GABLS4: AN INTERCOMPARISON CASE TO STUDY THE STABLE BOUNDARY LAYER WITH SURFACE INTERACTIONS ON THE ANTARCTIC PLATEAU

Within GABLS (GEWEX Atmospheric Boundary Layer Study), inter-comparison studies are carried on for boundary layer parameterizations schemes in use by numerical weather prediction and climate models. Under stable stratifications, models still have large biases, which depend on the parameterizations used for boundary layer and for surface (Holtslag et al, 2013, BAMS). The first three GABLS inter-comparison studies (Cuxart et al. 2006, Svensson et al 2011, Bosveld et al 2012) dealt only with moderate stable conditions. Also in GABLS1 and 2 a prescribed surface temperature was utilised. In the next GABLS4 case, we aim to study the interaction of a boundary layer with strong stability (Ri>>1) with a snow surface (having a low conductivity and a high cooling potential). Such a case is explored using observations at the Antarctic Plateau at Dome-C.

Several types of in-situ measurements of the surface and the boundary layer are available at Dome-C, notably a 45m tower with 6 levels of sensors measuring temperature, wind and humidity (Genthon et al, 2010), the temperature profile in the snow pack from the surface to 2m depth, radiative fluxes (Lanconelli et al, 2011), and twice a day radio-soundings at specific times. Due to the extreme weather condition, the measurement of the meteorological and soil parameters is more difficult and the computation of the turbulent parameter, such as Turbulent Kinetic Energy, and surface fluxes is a challenge.

For the GABLS4 case, one difficulty was to find a “ideal or golden” day with all measurements available and validated during at least 24h with a clear sky, low wind, a high Richardson number and with a large diurnal cycle. On D day, a specific 4Dvar re-analysis has been made with the stretched global model ARPEGE (10km on the Antarctic Plateau). Moreover, an improved boundary layer analysis using all the vertical levels (every 2m) from the radio-sounding was implemented to provide better initial and lateral boundary conditions to the non-hydrostatic model (2.5km) used to compute the forcing term (advection, geostrophic wind) for 1D models.

Finally, the set-up of the GABLS4 inter-comparison for the 1D model will be explained and presented together with a possible LES inter-comparison.