River training works performed in the last couple of centuries constrained the natural dynamics of channel networks in locations that include the confluences between tributaries and main channels. As a result, the dynamics of these confluences are currently characterized by homogeneous flow depths, flow velocities, and morphologic conditions, which are associated with impoverished ecosystems. The widening of river reaches is seen as a useful measure for river restoration, as it enhances the heterogeneity in flow depths, flow velocities, sediment transport, and bed substrates. The purpose of this study is to analyze the effects of local widening of the tributary mouth as well as the effects of the ratio between the width of the tributary and that of the main channel on the flow dynamics and bed morphology of river confluences. For that purpose, 12 experiments were conducted in a 70° laboratory confluence. In these experiments, three unit-discharge ratios were tested (qr=0.37, 0.50, and 0.77) with two width ratios and two tributary configurations. The unit-discharge ratio is defined as the unit discharge in the tributary divided by that of the main channel, measured upstream of the confluence. The width ratio, which is defined as the width of the tributary divided by that of the main channel, was modified by changing the width of the main channel from 0.50 to 1.00 m (corresponding to Br=0.30 and 0.15 respectively). The tributary configurations consisted of (i) a straight reach with a constant width (the so-called reference configuration) and (ii) a straight reach with a local widening at the downstream end (the so-called widened configuration). During the experiments, a uniform sediment mixture was continuously supplied to both channels. This experimental setup is novel among existing experimental studies on confluence dynamics, as it addresses new confluence configurations and includes a continuous sediment supply to both channels. The experiments were run until the outgoing sediment rate was nearly the same as the incoming; i.e., equilibrium had been achieved. The bed topography and water surface were then recorded in both channels. The results reveal that the width ratio and the locally widened tributary reach influence the dynamics of the confluence. The different width ratios influenced the size of the bank-attached bar at equilibrium, which was wider and longer for Br=0.15 than for Br=0.30. Other morphological differences were observed at equilibrium for the different width ratios, such as deeper scour holes and increased penetration of the tributary into the main channel. These differences were attributed to the different values of the ratio between the unit momentum-flux of the tributary and that of the main channel that were noted at equilibrium for the different width ratios. The local widening of the downstream reach of the tributary significantly enhanced the heterogeneity in flow depth, flow velocity, and bed morphology within the widened reach. This heterogeneity contrasts with the homogeneity observed in the tributary without widening (reference configuration). Additionally, the effects of the local tributary widening were limited to the tributary, with minor or negligible effects on the main channel.