LATE HOLOCENE PALEOFLOODS ON THE MIDDLE FORK WILLAMETTE RIVER: IMPROVING HYDROLOGIC LOADING INPUT FOR USACE DAM SAFETY EVALUATIONS

USACE dam safety assessments consider potential hazards (loading), system responses (fragility), and associated consequences, as part of the risk assessment of its entire dam portfolio. The country-wide dam portfolio covers a vast range of hydrologic and geomorphic settings, many of which are suited for paleoflood data development. Where possible, paleoflood data are incorporated in dam safety evaluations to reduce uncertainties in the hydrologic loading component of the risk assessment.

An initial paleoflood analysis along the bedrock-floored Middle Fork Willamette River (MFWR) in central Oregon identified multiple discontinuous terrace remnants that were used as paleo-stage indicators (PSI) and a non-exceedance bound (NEB). Field geomorphic mapping using high-resolution topographic data shows the presence of paleoflood deposits at elevations higher than the historic flood of record (est. 87,000 cfs, in 1861); age estimates from relative soil development, radiocarbon, and OSL analyses suggest deposition (of terrace “Qt3”) 200 to 500 years ago. Similarly, a higher, abandoned NEB terrace is at an elevation near the stage of the calculated Probable Maximum Flood (PMF; 270,000 cfs). Soil stratigraphy and numerical age estimates indicate that this terrace (“Qt2”) has not been inundated since deposition 2,200 to 2,400 years ago.

Paleodischarge magnitudes are estimated via HEC-RAS 2D hydraulic modeling using high-resolution topographic data, and include multiple sensitivity runs to capture a range in hydraulic roughness. The resistant bedrock floor of the MFWR valley allows an assumption of channel stationarity over the late Holocene time scale. The paleodischarge magnitudes and ages will be incorporated into inflow frequency analyses using USGS Bulletin 17C procedures, and will be used to assess the probabilities and durations of rare reservoir-pool elevations for dam safety risk assessment. We are encouraged by available information to constrain discharge magnitudes and frequencies, and thus reduce uncertainties in rare hydrologic loading estimates.