Should we take into account distance from center of time zone?

[arbennett]

In the original code the distance from the center of the time zone is calculated by

hour_offset = (theta_l - theta_s) * 24 / 360

and is used in the disaggregation of shortwave radiation:

tiny_step = j*tinystepperhour + k - tiny_offset

Where tiny_offset is just a scaling of hour_offset.

The theta_l and theta_s parameters come from a struct returned by read_soilparam.c, which doesn't have any implementation in MetSim right now.

[arbennett]

@bartnijssen @jhamman Do either of you have thoughts on whether/how we should implement this? Perhaps these parameters can be set in the configuration file?

[jhamman]

Yes, I do think we should maintain this feature. For now, I'm fine with specifying this as a configuration file option.

[arbennett]

Would it be reasonable to pull the lats and lons from the forcing file name? Or rather, is it to onerous to enforce the forcing naming standard $NAME_$LAT_$LON? If we enforce this, the theta parameters could be calculated from these directly.

[jhamman]

I don't really like using the filename as metadata. For the traidtional VIC input, this may be okay but the most common application of this tool will probably be netcdf->netcdf.

[bartnijssen]

We should not pull this from the forcing file name. That's an old VIC habit in classic mode, but we should not assume that going forward. For example, it won't be available when we write in netCDF mode

[arbennett]

Solved in #10.