snow depth (and density) threshold for 100 % snow coverage: inconsistency between documentation on ECMWF charts and cy49r1 physical processes documentation ?

AIFS, IFS and OpenIFS IFS
1

Hello.

I noticed a potential inconsistency regarding the snow depth product documentation on::

The documentation states that:

“Green shades relate directly to model assumptions regarding proportion of the ground assumed to be covered by snow - when the diagnosed depth is less than 10cm the model assumes that there is only partial cover, and surface fluxes are treated accordingly. Thus the green shades (< 10cm) aim to signify, visually, that some of the underlying vegetation is showing through. For depths above 10cm a total cover is assumed (so no green hues).”

However, in the Physical Processes documentation of the IFS cy49r1:

Equation 8.2 of the Page 153 describes the snow cover c_sn as depending explicitly on snow depth (S_tot/rho_sn,eq) AND snow (whole) layer equivalent density rho_sn,eq. Following this equation 8.2, the minimum snow depth leading to a 100 % snow coverage (c_sn=1), assuming a snow layer equivalent density of 300 kg/m³, is 20 cm (which indeed gives c_sn=0.99), and not 10 cm as indicated in the documentation of the above mentioned snow depth charts.

This 10 cm snow depth threshold seems to come from a previous version of the above mentioned relationship in a older IFS cycle:

https://www.ecmwf.int/sites/default/files/elibrary/2016/17117-part-iv-physical-processes.pdf#section.H.4

See Equation 8.2, page 129.

So could you clarify which value should be considered authoritative for interpreting the snow depth charts? Or may be I missed something? :wink:

Thank you for your time and guidance.

Best regards,

Fabian Debal,
Royal Meteorological Institute of Belgium.

2

That’s a very good observation. Documentation can sometimes lag behind changes introduced in newer model cycles, so it’s always worth checking the underlying physics instead of relying solely on chart descriptions. Small differences in how snow cover fraction is parameterized can have a noticeable impact on surface temperature, albedo, and ultimately snow forecasts. It’s discussions like these that help users better understand model limitations.