Part IV - Forecasts
|Foreword |||ARPEGE and AROME |||Tarbes CDM local forecast|
AROME and ARPEGE
During the field campaign, specific outputs from two French operational Numerical Weather Prediction models will be saved.
The first model is ARPEGE, this is a global model with a stretched horizontal resolution varying from 10 km over France up to 55 km over Australia. It is run every 6 hours for 96h. There is a 4D var data assimilation system.
The second model is AROME, this is a non-hydrostatic model with a regular horizontal resolution of 2.5km. It is also run every 6 hours for 30 hours. There is an assimilation system for the surface (optimal interpolation) and for the atmosphere (3D var assimilation system with mesoscale data assimilation, Doppler radar reflectivities and winds). This model is forced at the lateral boundaries by the forecast of ARPEGE. During the campaign, this model will be rerun on a smaller domain covering the South-West of France with lateral boundaries from the operational AROME model. This will allow the computation of diagnostics. This run will be used for the real-time and time-delay outputs.
The real-time outputs:
During the field campaign, specific outputs from those two operational Numerical Weather Prediction models will be available on the BOC. These will concern model runs from 00h and 12h.They consist of horizontal map of cloud cover (low, medium, high and total), wind, temperature and humidity at 2/10m, 100m, 500m 1000m and 4000m, the boundary-layer height diagnosed as the first level where the turbulent kinetic energy gets lower than 10-2 m²/s² and hourly cumulated rainfall. The principal area covers a zone from 41°N to 46.5°N and from 2.5°W to 4°E. A zoom on a domain of 1°x1° centered over Lannemezan will also be available for the AROME outputs as well as temporal series of vertical profiles of cloud cover and relative humidity at Lannemezan and Toulouse.
The time-delayed model products:
Supplementary products will be available after the field campaign. This includes :
- vertical profiles of prognostic variables (t, q, ,u, v, tke) and their budgets
- horizontal cross-section of tke
- u*, w*
- surface fluxes, surface variables and soil variables
- horizontal variability on a box of 10x10km² (5x5 points in AROME)
- very high-frequent outputs (every timestep): for T,q, u, v to compare with tower observations (-> 65m), tethered ballons, lidar...