openearth / aeolis Goto Github PK

Soil moisture module currenly used from HYDRUS1D should be simple to implement and optimize for top layer moisture content rather than moisture profile.

Support wind direction with a simple cosine rule.
Wind direction should be added to the wind input files.

Ignore grid cells below water surface to increase computational efficiency

Implement BMI interface.
See https://github.com/openearth/bmi for an example

See for example Masselink (2007)
http://onlinelibrary.wiley.com/store/10.1111/j.1365-3091.2006.00825.x/asset/j.1365-3091.2006.00825.x.pdf?v=1&t=i60kfihk&s=42413ac9034d6d2d29fea8ec05c9bc37850a4dd0

Support variable time step.
Mainly important for implicit computations that can adapt to CFL condition.
Interpolation of time series should be done within the step function rather than the initiatilization.

Add input options to disable processes (e.g. bed update) and/or threshold effects (e.g. bed slope).

Currently meteorological conditions are constant in time. Support a time series file for this.
A single file should contain columns with all relevant parameters for the Penman-Monteith equation.
This can also be used for supporting the relative humidity threshold from Arens (1996).

Implement the support for variable grid sizes, 2DH grids and curvilinear grids.

Gathering variable stats (min, max, sum, avg, var) via Python is very time-consuming.
Better to gather the stats in Fortran and only exchange upon output.

... and implement right forumlation

See test case "u" where with constant wind speed, the avg if fluctuating anyway!