Abstract
To estimate the downslope distance of leaf-litter transport in a given time on the forest floor, we conducted a field experiment with artificial leaves and used the data to construct two models for the transport rate of those leaves. A statistical model with three independent variables was derived from stepwise multiple regression analysis of the field experiment data : the component of wind speed at 1 m height that was perpendicular to the contour line, the forest floor vegetation cover, and slope gradient. An empirical model having the same variables was developed with additional consideration given to the process of leaf movement. The empirical model is regarded as more accurate than the statistical model, because it has a lower Akaike Information Criterion (AIC) value. To derive the transport rates of real leaves, we calculated the ratio of transport rate between the real and artificial leaves for nine species. These ratios ranged from 0 to 6.1. For leaves of deciduous species, the ratio tended to be higher the larger was the area and the lower was the falling velocity. The empirical model and the ratios of the transport rates promise to be useful for predicting leaf transport on a forest floor similar to our research site and for determining the size of the area that contributes leaf-litter to streams.
