River Patterns

Experimental & Numerical modelled rivers (2009-2013)

For my PhD-research, I aimed to understand how meandering rivers are formed. To obtain insights in the necessary and sufficient conditions for meandering, I conducted several experiments. The first experiments started with a basic set-up and were used to test scaling issues (see Earth-Science Review article). Then experiments were conducted with different initial conditions. I tested how the meanders were formed, while Wietse van de Lageweg focussed his PhD on the sedimentary architecture of these systems.

1. Constant discharge (info HERE)

2. Addition cohesive silt fraction in the feed (info HERE)

3. Upstream fixed perturbation

4. Continuous upstream perturbation (info HERE)

5. Varying discharge (info HERE)

6. Addition of vegetation (info HERE)

Meander bend with cutoffs during an experiment with constant discharge.
WMvanDijk2012_experimentalmeandering.pdf
VanDijketal2013a_FloodplainFormationMeanderingRiver.pdf
WMvanDijk2013_vegetationdistribution.pdf
VanDijketal2014_BifurcationsChuteCutoffsMeanderingRivers.pdf
Kleinhansetal2014_ScalingExperimentalSettings.pdf
Kleinhansetal2015_ScalingExperimentalVegetation.pdf
WIvandeLagewegetal2012_ChannelBeltArchitectureMeanderingRiver.pdf
VandeLagewegetal2015_MeanderPreservation.pdf
VandeLagewegetal2014_ScrollBarFormation.pdf

For additional figures and images of the experiments you can find them meander experiments, vegetation experiments, river Allier and PhD thesis.

Next to the experiments we tested our hypothesis with a sophisticated numerical Delft3D. I used this model to understand the behaviour of chute cutoffs, which mainly occurred in our experimental set-up. Therefore, I looked at natural examples of the river Allier in France (see more here). I found that chute cutoffs are instable bifurcation that form mostly on the outer side of a bend. The success of a chute cutoff depends on the migration of the bend upstream, several small chute cutoffs were not successful as the bend migrated downstream and the chute channel became part of the inner bend.

The experiments were compared with numerical model outcomes of Delft3D. Filip Schuurman showed how braided bars developed in Delft3D and compared to the field. He showed how sensitive the model is for several boundary conditions in his first article.

Here is an video on our flume experiment. This was for a Dutch television program (RTV Utrecht) and shows to the viewer what kind of research is going on at Utrecht University. Maarten Kleinhans shows here an experiment with vegetation (alfalfa) and an experiment without vegetation.

Locations of some natural rivers

List of rivers and publications, from

Kleinhans, M. G. and van den Berg, J. H. (2011), River channel and bar patterns explained and predicted by an empirical and a physics-based method. Earth Surf. Process. Landforms, 36: 721–738. doi: 10.1002/esp.2090

Braided rivers

Braided rivers are characterized by a multiple bars, mainly mid-channel. There is one dominant channel, but the system remains multi-thread. Here an example is given of a braid compound bar in the Brahmaputra, with chute channels on the top of the bar. In the maps several locations are given of braided rivers, with a variety of braid bars.

Meandering rivers with chutes

Meandering rivers with chutes are characterized by constant cutoffs of the meanders without that the neck of the bend reaches eachother. The point bar is cutoff by water that flows over the bank and takes a shortcut. An example which we visit more often is the river Allier in France. On the image the bend was cutoff, but later the bend would redevelop.

Meandering rivers with scroll bars

Typical in meandering rivers with scrolls are the clear lateral migration of the channel. Meander growth resulted in lateral displacements and eventually river meander bends are connected by there neck, forming neck cutoffs. An example of these nicely formed scroll bars are found in Alaska, were rivers develop in the permafrost.

No-bars sinuous channels

These rivers have a sinuous channel, but over time the channel does not migrate and does not form chute bars nor scroll bars. These rivers seem to be meandering, but we would not describe them as meandering. Here is an example of the Mississippi, which was dynamic, and shows some old channels, but these old channels were not nicely filled and the lateral displacement in the greenish forest is insignificant to form bars.