Hydrologic changes due to the imposition of dams can affect the composition and distribution of plant communities in floodplain riparian areas. In the Lower Columbia River Basin in the Pacific Northwest of North America, hydrologic regime shifts due to upstream dams have in part resulted in the abundance of invasive Phalaris arundinacea (reed canarygrass) in riparian and palustrine plant communities throughout the floodplain. We analyzed the effect of using a water control structure to mimic historical snow-melt driven spring-summer high-water levels to better match up with the growth phenology of native riparian plants in an urban wetland in the Columbian River floodplain in Portland, Oregon, USA. Vegetation monitoring was carried out in three phases since project initiation (2003-2004, 2008-2009, and 2015-2016) to assess restoration efforts. Using line-intercept and differential leveling methods, we measured 25 randomly established transects perpendicular to shoreline, ranging from 21.5m to 280.7m (mean: 92.9m), during monitoring years for vegetation and elevation to determine changes in vegetation in relation to seasonally varying water levels.
Results/Conclusions
Over this 14-year period in the 809 ha Smith and Bybee Wetlands Natural Area, hydrologic management affected a dramatic decrease in invasive Phalaris arundinacea (reed canarygrass) cover from 46.5% to 17.6% (McNemar’s test, p<0.05). Native Persicaria amphibia (smartweed) replaced reed canarygrass as the dominant species, increasing in cover from 20.2% in 2003 to 67.9% in 2016, and replacing all other common species except for Salix lucida lasiandra (p<0.05). Other common native species (Bidens cernua, Eragrostis hypnoides, Eleocharis palustris, and Cyperus spp.) experienced declines in cover between 2003 and 2009; after further altering the hydrologic regime, these species have since increased in cover in low transect elevations due to lower inundation rates during their early growing seasons. Species diversity declined significantly from H’ = 2.44 in 2003 to H’ = 2.10 in 2016 (Hutcheson p<0.05). Diversity was negatively correlated with reed canarygrass presence in 2003 (Spearman’s rho = -0.79; p<0.05) and Persicaria presence in 2016 (Spearman’s rho = -0.70; p<0.05). These findings demonstrate that hydrologic management, based on phenological responses of the plant species and on recreating historic hydrological conditions, can be effective at reducing the presence of invasive plant species and in promoting native wetland vegetation in floodplain wetland ecosystems.