QUATERNARY STRATIGRAPHY OF THE EEL RIVER AND VAN DUZEN FLUVIAL SYSTEMS: GEOCHRONOLOGY TO SUPPORT TECTONIC INTERPRETATIONS AND SLIP RATE CALCULATIONS FOR LATE PLEISTOCENE ACTIVE FAULTS

The Mendocino triple junction (MTJ) in California has overlapping tectonic regimes; the San Andreas fault (SAF) to the south and Cascadia megathrust to the north. Late Pleistocene northward migration of the MTJ juxtaposes north-south convergence from Pacific-North America (PAC-NAM) plate motion at the northern on-shore termination of the SAF with E-W compression from the Juan de Fuca-North America relative plate motion. Active N-NW striking faults, like those in the N30W striking Mad River fault zone near Arcata, CA, represent deformation related Cascadia convergence. Structures striking ~E-W, like the Table Bluff fault south of Eureka, CA, represent deformation related to PAC-NAM N-S convergence. Slip in this area is transferred from the termination of the SAF and strain is partitioned onto structures to the east. These crustal faults contribute to regional seismic hazard and may comprise one-third of the tectonic strain in coastal northern California.

We locate a topographic scarp adjacent to the Russ fault zone that may represent Holocene slip on a west striking reverse fault offsetting late Pleistocene to Holocene fluvial terraces. We identify this structure as the Lahsāséte fault. The south facing scarp crosses multiple terraces. Scarp heights increase on progressively older terraces. Using regionally derived incision rates as a proxy for terrace age, we use topographic swath profiles to measure scarp heights and calculate a late Pleistocene slip rate of about 0.75 mm/yr.

To better understand the stratigraphic setting and to provide relative age control for the geomorphic surfaces offset by the fault, we conduct a terrace mapping campaign. This chronostratigraphic framework will form the basis for updated slip-rate calculations made for the scarp forming Lahsāséte fault. We use LiDAR derived slope rasters to delineate fluvial terrace treads using maximum slopes up to 10°. We calculate the relative elevation for the treads using a constructed digital elevation model that represents the modern floodplain. Using the distribution of relative elevations for each tread, and the vertical spacing between these treads, we correlate terraces along the lower Eel and lower Van Duzen rivers. Once we acquire additional numerical ages, we will be able to calculate relative ages for the other terraces using incision rates.