Albedo problem in the Trunk

[peylin]

Christophe Dumas reported a problem with the snow albedo calculation in the Trunk. There are two options: i) using the CLM model parametrization or ii) using the previous ORC parametrization (referred as "Chalita and Letreut".

First note that the Chalita initial parametrization for snow only concern the nobio.

The problems that he founds:

1) With CLM snow scheme
The albedo is not calculated when the sun is down (ex. high latitude in winter). In principle the routine should return through XIOS an "undef" value under these condition
=> However it induce a model error when output the snow albedo !

2) With previous ORC snow scheme

• It works fine over the nobio regions
• But it leads to very strange results over land ! (like if the snow albedo was not properly accounted for)

=> This needs to be "debug" (btw CO, PP, CB, CD)
=> When this will work we could use the optimised parameters for the snow albedo (done with the version V2) in the TRUNK to see if it improves the overall albedo.