Kevin Bachmann, Hans-Ertel-Zentrum für Wetterforschung (HErZ)

Predictability of convective precipitation using the idealized COSMO-KENDA forecasting system

Heiner Lange, Christian Keil


The predictability of moist convection and therefore precipitation is strongly influenced by the interaction between large-scale forcing and local-scale flow characteristics. Predictability is increased in the presence of strong large-scale forcing, whereas cases with a weak large-scale forcing tend to be sensitive to the availability of triggers for convection. Particularly, the presence of orography can serve as a trigger for convective precipitation under weak large-scale forcing, thus enhancing its predictability. Following Keil et al.(2014) the convective adjustment timescale is used as an objective indicator for the prevalent weather regime allowing for a regime dependent inspection of the forcing mechanisms of moist convection and its predictability.

The idealized COSMO-KENDA test bed is used to investigate the influence of orography and synoptic conditions on the predictability of precipitation. As in Lange\&Craig(2014) convective-scale data assimilation of simulated Doppler radar observations is conducted. In addition, orography and large-scale forcing is implemented and the forecast quality is assessed with special regard to differences of amplitude and localization errors of precipitation areas. In a first step, the ensemble forecast quality is evaluated under the assumption of a perfect model.

File available here