Effects of leaf type, litter mass and rainfall characteristics on the interception storage capacity of leaf litter based on process simulation

Forest litter plays an important role in the hydrological process of an ecosystem, but the mechanism of inter-ception is not well understood and has not been fully evaluated. We conducted a process-based experiment under five simulated rainfall intensities. We directly quantified the interception of leaf litter and determined the effects of litter mass (MA), rainfall characteristics (RI) and leaf type (SP) on the interception of six common subtropical tree species, including two coniferous (Cunninghamia lanceolata (Lamb.) Hook. and Pinus massoniana Lamb.) and four broadleaf deciduous (Phoebe bournei (Hemsl.) Yang, Schima superba Gardn. et Champ., Liquidambar for-mosana Hance and Cinnamomum camphora (L.) Presl) species. The results demonstrated the following. (1) The interception process of litter can be divided into three phases: the wetting phase, the saturation phase, and the postrainfall drainage phase. The first phase of which arrived approximately 95% of the maximum interception storage (Cmax) and the duration was longer in broadleaf species than in coniferous species. This phase was significantly different among the species and rainfall characteristics, but it was not affected by litter mass. (2) Cmax and the minimum interception storage capacity (Cmin)of litter were not significantly different between the two forest types, but among the six species, especially between P.bournei and each of the two coniferous species, these values showed the greatest differences. (3) Cmax and Cmin increased significantly with litter mass and rain intensity. (4) Based on commonality analysis, the interaction between MA and RI largely outweighed the other variables in terms of their contribution to Cmax; however, MA surpassed all the other variables in its contribution to Cmin. We also found that the effects of the individual variables on Cmax and Cmin as well as their interactions, were significantly different; the three most influential variables for Cmax ranked in the descending order of their contributions were the interaction between MA and RI (41.07%), RI alone (23.69%), and MA (20.81%), and those for Cmin were MA alone (52.59%), the interaction between MA and RI (25.81%) and leaf types (14.58%). The quantitative and mechanistic explanations regarding the effects of leaf types, rainfall characteristics, litter mass and their interactions on litter Cmax and Cmin presented in this study are expected to increase our under-standing of the mechanism underlying litter interception.