Peak Performance

Technical Environment

FMI NWP Activities

Model Configuration Data assimilation Upper air analysis3-dimensional variational data assimilation VersionHIRVDA 6.2.1, FGAT option ParametersSurface pressure, temperature, wind components, humidity Surface analysisSeparate analysis, consistent with the mosaic approach of surface/soil processes, for: SST, fraction of ice snow depth screen level temperature and humidity soil temperature and humidity at two layers Grid length 0.2  in the horizontal Integration domain 438 x 336 gridpoints in rotated lat/lon grid Levels40 hybrid levels defined by A and B. Observation typesTEMP, PILOT, SYNOP, SHIP, BUOY, AIREP First guess3 h forecast, 3 h cycle InitializationDigital filter (IDFI) Cut-off time2 h for main cycles, 4 h 20 min for intermediate cycles Assimilation cycle3 h cycle Forecast model Limited area gridpoint model VersionHirlam Basic equationsPrimitive equations Independent variables,  (transformed lat-lon coordinates), η, t Dependent variablesT, u, v, q, p s, cloud water, turbulent kinetic energy DiscretizationArakawa C grid Grid length 0.2  in the horizontal Integration domain438 x 336 gridpoints in rotated lat/lon grid Levels40 hybrid levels defined by A and B Integration scheme Semi-Lagrangian, semi-implicit,  t=360 s. OrographyHirlam physiographic data base, filtered PhysicsSavijärvi radiation scheme Turbulence based on turbulent kinetic energy STRACO condensation scheme Surface fluxes using the drag formulation Surface and soil processes using mosaic tiles No gravity wave drag Horizontal diffusionImplicit fourth order Forecast length54 hours Output frequency1 hour Boundaries 0.2  “frame” boundaries from ECMWF received four times a day 3 h temporal resolution

Sodankylä Mast Comparison: HIRLAM vs. ARPEGE HIRLAM ARPEGE 32m Temperature