|Dec 30, 2013|
|6 years 1 month ||This is now a hopefully well tested commit that finalizes the
introduction of a SNOW_EROSION switch acting only on the main
station and SNOW_REDISTRIBUTION acting on virtual slopes only.
Unless NUMBER_SLOPES is set to 1 and SNOW_EROSION is true, the
|Dec 21, 2013|
|6 years 2 months ||A SOS Xmas commit (will be described later by H.C.G. Fierz)
|Dec 20, 2013|
|6 years 2 months ||Switching to DGESVD solver when DGTSV fails is not always preferred,
as DGESVD can be so slow, that a smaller time step and trying DGTSV
again works faster. This is especially the case in simulations with
many layers. Now this can be controlled with
|Dec 19, 2013|
|6 years 2 months ||Implementation of the liquid water index for virtual north and south
slopes (dumped to SDBO only). Surface hoar at the surface (nodal
property) also shows up in dumped profiles (PRF and IMIS). Rename
new_snow_grain_rad and new_snow_bond_rad to *_size
|Dec 16, 2013|
|6 years 2 months ||Implement Edgar Schmucki's new albedo parameterization (SCHMUCKI,
version X32 is set) thereby moving element property rg_opt to ogs,
that is twice the former value ;-) New keys SNOW_ALBEDO,
Renamed quite a n
|Dec 15, 2013|
|6 years 2 months ||Disentangle HN_DENSITY and HN_DENSITY_MODEL to HN_DENSITY,
HN_DENSITY_PARAMETERIZATION and HN_DENSITY_FIXEDVALUE
|Dec 11, 2013|
|6 years 2 months ||New approach for determining the residual water content when solving
water transport in snow using Richards equation. Just taking
theta_r=0 was causing a too big error in wet snow covers, so
something different was needed.
|6 years 2 months ||When CHANGE_BC=false, there is no need to make THRESH_CHANGE_BC
|Dec 10, 2013|
|6 years 2 months ||The Calonne (2012) parameterization of saturated hydraulic
conductivity provides a better agreement with measured runoff at CDP
and WFJ, so I make it the default. Furthermore, it is better to use
the rg_opt for Calonne, instead of rg, as the first on
|Dec 9, 2013|
|6 years 2 months ||Introducing the Calonne (2012) parameterization for hydraulic
conductivity. So now, one can choose between Shimizu (1970) and
|6 years 2 months ||The van Genuchten parameters used for Richards equation needed
adequate referencing, so a revision was necessary, as I could not
find the original source. I now use the ROSETTA data for the soil
types, which is widely used.
|6 years 2 months ||Making the air entry pressure variable. It depends on a maximum pore
size, which can be specified now differently from the standard 5mm
for each soil layer or the snowpack.
|Dec 4, 2013|
|6 years 2 months ||A follow up on my commit in rev 479: a better approach turned out to
be to only limit settling in case we use richards equation for snow
in an alpine3d simulation. All other cases seem to work fine with
normal settling treatment.
|Nov 27, 2013|
|6 years 2 months ||Introducing a finer control of boundary conditions for solving
Richards equation, depending on whether snow is present or not.
|6 years 2 months ||Setting the predefined soil properties to more realistic values and
update the CDP-soil.
|Nov 25, 2013|
|6 years 2 months ||Make the SOIL_FLUX configuration key optional (set to FALSE by
default) in order to be consistent with inishell
|Nov 20, 2013|
|6 years 3 months ||Forgot a ";"...
|6 years 3 months ||Adding a soil definition for Col de Porte (CDP) in Richards equation
|Nov 17, 2013|
|6 years 3 months ||Closing issue 304. Amount of settling is now limited to theta[AIR],
while leaving also some space in case all liquid water would
refreeze. This makes the loop (hopefully) obsolete.
An additional tuning factor was added to ensure stability in the Rich
|6 years 3 months ||A small correction on Rev 469: when there is no snow, but we have
soil layers, AsciiIO should write out the soil layers, and not close
the file directly.