This thesis treats the understanding and prediction of the stable atmospheric boundary layer over land. This is the air layer that is next to the Earth surface, where humans live, where we grow our crops, extract wind energy, but also where we emit air pollution. A stable boundary layer develops when it is colder at the surface than aloft. This occurs predominantly at night, in wintertime also during the day, in polar areas, and surprisingly also over irrigated terrain in semi-arid areas.
Weather forecast and climate models make large systematic errors for wind speed and temperature in stable conditions. In polar areas these models predict a winter climate that is 10ºC warmer than observed. Thus climate models need substantial improvement. In the stable boundary layer many physical processes play a role, that together with their interactions should be well represented in models to obtain reliable forecasts. These processes are turbulent mixing, radiative transport, the coupling between the atmosphere and the underlying soil and vegetation, gravity waves, effects of heterogeneity of the land use.
We evaluate and improve weather forecast models through analysis if detailed observations of wind speed, temperature and radiation and turbulent fluxes. A realistic description of the coupling between the atmosphere appears to be essential for successful night-time forecasts, especially for calm conditions. A new aspect is that also small scale orography shows considerable impact on the wind speed profile in the boundary layer. Finally, we propose an alternative and robust expression for the stable boundary layer height, for use in air quality modelling.
Titel thesis: "Understanding and Prediction of Stable Atmospheric Boundary Layers over Land"