@Tomber42
@idefix37
@ivo

I think it answers your/our question!

btw how does AROME handle the subject?

Yes, I think... but to clarify the "Freezing altitude" layer on Windy, should we understand that this level is counted as Altitude above sea level, but when this level intersects terrain located at a higher altitude , then this level is measured in relation to the ground itself, as Height above the ground.
So this layer should be called FREEZING LEVEL, as sometime it is measured as ALTITUDE above sea level (asl), sometime as HEIGHT above ground.

In Windy, if you check the Freezing level with AROME it switches to ECMWF. So this layer is not available with this model

In the example below (left map) the Freezing level is at 2400m in West of the Alps and at 0m just above the high summits, as described here above for ICON.
On same locations, the Freezing altitude from AROME data on Meteociel (Right map) is shown at 2700-2600m even when this altitude intersects mountains at 3000 or 4000m.

So there are different ways to display the Freezing level as we could try to summarise with this sketch:

I am more familiar with Freezing altitude asl, which is much more precise for outdoors activity in mountains. The Height above ground is too coarse due to the low altitude resolution of the orographic model integrated in the model itself.

Anyway, at least, it clarifies the way how this layer must be interpreted...

Thanks for your investigation

Alain

DWD answered:
" the freezing level (HZEROCL) is provided in m ASL,
but it is set to zero (rather than the height of the model topography)
if the freezing level is at the surface.

However, I see that this should be changed in our diagnostic
calculation, as well as the scanning direction (which should be top-down
rather than bottom-up) in order to detect the uppermost freezing level."

So, dear idefix37, I think it answers your/our question!

btw how does AROME handle the subject?

Sehr geehrte Damen und Herren,
Ich hätte eine Frage zur Dokumentation des Icon Modell. In der Doku "insert link to dokument " sprechen Sie bei der Beschreibung von HZEROCL, Tabelle 7.5, Seite 46,Version 1.2.6 von MSL oder über Grund Höhe?

Mit freundlichen Grüßen
"your name"

1. DWD ICON ref. manual is wrong
   

2. Windy uses a different parameter!

I hope Windy team will investigate that.

But how do you explain that this morning in Greece ICON shows on Windy a Freezing « altitude » at 0m in mountains of 1800m alt. approx. and, at close distance, in low land of 100m alt. approx., a Freezing « altitude » at 1300m.

Switching to ECMWF we see the same figures at same places, but just with a lower resolution map.
For me both ICON and ECMWF on Windy seem to display a Freezing HEIGHT AGL.

Yeah I think those changes wouldn't be hard to make... But that's a non-developer point of view :)

for ICON model, the Height of 0 degree Celsius isotherm (short name HZEROLOC)
has MSL as a reference.

Thus, the frz. alt. from ECMWF and frz. alt. from ICON, are different!

Nuremberg's altitude is 314m

Model's "surface" (named Orography(Z) by ECMWF with ID=129)
over rough terrain may vary a lot compared with actual altitude from MSL.
https://www.ecmwf.int/en/forecasts/datasets/set-i#I-i-a

It might be a good idea to have a Freezing Level chart using altitude from MSL,
where FRZ LVL = DEG0L+Z

First, it might be a good idea to explain what are height and altitude to other users => height has the ground level for reference (AGL-above ground level). Altitude has the main sea level for reference (MSL)

There are in my opinion two options regarding aeronautical weather data :

-Forecaster point of view: Using charts, he gives the ALTITUDE of clouds (from MSL - main sea level) and doesn't care how you use the info...

-Pilots or controllers point of view: They don't want you to crash and provide the HEIGHT (from AGL - above ground level) of cloud base. So you have an idea of what you're going to see (or not...) while arriving or departing from an airport, whatever the altitude of the airfield.

Considering this, I think that giving the HEIGHT of cloud base is the best option (AGL), which is already the case. It is also good for non pilot users as it provides a useful and easily understandable info.

Regarding Freezing level... It's indeed more useful for pilots when using ALTITUDE (MSL) for other technical reasons. And... It's more logical for pilots but also for other users ... For hikers for exemple. They want to know the freezing ALTITUDE and not height, so they can compare it with the ALTITUDE of a mountain for exemple, because what is depicted on topo charts are ALTITUDEs.

A good solution in my opinion would be to use the words "height" and "altitude"in windy layers names and provide a clear definition of these 2 terms in the help forum section/legend, and obviously use MSL for freezing level and AGL for cloud base.

We switched to ECMWF parameter Ceiling, because we found it more usefull.

In Windy it has the same name (cloud base), but now it shows height of clouds where the cloud cover is >50%.

I think ceiling would be useful from pilots perspective. Currently I check cloud base and low cloud to get an idea of the base and amount of cloud. Ceiling seems to combine both these with the >50% filter. I think I would use ceiling first and then maybe base and low cloud if I needed more info.

More about Cloud bases: Can you tell us what do you use it for? Are you a pilot? We are thinking of adding new ECMWF parameter called Ceiling, which is similar to present Cloud base parameter. It may be interesting for you. If you want to know more, you can read this short description: https://software.ecmwf.int/wiki/display/FCST/43r1+new+parameters%3A+Ceiling

What do you think about this parameter? Would it be more useful for you than Cloud base?