Experimental operational WRF forecast for Wageningen

The WRF model (ARW core) is run daily, based on initial conditions and boundary conditions taken from the NCEP GFS model (initialized at 00 GMT). The two domains of WRF are centered around Wagenigen, and have grid sizes of 25 and 5 kilometers (and domain sizes of both domains are 105 times 105 and 121 times 121 points, this resolution is in use as of May 25, 2015). In the vertical 30 levels are used. Output is normally available at approx. 6:30 GMT. Since 2014 an extra domain, centered around Ireland has been added to feed data into a decision system system for the management of Late Blight (Phytophotora) in potatos.

In the model the following parameterizations are used:

• Land surface model: Noah LSM (sf_surface_physics=2)
• Boundary layer turbulence: YSU (bl_pbl_physics=1)
• Microphysics: ETA microphysics (mp_physics=5)
• Cumulus parameterization: Grell Devenyi (cu_hysics=3)
• Longwave radiation: RRTM (ra_lw_pysics=1)
• Shortwave radiation: Dudhia (ra_sw_pysics=1)

The output for the individual domains van be found at:

• Domain 1
• Domain 2

Time series for near surface quantities, as well as time series of vertical soundings, both at the location of Wageningen, can be found at:

• Timeseries Haarweg
• Vertical timeseries above Haarweg

Disclaimer

Note this is experimental so do not use this information for anything really important.

This is only model output. It takes a skilled meteorologist to make a forecast out of this. He/she knows about the weaknesses of atmospheric models and can interpret the numbers produced by the model correctly. Furthermore, this MM5 forecast is initialized only with data from a global model, with a coarse resolution of approximately 50 kilometers. To make a real forecast, one should also use as many observations (ground-based, upper-air and satellite-based) as possible to fix the initial state of the model well.

Note that this WRF forecast replaces the MM5 forecast as of May 1, 2009.

More information on atmospheric modelling

The background and application of mesoscale models is taught in our course Atmospheric Modelling (MAQ-31806).

More information on numerical weather prediction can be found in the online course from Euromet.