Poster Session P1.1 The Quality of Horizontal Advective Tendencies in Atmospheric Models

Monday, 9 June 2008
Fred C. Bosveld, KNMI, De Bilt, Netherlands; and E. Van Meijgaard, C. De Bruijn, and G. J. Steeneveld

Handout (1.3 MB)

Advective tendency terms in 3D atmospheric models form an intimate part of the model description. Although they represent real physical processes, it is not clear to what extent they correspond to the actual advective tendencies in the real atmosphere.

The accuracy of calculated advective tendencies becomes important in the simulation of atmospheric profiles with single-column models. In some experiments these models are driven with advection terms derived from the analysis of 3D NWP models. The same issue also becomes important when interpreting column information from advanced meteorological profiling sites. In general, one would like to interpret the observed changes in the column parameters in terms of physical tendencies. This can only be done if we can correct, with a reasonable accuracy, for the contribution to the changes due to advection.

Over the years, improvements in the parameterization of atmospheric processes, increase in model resolution, and the assimilation of more detailed observations have led to a better representation of the state of the atmosphere. Bosveld et al. (2004) compared horizontal advective tendencies as given by RACMO, the regional climate model of KNMI, with tendencies derived from flux divergence observations from the 200 m tower. A reasonable agreement was found for two selected days, one with advection and one without advection.

In this study we analyze advection for the GABLS 3rd stable boundary layer case (Baas et al, 2008). During this night a small clear air disturbance passes over the Cabauw site resulting in changes in temperature, humidity and momentum not accounted for by local vertical physical processes. We compare three 3D atmospheric models (RACMO, HIRLAM and WRF) and we evaluate the models with observations of Cabauw. Qualitative agreement is found among models and between models and observations. However, differences in magnitude and timing are found.

Baas, P., F.C. Bosveld, and G.-J. Steeneveld (2008). GABLS 3rd case Cabauw case selection and description. Presented at this Symposium.

Bosveld, F.C., E. van Meijgaard, E. Moors, and C. Werner, Interpretation of eddy-correlation flux observations at different levels along the Cabauw 200 m meteorological mast. 16th AMS symposium on Boundary Layer and Turbulence, 9-13th Aug 2004, 2004, Portland (ME), USA, American Meteorological Society.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner