Deep‐Seated Landslides Drive Variability in Valley Width and Increase Connectivity of Salmon Habitat in the Oregon Coast Range

Abstract

Declines in populations of Pacific salmon have prompted extensive and costly restoration efforts, yet many populations are still in peril. An improved understanding of landscape‐scale controls on salmon habitat should help focus restoration resources on areas with the greatest potential to host productive habitat. We investigate the contribution of deep‐seated landslides (DSLs) to Coho Salmon habitat by comparing the quantity and connectivity of potential seasonal habitat observed in five streams with extensive DSLs to five lacking significant landsliding. Further, we measure valley width in these streams and relate it to connectivity. We show that median fractions of stream length identified as spawning, summer‐rearing, winter‐refuge habitat, and as having high connectivity among seasonal habitat types are greater in streams with DSLs and that distances between units of each seasonal habitat type are significantly lower in DSL terrain. The median R2 value for the relationship between drainage area and valley width is lower in landslide terrain and we observed that high connectivity among seasonal habitat types tends to occur where valley width is variable. Our results suggest that DSLs promote connectivity among seasonal habitat types for Coho Salmon and that prioritizing restoration projects in streams in DSL terrain could improve the effectiveness of salmonid recovery programs.

The influence of lithology on channel geometry and bed sediment organization in mountainous hillslope‐coupled streams

“…In the sandstone basin the slope of the drainage area-valley width relation was 2 times larger than the slope of the drainage area-channel width relation (Table 1). The drainage area-valley width relations were similar to those derived from LiDAR for sandstone basins in the Oregon Coast Range (Beeson et al, 2018;May et al, 2013).…”

“…channel width through long-term sediment supply (Beeson et al, 2018;May et al, 2013;Roering et al, 2005).…”

“…Lithology may additionally influence hillslope-fluvial coupling through contrasting hillslope evolution and mass wasting processes (Beeson et al, 2018;Johnstone & Hilley, 2015;Marston et al, 1997;May et al, 2013;Miller & Burnett, 2008;Roering et al, 2005). This in turn may influence the presence of alluvial or non-alluvial conditions and contrasting longitudinal adjustments of channel geometry and grain sizes through differential hillslope sediment inputs.…”

“…For example, an investigation of debris torrents in the Oregon Coast Range indicated that they were twice as frequent in basins draining rocks with low friability (such as basalt) compared to basins draining rocks with high friability (such as sandstone) (Marston et al, 1997). In addition, deep-seated landslides often found in streams draining softer lithology (sandstone) appear to be associated with wider streams as a result of long-term sediment inputs in these regions (Beeson et al, 2018;May et al, 2013;Roering et al, 2005).…”

The Preservation of Climate‐Driven Landslide Dams in Western Oregon

“…Furthermore, landslides that impound valleys and form lakes and/or become sediment‐filled serve as a secondary drivers of landscape evolution, potentially over long timescales (>10 4 year; e.g., Korup, 2004; Korup et al., 2006; Ouimet et al., 2007). Landslide dams locally store immense quantities of sediment that cover and armor bedrock channels (e.g., Hewitt, 1998; Korup et al., 2004; Lancaster & Grant, 2006), and upstream low‐gradient valley reaches foster ecological diversity and sequester organic carbon (Beeson et al., 2018; Bilby & Likens, 1980; Mackey et al., 2011; May et al., 2013). As a cascading hazard, when landslide dams become unstable, they often release large outburst floods (Costa & Schuster, 1988; Fan et al., 2019; Shang et al., 2003), making characterization of the timing and persistence of landslide dams critical for hazard mitigation.…”