Instream wood transport, and effects of forest harvest on geomorphology and fish, in northern Minnesota streams.

Abstract

Trees provide critical functions to the ecology of streams. Trees affect hydrology, mitigate sediment inputs, and buffer water temperatures by providing shade. Watersheds with a higher proportion of mature forest tend to have less variable hydrographs, and older trees provide larger pieces of wood to streams. Instream wood itself affects nearly every process in stream ecology.

Despite their ecological importance, trees and instream wood have been greatly modified by humans. The studies described in this dissertation advance current knowledge as follows: We demonstrate that headwater streams in northern forests can require ten years to recover from a large input of fine sediment, depending on the occurrence of stormflows.

Our analyses suggest that, at the basin scale, warmer air temperatures in summer are more important to the abundances of some headwater fish species than instream habitat or spring precipitation. The analyses also lend support to previous findings that riparian forest harvest can cause local stream warming.

Wood transport in streams is a dynamic process. Forty-one percent of over 800 wood pieces were mobilized (at least 10m) during a study period by a single high flow event. Thirty-two percent of the mobilized pieces became entrapped again before leaving their study reach.

Mobilization of wood in streams is a complex function of both mechanical and hydraulic factors. Eleven potential predictor variables were studied, and seven were identified as significant to wood mobilization using multiple logistic regression. The seven predictors were burial, effective depth, length ratio, bracing, rootwad presence, downstream force ratio, and draft ratio.

Entrapment of wood in streams is related primarily to the length ratio and weight of the wood pieces. The mechanisms for entrapment are not always clear; wood pieces may simply be entrapped wherever they are located when high water recedes.

Together, this dissertation suggests that forest harvest should avoid excess sediment inputs (due to persistence) and stream warming (due to effects on fish). It also develops models that can be used for more informed management of instream wood. Stream managers and restorers can apply the results presented to reverse the impacts of historic logging and wood removal on streams.