Dyamond2 Example

[sehnem]

No description provided.

[sehnem]

@RobertPincus I added the dyamond example to a notebook and loaded using dask, I am just not sure about the data transformations that need to be done for rel and rei, probably will need some help with it.

I created a function to download the data and created this new PR as it rebases the latest changes.

[RobertPincus]

@sehnem I only have time for a quick peek but it seems the units for cloud optics should be g/m^2 for LWP/IWP and microns (10^-6 m) for particle sizes. I suspect that means you need to compute LWP = (DEPLP * QL)/sci.g * 1000 and pass RL*10e6 to the Python calls.

[RobertPincus]

@sehnem I've just pushed a commit to address units as per the comment above. You can see if this runs and give reasonable results?

[RobertPincus]

Seems like this supersedes #114

[sehnem]

@RobertPincus I changed all the calls to stack the dimensions that are not core, so you can run dyamond2 without renaming the dimensions Xdim, Ydim and nf and it will call the fortran functions sending the chunks as cols. This was expanded to all the functions making this PR not a small one.

I also made some tests running it with the full Xdim and 300 columns from Ydim and it worked.

I think it should be good for merging.