The effect of a sub-grid statistical cloud-cover scheme applied to the COSMO local numerical weather prediction model over the wider geographical domain of Greece

The COSMO model uses operationally two sub-grid schemes for the evaluation of stratus clouds. A semi-empirical scheme based on relative humidity is used in the radiation module while a statistical scheme is used in the turbulence module. The objective is to investigate the possibility of the implementation of the statistical scheme also in the radiation module. The relative impact is presented in reference to a spring test case with synoptic conditions that favor stratiform clouds. The domain considered is the wider Balkan region around the Hellenic geographical area and is characterized by comparable sea and land partitions. This particular domain choice gives rise to a strong coexistence of continental as well as marine clouds which is one of the most challenging features regarding the operational use of numerical weather prediction models by the Hellenic Meteorological Service. The results are evaluated through direct comparisons with satellite data as well as the observed 2-m temperatures for an approximate total of fifty Greek synoptic meteorological stations. plementation of the statistical scheme led to an underestimation of low cloud-cover by the model in contrast to the implementation of the default relative-humidity scheme, while regarding medium cloud-cover the situation was reversed. Also, the daily 2-meter minimum and maximum temperatures were slightly better simulated, but not conclusively, when the statistical scheme was implemented in the radiation module. Although the statistical scheme cannot in its present form replace operationally the relative-humidity scheme in the radiation module, it is an important asset to COSMO model invoking valuable insight to the physics of the model and can be used as a basis to support the ongoing research in this crucial area of atmospheric sciences.