[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[python #OCF-850030]: Hello?i have a question for Metpy 's calc on point and array

Hi, and thanks for reaching out!

Right now, most of MetPy's iterative vertical calculations (profile 
calculations, layer depths, etc.) must be limited to use on 1-dimensional 
profiles. Looping across large grids for these calculations can be done, but 
can cost time or computation resources. An example:

    import xarray as xr
    from metpy.calc import dewpoint_from_relative_humidity, showalter_index
    from metpy.cbook import get_test_data
    from metpy.units import units
    
    ds = xr.open_dataset(get_test_data("NAM_test.nc", False))
    
    ds["Dewpoint_isobaric"] = dewpoint_from_relative_humidity(
        ds["Temperature_isobaric"], ds["Relative_humidity_isobaric"])
    
    showalter_indices = []
    
    for x in ds.x:
        for y in ds.y:
            column = ds.sel(x=x, y=y)
            si = showalter_index(column["isobaric"],
                             column["Temperature_isobaric"].squeeze(),
                             column["Dewpoint_isobaric"].squeeze())
            
            da = xr.DataArray(si, dims=["time"], coords={"time": column.time, 
"y": column.y, "x": column.x})
            da = da.expand_dims(["y", "x"])
        
            showalter_indices.append(da)
    
    xr.combine_by_coords(showalter_indices)

And there are some tools in xarray like `apply_ufunc` and `map_blocks` that can 
somewhat parallelize this for you with dask 
(https://docs.xarray.dev/en/stable/user-guide/dask.html#automatic-parallelization-with-apply-ufunc-and-map-blocks).
 I may be able to throw together some example code for this in the future. As 
part of funded NSF grant work, we have high priority efforts to speed up our 
profile calculations and explore making more of our calculation suite 
grid-friendly. In the mean time, I hope this helps! Let me know if I can 
provide any further help.


All the best,

Drew


> I appreciate using this package.
> Now, i have a question when I use this package to calculate the
> showalter index, I found that I could only count single points. But I want
> to calculate the array, it seems that this package does not support, is
> there a good way. I wish I could get a little help
> 


Ticket Details
===================
Ticket ID: OCF-850030
Department: Support Python
Priority: Low
Status: Closed
===================
NOTE: All email exchanges with Unidata User Support are recorded in the Unidata 
inquiry tracking system and then made publicly available through the web.  If 
you do not want to have your interactions made available in this way, you must 
let us know in each email you send to us.