Python 3 stochastic Galerkin shallow flow model with well-balancing, wetting-and-drying, and uncertain friction and topography

This Python 3 code includes

1. a one-dimensional shallow flow model which can be configured as an intrusive or nonintrusive stochastic model;
2. preconfigured steady-state flow, lake-at-rest, and dam break test cases.

First ensure that Python 3, NumPy and SciPy are installed.  Then, install the stochastic Galerkin library and the shallow flow models:

pip3 install --user --editable pysg
pip3 install --user --editable pysgswe

To run the intrusive model:

pysgswe-intrusive <outputDir> <testCase> <basis> --max-basis <integer>

where testCase is either criticalSteadyState, lakeAtRest, flumeFriction or flumeFrictionTopography,# basis is either haar or hermite, and the --max-basis value is the maximum refinement level L or Hermite polynomial order P, according to the choice of basis. For test cases involving two sources of uncertainty, the flag --basis-dimensions 2 must also be included. A tensor product basis is used unless --truncate-basis is specified. Running the nonintrusive model follows a similar procedure:

pysgswe-nonintrusive <outputDir> <testCase> --max-level <integer>

where --max-level specifies the maximum refinement level. Note that the nonintrusive model can only be used with a Wiener–Haar basis. Other command line options are the same as for the intrusive model.