Abstract:
The natural fall of trees in riparian areas is an important source of coarse woody debris for mountain streams, improving fish habitat and influencing stream morphology. Existing models consider the probability of coarse woody debris entering a stream channel based upon trees having a random direction of fall without consideration of tree lean or wind direction. This research presents (1) the results of a field study to document tree lean of conifers near streams in two stands in the Oregon's Coast Range and (2) a physical and probabilistic model to estimate the probability of a tree falling into the stream including the effects of tree lean and
wind direction. The measurement of 200 conifers along two creeks located in McDonald Forest found that tree lean varied from 1 to 34 percent uphill and 1 to 29 percent downhill on slopes of 1 to 88 percent. Approximately 75 percent of the trees leaned
downhill and 25 percent of the trees leaned uphill. A significant linear relationship was found between lean and slope although there was considerable scatter around the
regression line. In general, the steeper the slope, the greater the tree lean downhill. When tree lean data was stratified by aspect, the linear relationship was higher for the NE and SW aspects, slightly weaker for the NW aspect and not related with slope for the SE aspect. A physical model was developed for calculating the critical wind speed required to overturn a tree. This critical wind speed is a function of maximum resisting moment of the tree root structure, crown cross sectional area, initial tree lean and the angle formed between wind direction and lean
direction on the horizontal plane. A probabilistic model was developed for determining the probability that a tree could fall and reach the stream. This probability is a function of exceedance probability for a
particular period of time, wind direction probability, tree location and tree height.
The models were applied to two old-growth coniferous stands. Douglas-fir [Pseudotsuga menziesii (Mirb.) Franco] trees were selected to illustrate how the models can be used.
Results of this study indicate that tree lean is not a major factor with respect to influencing tree blowdown for the range of tree lean data collected from coniferous trees along streams in the study area. Tree lean could be a major factor if it was greater than that observed in this study.
