Selecting hail at surface from HRRR model

[btickell]

I am trying to select specifically hail diameters at the surface.

H = Herbie(
    "2021-07-19",
    model="hrrr",
    product="sfc",
    fxx=0
)

x = H.xarray()

The above code does work but I'd like to select just the hail field for performance / download speed reasons.

I tried using :hail: when calling xarray but no dice. I admit I am not very familiar with the underlying GRIB structure. Is selecting subfields like this supported by Herbie?

[blaylockbk]

Good question.

Here is a table of some grib variable names...
https://www.nco.ncep.noaa.gov/pmb/docs/grib2/grib2_doc/grib2_table4-2-0-1.shtml

I'll have to sit at my computer to see what is actually in the HRRR sfc output, but I'd start to look for EFRHAIL which is effective radius of hail.

If you beat me to it, let me know what you find.

[btickell]

Ah thanks! Trying both EFRHAIL and HAIL (caps being important here) I do get results for HAIL but not for the effective hail.

I recieve an array with a step index of size 3 (all with value 0), but seems like only the first step has meaningful data. Do you have any insight into what this step refers to?

[blaylockbk]

It looks like there are three different levels the HAIL variable is output for.

from herbie import Herbie
H = Herbie("2023-10-01", fxx=6)
H.inventory("HAIL")

	grib_message	start_byte	end_byte	range	reference_time	valid_time	variable	level	forecast_time	search_this
52	53	36161743	36429492	36161743-36429492	2023-10-01	2023-10-01 06:00:00	HAIL	entire atmosphere	5-6 hour max fcst	:HAIL:entire atmosphere:5-6 hour max fcst
53	54	36429492	36466353	36429492-36466353	2023-10-01	2023-10-01 06:00:00	HAIL	0.1 sigma level	5-6 hour max fcst	:HAIL:0.1 sigma level:5-6 hour max fcst
54	55	36466353	36472988	36466353-36472988	2023-10-01	2023-10-01 06:00:00	HAIL	surface	5-6 hour max fcst	:HAIL:surface:5-6 hour max fcst

I think you are looking for the surface parameter, so filter by that

ds = H.xarray("HAIL:surface")
ds.hail.plot()

Now, I'm not too certain about the units. It might be worthwhile contacting the HRRR developers and ask them to clarify what this field represents.
https://github.com/NOAA-GSL/GSL-Regional-Model-Forum/discussions/