Post-Fire Erosional and Hydrological Processes Promoting Debris Flow Initiation in a Douglas Fir and Western Hemlock Forest in the Riverside Burn Area, Oregon

Abstract

Post-fire debris flows initiated by overland flow in the Pacific Northwest (PNW) are largely undocumented. Instead, debris flows are typically initiated by shallow landslides that result in a mud slurry of water and sediments traveling downhill under the force of gravity. However, because of the Fall 2020 fires in Oregon, the typical initiation style and erosional patterns in burned catchments may have changed because of unusually high burn severity. Due to the intensity of these fires, we set out to determine how hydrologic processes and erosion occurred, when they occurred, and what process was primarily responsible for the erosion that ensued. To investigate this, we studied the South Fork Clackamas River and Memaloose Creek catchments within the Riverside Fire perimeter. Here, we collected hydrologic measurements and consecutive lidar scans with a terrestrial laser scanner. The terrestrial lidar scans were pre- and post-PNW wet season as to capture the post-fire surface changes due to rainfall and runoff. Additionally, we administered infiltration tests and runoff monitoring. From these data and collected precipitation, we developed a runoff response model to delineate the timing of probable erosion during a rain event and to define the relationships between different runoff pathways: Hortonian overland flow, saturation excess overland flow, and baseflow. From our lidar scans, we found erosion was patchy throughout our sites without any rilling or indicators of debris flows from November 2020 to March 2022. This erosion, which averaged 2.13 mm, was likely due to rainsplash and Hortonian overland flow on December 20th-21st, 2020, and January 12th-13th, 2021, as depicted in our runoff response model. According to volume estimates, it's likely the greatest amount of erosion occurred on the January 12th-13th event. Specifically, this erosion occurred in the central, planar to convergent parts of our catchments with sediment being captured in the rocky borders. The lowest averaged saturated hydraulic conductivity was 61.5 (± 33.8) mm/hr. Therefore, for a debris flow to be initiated by runoff in glacial tills in the Oregon Western Cascades, we can conclude that saturated hydraulic conductivity will need to be less than 61.5 mm/hr and peak rainfall intensity will need to be greater than 85 mm/hr.