Can ROMS read in albedo field from external files?

[LeaF_7]

Hello,

I wonder if I can map albedo values (from MODIS) to ROMS grid, is there a way to read in the albedo value when the model runs?

Thanks in advance for any suggestions.

Fan

[kate]

The ROMS trunk code does not read albedo, but my branch does. Check out the ALBEDO_FILE option.

[LeaF_7]

Thanks a lot, Kate.

I am currently working with your branch, will try to find out how to use it.

Thanks again!

Fan

[LeaF_7]

Sorry Kate, I think i still need you help on how to use ALBEDO_FILE.

Should I provide it as a forcing file? Is it able to use a constant albedo field or the file need to be like the forcing files with time index?

Thanks a lot!

[kate]

ALBEDO_FILE is looking for a forcing file. I have only ever used it with the MERRA albedo, which is a function of time and space. I believe forcing fields can have files with only one record, which eliminates the time interpolation, though I haven't tried it.

If what you want is a really simple (uniform?) field, you can use ANA_ALBEDO instead.

[LeaF_7]

Thanks again!

I am using the reflectance values from MODIS MYD09A1, and mapping them to my model grid.

I am modelling the Yellow Sea with Solar radiation from ERA-interim(0.75 degree resolution).

Coastal region is always being over heated, I was suggested to look into the albedo distribution as the sediment suspension in this region.

Great thanks for the help!

[LeaF_7]

I tested the ALBEDO_FILE.
It just works as the forcing files, which need the time variable.
To use a temporal invariant field, just provide the albedo at starttime and endtime (2 time index). I guess ROMS will do a linear interpolation to the albedo value.

[m.hadfield]

Forcing files require a time variable, but that can be given the attribute "calendar" with the value "none", in which case only a single value is required and will be applied to all times.

I assume Kate's albedo files work the same way, and if they don't it wouldn't be hard to fix the code so they do.