UC Davis

Previous estimations of sediment flux for the Salinas River of central California were based on data collected in the 1970s and assumptions of time invariant suspended sediment-discharge behavior. The goals of this study were to estimate sediment flux from the Salinas River using data from 1967-2011 by incorporating time dependent behavior and reassess the role of El Niño Southern Oscillation patterns in inter-decadal sediment load. This study builds on previous findings that time-dependent suspended sediment behavior in this system is controlled in part by antecedent hydrologic conditions. The condition of temporal dependence was further tested herein through comparison of flux estimates obtained using time-dependent formulations and a multivariate approach incorporating hydrologic factors. Longer sampling records and incorporation of decadal scale behavior or antecedent hydrologic conditions resulted in average annual load estimates of 2.0-2.9. Mt/yr with 95% confidence intervals of ±25 to 202%, in comparison to earlier estimates of ~3.3. Mt/yr. Previous overestimation of sediment load is due largely to the extrapolation of suspended sediment behavior from a decade of high sediment concentrations to the entire record, and the use of log-linear regression techniques on a non-linear system. The use of LOESS methods lowered QSS estimates and decreased confidence interval size. The inclusion of time-stratified and antecedent flow indices further decreased QSS estimates, but increased confidence interval size. However, temporal dependence of the CSS- Q relationship violates the assumptions of single base period regression, which suggests that time-stratified rating curves provide more realistic estimates of sediment flux means and uncertainty. The majority of suspended sediment was transported by flows of ~25-90 times mean discharge depending on transport constituent (fines or sand) and estimation method. Periods of differential suspended sediment behavior changed the relative importance of rare floods due to changes in the relationship of suspended sediment concentration vs. discharge. El Niño years dominated the sediment budget by producing on average ten times more sediment than non-El Niño years. Sediment load estimates provided further evidence that antecedent hydrologic conditions appear to have caused much of the temporal dependence of suspended sediment behavior.