NABS Home | What's new? | Search | Contact

  
  email password   Forgot your login information?

About NABS

Membership application

Taxonomic certification

Classified Ads

Students & Postdocs

• Publications

Journal

Bulletin

Membership directory

• NABStracts

2008

2007

2006

2005

2004

2003

• 2002

2001

2000

1999

1998

1997

1997-2008

Bibliography

NABSLinks

Education & Outreach

Annual meeting

Journal (J-NABS)

Society Business

Members only

NABSWeb Admin

 
 

Presented at the NABS Annual meeting, Pittsburgh, Pennsylvania, 2002 in Disturbance Ecology IV

IMPLICATIONS OF TREE FALL DIRECTIONALITY FOR WOOD RECRUITMENT TO MOUNTAIN STREAMS OF THE PACIFIC NORTHWEST, U.S.A.

D.J. Sobota and S.V. Gregory. Department of Fisheries and Wildlife, Oregon State University, Corvallis, Oregon, U.S.A. 97331

Delivery of wood to stream ecosystems is conceptually well defined, but empirical data are scarce. This research investigated tree fall directionality along mountain streams to analyze wood recruitment to channels. Environmental factors that potentially influence fall directionality (i.e., hillslope steepness, geomorphology, and forest dynamics) were also evaluated. We inventoried downed trees in 28 second- to fifth-order stream valleys of the Coast Range, West Cascades, and Intermountain Region of the Pacific Northwest. Simulations of wood recruitment based on fall directionality and riparian forest structural characteristics (produced by OSU STREAMWOOD, an individual-based stochastic model) were compared to observed amounts of in-stream wood. Modes of tree fall were oriented towards the channel (78% of sites), up the stream valley axis (17%), or randomly (5%) (based on P<0.05). Using multiple ANOVAs, we determined fall directions of toppled trees from fluvial landforms were 50% more variable compared to those from hillslopes and overall patterns were 15-20% more variable in stem-exclusion riparian forests compared with older structures (understory reinitiation and old growth). Site-level predictions of average in-stream wood amounts overestimated observed values by 11-27%, yet site measurements fell within confidence intervals of simulated loads.