Riparian cover changes associated with flow regulation and bank stabilization along the Upper Willamette River in Oregon between 1939 and 1996

Abstract

This study investigates riparian cover change between 1939 and 1996 utilizing aerial photography, bank conditions and stream gauge data. Three sites under 1.0 km2, along a 10- km reach north of Harrisburg, Oregon, were evaluated. The cover classification scheme included 18 classes composed of 3 canopy densities and 8 structural categories: water, bare gravel bars, herbaceous, shrub, small trees, medium trees, closed trees, and agricultural fields. Net cover changes and shifts between classes were analyzed for each site as a whole and as terrestrial vs. fluvial changes. Terrestrial changes (i.e. areas unaffected by channel movement) were discriminated from fluvial changes (i.e. areas where a channel migrated out, in or both) by inspection of 14 air photo coverages between 1939 and 1996. Over 75% of each site experienced net cover change between categories of water, bare, cleared land, low vegetation, open-canopy trees, and closed-canopy trees. Water cover composed over half the area that remained the same. Net progressive structural change (i.e. an increase in cover maturity, e.g. bare changing to trees) exceeded or equaled net regressive structural change at each site. Progressive structural change was only slightly favored in fluvial areas, while terrestrial areas favored progression 4 to 1 at each site. Bare ground decreased by 35%, 65% and 90%, while forest cover increased by 10%, 50% and 150% at the three study sites. The only areas of non-forest cover remaining in 1996 existed on land disturbed by the channel since 1939. The maturation of structural cover was associated with decreases in channel dynamics and fluvial disturbance between 1939 and 1996. Channel migration affected 50% to 70% of each study area during this period, but most shifts occurred before 1956. Sixtyeight percent of the revetments present in 1996 were established by 1956. These revetments hardened portions of all outer bends along the study reach, limiting lateral channel migration. Upstream dams held 42% of the 1996 storage by 1956, which reduced peak flows.