Research & operational forecasting with the NMM: Radiation & fog in complex topography

Matthias Müller

University of  Basel

At the University of Basel, numerical weather forecasts for Europe are computed twice daily using the NMM, and are made available on the internet []. A visualization system was developed and
some animations are now used daily by national television. Special focus lies on the Alpine region, which is covered by two nested domains running at resolutions of 4 and 2 km, respectively.

A parameterization for radiation in complex topography was developed and tested. The interactions of
radiation fluxes with the terrain due to sky view restriction, slope and aspect angle as well as shadows are computed based on a digital elevation model that has a higher resolution than the mesoscale model. Runtime computational costs are negligible, and temperature differences as high as 3 K are found in complex topography. Especially at night, the cold bias in valleys is significantly reduced.

Current research efforts focus on fog and low stratus clouds. An assimilation and ensemble system for 1D-
modelling was developed, but a 3D approach seems more promising for the topography in Switzerland. Therefore, detailed microphysics of the 1D fog model PAFOG were implemented into the Non-hydrostatic Mesoscale Model (NMM). This introduces total droplet number concentration as a new prognostic variable and a sophisticated determination of condensation/evaporation as well as size dependent droplet settling. The nature of radiation fog requires a high vertical resolution to model the steady growth from the surface. A thickness of approximately 4 meters in the lowest layers combined with the stable stratification during a fog event can cause problems in the standard turbulent exchange, resulting in a cold bias. In the case of liquid water advection, boundary conditions for the droplet number concentration have to be specified. In general, the new microphysics decrease fog water content to a more realistic value and the complex spatial distribution of simulated fog underlines the importance of the 3D approach.