%0 Journal Article
%T An online trajectory module (version 1.0) for the non-hydrostatic numerical weather prediction model COSMO
%A A. K. Miltenberger
%A S. Pfahl
%A H. Wernli
%J Geoscientific Model Development Discussions
%D 2013
%I Copernicus Publications
%R 10.5194/gmdd-6-1223-2013
%X A module to calculate online trajectories has been implemented into the non-hydrostatic limited-area weather prediction and climate model COSMO. Whereas offline trajectories are calculated with wind fields from model output, which is typically available every one to six hours, online trajectories use the simulated wind field at every model time step (typically less than a minute) to solve the trajectory equation. As a consequence, online trajectories much better capture the short-term temporal fluctuations of the wind field, which is particularly important for mesoscale flows near topography and convective clouds, and they do not suffer from temporal interpolation errors between model output times. The numerical implementation of online trajectories in the COSMO model is based upon an established offline trajectory tool and takes full account of the horizontal domain decomposition that is used for parallelization of the COSMO model. Although a perfect workload balance cannot be achieved for the trajectory module (due to the fact that trajectory positions are not necessarily equally distributed over the model domain), the additional computational costs are fairly small for high-resolution simulations. Various options have been implemented to initialize online trajectories at different locations and times during the model simulation. As a first application of the new COSMO module an Alpine North F hn event in summer 1987 has been simulated with horizontal resolutions of 2.2 km, 7 km, and 14 km. It is shown that low-tropospheric trajectories calculated offline with one- to six-hourly wind fields can significantly deviate from trajectories calculated online. Deviations increase with decreasing model grid spacing and are particularly large in regions of deep convection and strong orographic flow distortion. On average, for this particular case study, horizontal and vertical positions between online and offline trajectories differed by 50¨C190 km and 150¨C750 m, respectively, after 24 h. This first application illustrates the potential for Lagrangian studies of mesoscale flows in high-resolution convection-resolving simulations using online trajectories.
%U http://www.geosci-model-dev-discuss.net/6/1223/2013/gmdd-6-1223-2013.pdf