Simulated wood budgets in two mountain streams

Abstract

Large wood (LW) recruitment, transport, and storage were evaluated over a century in Gregory and Riley creeks (Haida Gwaii, British Columbia) by modeling a reach-scale LW budget using two frameworks for output: LW loss through decay and downstream transport, and loss through depletion. At reach and at watershed scales, mass movement and bank erosion dominated inputs, and fluvial transport was an important flux term in several reaches. Large wood recruitment by mortality was relatively minor in comparison. Large proportions of the in-channel LW were stored in jams with a mean age of 40–50years. Overall, both modeling approaches yielded reasonable stored LW predictions in the study creeks, with the omission/inclusion of transport responsible for the largest differences between models. Modeled storage generally was within 30% of that measured in the field, and our results illustrate the large temporal variation in storage resulting from episodic inputs of LW from hillslopes.