as the two dominant water transport mechanisms regulating terrestrial water and solute cycles. However, it is difficult to accurately separate the two water flow patterns because preferential flow is not easy to capture directly in field environments. In this study, we take advantage of the afforestation induced desiccated deep soil, and directly quantify piston and preferential water flow using chloride ions (Cl⁻) and soil water profiles, in four deforested apple orchards on the Loess Plateau. The deforestation time ranged from 3 to 15 years. In each of the four selected orchards, there was a standing orchard that was planted at the same time as the deforested one, and therefore the standing orchard was used to benchmark the initial Cl⁻ and soil water profiles of the deforested orchard. In the deforested orchards, piston flow was detected using the migration of the Cl⁻ front, and preferential flow was measured via soil water increase below the Cl⁻ front. Results showed that in the desiccated zone, Cl⁻ migrated to deeper soil after deforestation, indicating that the desiccated soil layer formed by the water absorption of deep-rooted apple trees did not completely inhibit the movement of water. Moreover, there was an evident increase in soil water below the downward Cl⁻ front, directly demonstrating the existence of preferential flow in deep soil under field conditions. Although pore water velocity was small in the deep loess, preferential water flow still accounted for 34–65% of total infiltrated water. This study presented the mechanisms that regulate movement of soil water following deforestation through field observations and advanced our understanding of the soil hydrologic process in deep soil.

Land-use change could substantially alter the soil water balance and hydrological cycles; however, little is known on the changes in deep soil water following a cycle of afforestation and deforestation. The purpose of this study was to quantify the soil water deficit in an apple orchard and subsequent replenishment of deep soil water after the orchard was felled. Soil water changes were quantified using the “space-for-time” method through a paired plot design. The results showed that the water storage in deep soil (>3 m in depth) began to decrease when the apple tree reached about 10 years of age. The cumulative deficit of deep soil water storage in the 3–18 m soil depth could reach about 1200 mm; however, deep soil water was so depleted that apple trees can no longer adsorb water from the deep soil when apple trees are older (>22 years old). After the apple orchard was converted to cropland, precipitation replenished the desiccated deep soil to a depth of about 7 m in the first two years, but thereafter, both water recovery amount and the advance rate of the wetting front were slowed down. After 15–16 years of recovery, soil water storage increased by 512–646 mm, accounting for 42.7–53.8% of the total cumulative soil water deficit caused by the apple orchard. However, it will take more than 26 years for soil water to be replenished to the level of the original cropland prior to planting apple trees. The considerable water deficit after afforestation and subsequent long water recovery time following deforestation extend our understanding of the effect of deep-rooted trees on water balance at the decade scale.