An Online-Trajectory Module for the Nonhydrostatic Numerical Weather Prediction Model COSMO - Case Study of a Foehn Event

The Lagrangian perspective on atmospheric processes has become a frequently used way to investigate air streams in the atmosphere since the middle of the 20th century. Usually the path of air parcels is calculated based on wind fields from some numerical model stored at a certain time interval (most often between 1 and 6 hours). This time interval as well as the grid spacing of the model influence the accuracy of the numerical solution to the trajectory equation as the wind field has to be interpolated in space and time. In recent years the spatial resolution of numerical models has increased strongly and it is often argued that the classical approach to calculate trajectories offline is not capable to capture the full variability that is represented in the Eulerian model. In addition spatial and temporal resolution seem to be no longer balanced. We therefore implemented an online-trajectory module in the nonhydrostatic numerical weather prediction model COSMO, which calculates trajectories in parallel with the integration of the Eulerian model. As the wind fields are available at each model timestep, no temporal interpolation is required and this error source can be reduced to a minimum. In a first case-study we simulated a F¨ohn event in July 1987 with horizontal resolutions of 14 km and 7 km. A comparison to ”classical” offline trajectories shows that the large-scale flow pattern is not strongly altered. However, mesoscale features are better captured by the online-trajectories and the path of individual air parcels is quite strongly affected: The average distance between an online- and an offline-trajectory is about 800-1000 km in the horizontal and 700-1000m in the vertical after 96 h of integration. In addition, by comparing the distribution of temperature fluctuations, it becomes clear that the higher temporal resolution is crucial to capture the variability represented in the Eulerian model. It is evident from this case-study that the online calculation of trajectories can capture also large spatial or temporal changes of the wind field by greatly reducing the temporal interpolation error. Hence particularly the representation of air-streams in complex terrain can strongly benefit from online-trajectories.