Variability of surface runoff generation and infiltration rate under a tree canopy: indoor rainfall experiment using Japanese cypress ( Chamaecyparis obtusa )

To estimate the variability of surface runoff generation and the infiltration rate on a bare surface in a forested area, indoor experiments were conducted using 13 runoff boxes and a single transplanted Japanese cypress tree (9·8 m in height) in a large‐scale rainfall simulator with spray nozzles (at a height of 16 m). The surface runoff was measured for applied rainfall and for 12 kinds of throughfall with different intensities and kinetic energy (KE) (found among measuring points and canopy structures). While no surface runoff was observed for the applied rainfall, surface runoff was observed for throughfall in each runoff box. Compared with the applied rainfall, the throughfall had larger drops due to canopy drip generation and thus had higher kinetic energy, which decreased the infiltration capacity. The maximum stable infiltration rate (IR MAX ) was lowest for throughfall (44·2 mm h −1 ). Surface runoff generation and infiltration rates varied greatly under the canopy, even though the rainfall applications were identical and the runoff boxes had identical initial soil properties. The variability of IR MAX , ranging from 44·2 to 120·2 mm h −1 , was caused by the variability of the throughfall intensity and kinetic energy. The index showing the best correlation to IR MAX was the effective unit kinetic energy (KE 0 mm : J m −2 mm −1 ). The prediction of surface runoff generation in a forested area requires estimations of the spatial variations of the amount and kinetic energy of throughfall. Copyright © 2010 John Wiley & Sons, Ltd.