Catchment drainage network scaling laws found experimentally in overland flow morphologies

Publication:

Cheraghi, M., A. Rinaldo, G. Sander, P. Perona, and D.A. Barry. (2018), Catchment drainage network scaling laws found experimentally in overland flow morphologies, Geophys. Res. Lett., 45. https://doi.org/10.1029/2018GL078351

Data organization:

Surface elevation at different times for two different experiments: I) Average rainfall = 85 mm/h and 5% slope and II) verage rainfall = 60 mm/h and 10% slope. The details of the experimental setup are in the manuscript and its Supporting Information.

The morphology data are named according to: XYZ (SLOPE) % (Average Rainfall) mmh_ (time in hour) h_4mm.dat

For example, "XYZ5%85mmh_1.00h_4mm.dat" is the surface morphology at 5% slope, 85 mm h^-1 rainfall intensity and after 1 hour of precipitation.

The rainfall profile file (Rain4mm.dat) contains x(mm),y(mm) and the normalized rainfall distribution (with respect to the mean). To use the data, they have to be multiplied by either 85 or 60 mm h^-1 to give the corresponding rainfall distribution of the experiments above.

_____________________________________________________________________________________________________________

Experimental setup:

We used a  2-m \(\times\) 1-m erosion flume. The total sand depth was 15 cm. Uncohesive sediments with a mean diameter of 0.53 mm were placed in three successive layers of 5 cm within the flume. Each layer was compacted by 30-min rainfall (droplet size of 3-7 mm) with an intensity of 10 mm h^-1 followed by consolidation via a 600 kg m^-2 weight. Then, the sediment was air dried for 48 h. Heterogeneous rainfall was generated by two sprinklers located 3 m above the sediment surface. The average rainfall was 85 mm h^-1 with a Christiansen uniformity coefficient of 26 %. The rainfall was applied continuously except a 30-min break for each laser scan.
There was no drainage from the flume bottom and all surface flow was collected at a single, 4-cm wide outlet, located at 6 cm above the base of the flume. The experiment started with a smooth surface. At the outlet, the initial elevation difference between the sediment surface and the base level was 9 cm.  At the first 5 minutes, the sediments were being saturated and the flow rate was very low. Afterwards, there was a rapid elevation drop at the outlet until about t = 10 minutes. In this period, it was not possible to capture the flow rate due to very high concentration of sediments at the outlet. The scanning was started after 15 minutes when the flow rate became steady, the sediments were saturated and the whole precipitation flowed on the surface. A 3D laser scanner (Konica Minolta Vivid 910), with about 4 mm horizontal resolution and accuracy of 0.1 mm in the vertical direction, was used to extract Digital Elevation Models (DEMs) at 0.25, 0.5, 1, 2, 4, 8 and 16 h. The scanner was calibrated using 20 fixed points on two bars along the flume's lateral walls. Eight individual scans were taken to cover the entire flume. Following registration and post-processing, each DEM produced was trimmed 2 cm from the side walls and 10 cm from the upstream wall.

In order to investigate the scaling laws at different slope and rainfall intensity, another experiment was carried out at 10 % slope and average rainfall of 60 mm h^-1 for a duration of 20 h.