TESTING POTENTIAL TRIGGERING MECHANISMS OF LONG-RUNOUT CATASTROPHIC ROCK AVALANCHES IN THE NOOKSACK RIVER BASIN, WHATCOM COUNTY, WASHINGTON

The Nooksack River basin in central Whatcom County, WA, preserves a remarkable concentration of long-runout, catastrophic rock avalanche deposits. These deposits present an unusual opportunity to test ideas related to the triggering mechanisms for such landslides. Potentials triggers include large magnitude earthquakes, increases in pore-water pressure from heavy precipitation events, and recent glacial or fluvial debuttressing of hillslopes. We test the plausibility of each potential triggering mechanism by first determining the collapse age of each slide, then comparing those ages to known paleoseismic events (e.g., Goldfinger et al., 2012) to determine whether earthquakes are a dominate triggering mechanism or if other triggers must be invoked. Collapse ages are determined with two independent dating methods: radiocarbon (¹⁴C) and beryllium-10 cosmogenic radionuclide exposure dating (¹⁰Be-CRN). For ¹⁴C dating, 6 sediment cores were collected from bogs formed on top of landslide debris fields. Woody debris from the Racehorse Creek bog core was dated to 3065 ± 26 ¹⁴C yrs. B.P. and provides a minimum age for this deposit. This combined with a maximum age of 3840 ± 140 ¹⁴C yrs. B.P. (Pringle et al., 1998) further constrains the timing of this landslide. ¹⁴C dates from the Middle Fork bog yielded ages of 8855 ± 39 and 8816 ± 41 ¹⁴C yrs. B.P. These two dates define a minimum age for the landslide event and leaves open the possibility of glacial debuttressing as a trigger given the timing of deglaciation for the region (ca. 12-15 kya; Clark and Clague, 2020). For ¹⁰Be-CRN dating (ongoing), 16 rock samples were collected from the top of large (>3 m) boulders associated with the initial collapse of each deposit. Exposure ages of boulders will be used as proxies for landslide emplacement. If these slides were triggered by earthquakes then collapse ages should overlap with paleoseismic events from the Cascadia Subduction Zone (Goldfinger et al., 2012) or local crustal faults, such as the Boulder Creek Fault (Sherrod et al., 2013). Conversely, other triggering mechanisms should show different temporal distributions: glacial debuttressing should cluster immediately following deglaciation whereas precipitation triggering or fluvial debuttressing should display a more stochastic temporal distribution.