The case was launched in summer 2014 during the GEWEX conference in July 2014 and during the World Weather Open Science Conference (WWOSC) in Montreal (GABLS4 presentation)

The “gloden day”, based on a real case, is the 11th, Dec 2009 at 00UTC.
DomeC is situated at 3233m above the sea level at latitude -75.1° and longitude 123.35°.

The atmospheric forcing and initial conditions are available in the NetCdf format for a 36h forecast length, the case uses a geostrophic wind and advection for temperature and specific humidity (time variable for stage 1 and 2).

For the LSM, SCM and LES, a reference simulation is required with the following prescribed value for the surface and snow characteristics, for
all the stages:
- fixed albedo= 0.81 and fixed emissivity = 0.98
- z0m = 0.001 m and z0h/q = 0.0001m
- for the one layer snow scheme (bulk) snow density 300Kg/m3
- prescribed vertical grid (A, B, P ,z)
on half level (91) or (P, z) on full level (90)

Please specify the reference experiment with exp=FREF.
In case, you can not use all the prescribed surface value but ONLY the prescribed vertical grid: exp=GREF.
Finally, if you can not use the prescribed vertical grid but you have all the prescribed surface characteristics: exp=SREF


example :

Nevertheless, in addition, experiments with the “operational” snow characteristics or values coming from the stage0 (see below) can be send.

The GABLS4 intercomparison follows in some aspects the DICE intercomparison (DICE) with the same philosophy such as LSM forced by observations and SCM with LSM and driven by the surface.

The case is organized in two parts interactive surface and prescribed surface

Main contact for LSM : Ouput format for the LSM in stage 0 (lsm_output)
Main contact for SCM : Ouput format for the SCM in stage 1/2/3 (scm_output)
Main contact for LES : Ouput format for the LES in stage 1/2/3 (les_output)

Part 1 : Original timetable results returned for 01/12/2014 NOT ANYMORE VALID !
- stage0 : Land (Snow) Surface Model (LSM) driven by observations from 01/12/2009 up to 15/12/2009. Stage 0 has two objectives first provide the surface and the soil (snow) initial conditions for the 11 December 2009 at 00UTC for the Single Column Model (SCM stage 1a) and an intercomparison of the snow scheme for this period (two weeks). Inital condition (init_soil).Observed forcing (driver_stage0).
Email to and (list_stage0)

- stage1a : the single column model (SCM) will use the same LSM, used in stage 0, with the initial surface and soil conditions from stage 0 valid for the 11 december 2009 at 00UTC. Sensitivity experiments are welcome (3 max by participant) such as : operational vertical grid, time step, PBL parameterization, etc ..... Atmospheric forcing and initial conditions (driver_stage1).
- stage1b : the single column model (SCM) will use the same LSM, used in stage 0, BUT with the initial surface and soil conditions given in (driver_stage1). 3 maximum sensitivity experiments by participant.
stage 1a and 1b please send email to,, (list_stage1)

LES simulations are welcome for stage 1b and 1a however due to the cost of a LES simulation with the radiation code and a surface scheme, the effort should be put on stage 2 and stage 3 (“ideal” case see below) . People potentially interested should contact and (cc

Part 2 : Original timetable results returned for 01/05/2015 NOT ANYMORE VALID

- stage2 : (for LES and SCM), same atmospheric forcing as stage1 BUT the surface temperature is given (hourly value) driver_stage2. The version of the SCM used must be the one used for stage 1a/b. The aim is to study the impact of the surface interaction and feed-back between the surface scheme and PBL.Email to and

- stage3 : (for LES and SCM) is more an “ideal” or simplified GABLS4 case: no radiation, no humidity, prescribed surface temperature (same as stage2) and constant geostrophic wind in time . The surface evaporation/sublimation must be equal to zero with no available water in the surface layer driver_stage3. Stage 3 is probably easier for the LES community but also for other community like fluid mechanics. Email to and

For any questions, errors or general remarks please contact :,,,