Bedform Experiments Conducted at the University of Texas at Austin, 2018

This is the file-processing scripts required to generate meaningful figures for topography, grain size, and derived, modeled values. How to use these scriptsThere aren't any particularly unusual packages used in this analysis, but the key requirements are:numpyscipypandasmatplotlibseabornjupyteripympl The working environment is in the environment.yml file. The parent utility file is in tools/tools.py and should be accessible through relative imports.The analysis of the data in the data/ directory is all generated through the Jupyter Notebook. Loading the data In notebooks/Figure_plotter.ipynb, the utilities and plotters are imported and all relevant figures are accessable.If you prefer spyder (my own preference), the file tools.py can be executed in the IDE and all relevant data and figures are renderable and become more or less interactive, insofar as matplotlib allows for that. One note, orthogonal projections of the 3D scans does not look good in matplotlib, and there appears to be no way to effectively render a 45 degree view of the scan time series in an attractive way. I've preserved the original matlab figure, and have a script that plots the scans from the overhead view. Experimental Description Experiments were conducted in the Experimental Sedimentology Lab at the University of Texas at Austin, using an 11 m long recirculating flume. The flume is 0.6 m wide and 1.2 m deep. Using a flow diffuser at the upstream end, logarithmic flow-velocity profiles were generated, measured by a Nortek Aquadopp Acoustic Doppler Profiler. One meter from the flow diffuser, a 2 m long by 50 mm thick bed of sand was smoothed to cover the base of the flume. The sand mixture included 90 kg of medium quartz sand, D50 = 281 μm, ρqtz = 2616 kg/m3 and 1 kg of magnetic sediment, D50 = 330 μm, ρmag = 4990 kg/m3. This tracer concentration was chosen as a tractable, measurable concentration within bulk sediment The mean vertically-averaged centerline flow velocities associated with the four transport conditions were 0.155 m/s (Experiment 5), 0.159 m/s (Experiments 4 and 6), 0.161 m/s (Experiments 3 and 7), and 0.188 m/s (Experiments 2 and 8). Despite a modest range in mean flow velocities between the experiments, we were able to generate a range of transport rates for the median quartz sediment, and theoretical transport conditions (based on the derived Rouse numbers, and retained volume fraction) because our experiments operated at thresholds for different types of transport. Our experiments were paused periodically throughout their duration to take elevation scans of the deposits. The experiments were run until the bedform translated one full wavelength past its initial deposition location. This does not ensure that complete steady-state transport filtering occurred, but by enforcing the rule of every grain in the domain necessarily experiencing at least one transport event we have some assurance that aggregate transportability is being modeled by the experiment.