Catchmentwide erosion rates with riversand

riversand is a python package to calculate catchmentwide erosion rates from cosmogenic nuclide concentrations in river sand samples. The program computes the hypsometric statistics of the catchment area from a digital elevation model. It uses the online erosion rate calculator by Greg Balco (e.g. http://stoneage.hzdr.de/) to determine predicted nuclide concentrations $N$ for given erosion rates $E$, and calculates the erosion rate that corresponds to the measured nuclide concentration from a polynomial fit $N(E)$.

The method works for in situ Be-10 and Al-26 data. It is fast (few seconds for one catchment) for all production scaling methods implemented in the online calculator (St: Lal 1991/Stone 2000; Lm: Lal/Stone with a geomagnetic correction after Nishiizumi et al. 1989; LSDn: Lifton et al. 2014) and independent of the catchment size or the resolution of the digital elevation model. It is considered robust for catchments up to approx. 600 km x 600 km; for larger catchments the effect of latitude on cosmogenic production may become significant.

The approach is described in:

Stübner, K., Balco, G., and Schmeisser, N. (in review). Calculating catchmentwide erosion rates using an existing online calculator. Radiocarbon.

Definitely check out the documentation of the online calculator (e.g. here or here) and the publication Balco et al. (2008) before using this calculator.

Documentation

• quickstart.ipynb
• step_by_step.ipynb
• test_data/ : geotiffs of a 35m-resolution digital elevation model, a topographic shielding raster generated with TopoToolbox and a binary raster indicating quartz-bearing and quartz-free lithologies; shapefiles with catchment outlines; a spreadsheet with sample data.

Installation

Install riversand by running:

$ pip install riversand

Requirements

• numpy, scipy, pandas, xarray
• rasterio, fiona, pyproj
• matplotlib

License

GNU General Public License v3.0