Response of reach-scale bankfull channel geometry to the altered flow and sediment regime in the lower Yellow River

Abstract

The channel geometry in an alluvial river usually varies in response to the altered flow and sediment regime from upstream damming. Bankfull channel dimensions have significantly adjusted along the lower Yellow River (LYR) because of recent channel degradation caused by the operation of the Xiaolangdi Reservoir, which has led to longitudinal variability in the cross-sectional geometry. Therefore, describing the bankfull channel geometry of the LYR is more representative using reach-averaged variables. In this study, an integrated approach is proposed for calculating reach-scale bankfull channel dimensions based on the measured section-scale bankfull geometry in the LYR, and the proposed approach integrates a geometric mean based on the log-transformation with a weighted average based on the spacing between two consecutive sections. The post-flood reach-scale bankfull channel dimensions in three different channel-pattern reaches of the LYR were then calculated annually during the period from 1999 to 2012 using the proposed approach and the surveyed post-flood profiles at 91 sedimentation sections. Calculated results indicate that the channel geometry changed in the components of bankfull width and depth in the braided reach, with an increase of about 400m in the reach-scale bankfull width over the past 13years; and the channel evolution occurred mainly in the component of bankfull depth in the transitional and meandering reaches, with the increased reach-scale bankfull depths of 2.4 and 1.8m, respectively. Finally, empirical relationships for the three reaches of the LYR were developed between the reach-scale bankfull channel dimensions and the previous four-year average discharge and incoming sediment coefficient during flood seasons, with higher correlations being obtained when calibrated by the measurements from 1999 to 2010. Furthermore, the developed relations were also verified by the observed data in 2011 and 2012, with encouraging results.