noaa-oar-arl / canopy-app

Stand-alone/column canopy codes and parameterizations
MIT License
6 stars 7 forks source link

Develop #47

Closed drnimbusrain closed 1 year ago

drnimbusrain commented 1 year ago

This PR addresses a few small changes to adding a new Flameh_Opt (=2) that allows for using FRP to calculate flameh for active fire grid cells, AND uses the input user defined flameh elsewhere FRP=0. This PR also simply refines the VTYPE mapping to Massman for determining canopy parameters/shapes, and improves the roughness sublayer option (RSL_Opt) to be more consistent with Massman approach.

drnimbusrain commented 1 year ago

@zmoon Added two more small commits (8c4a56c and 0e388fb) that simply remove additional 1D variable calculation when 2D NetCDF input is used, because of earlier changes that remove text write/output when input 2D NetCDF is used. I feel this becomes unnecessary as input 2D would be associated with 2D output. Also, when a 2D input domain becomes large, the output 1D text files become much too large in size and slows the model down considerably to write to them. The README has been updated accordingly.

drnimbusrain commented 1 year ago

@zmoon Please note that we have merged in the new netcdf format and other bug fixes that pertained to previous new options and modifications.

drnimbusrain commented 1 year ago

@zmoon seems like this is looking good for merge?

drnimbusrain commented 1 year ago

@zmoon I agree with the changes, maybe some were there later in the PR for modifying the output netcdf formats. I also agree, and will add a levels coordinate variable.

@zmoon I have added the levels as a coordinate variable here.

variables: float levels(levels) ; levels:long_name = "height above ground level" ; levels:units = "m" ;

zmoon commented 1 year ago

@zmoon I have added the levels as a coordinate variable here.

Great! I would make the dim name level or lev (or z) instead of levels though. And the coordinate could be z. Would we call these the interface levels (they include z=0)? Doesn't matter right now, but might in the future.

drnimbusrain commented 1 year ago

@zmoon Is this better now?

dimensions:
        time = UNLIMITED ; // (1 currently)
        timestr = 1 ;
        grid_xt = 86 ;
        grid_yt = 43 ;
        level = 100 ;
variables:
        float z(level) ;
                z:long_name = "model interface level" ;
                z:units = "m" ;