Poster Sessions
P-22 |
|
Radar assimilation in COSMO-KENDA |
Michael Wuersch (University of Munich) |
| Abstract |
|
Using COSMO-KENDA, various experiments have been performed to study the impact of radar observations with an LETKF in a non-hydrostatic model. First, it has been analysed whether the general behaviour of the system in a real setup is similar to tests that have been done in an idealized setup in Lange & Craig (2014). The focus in these experiments lies on the impact the localization radius and observation averaging have on the analysis and forecast quality. Therefore the same methods are applied in our study. Additionally we look at balance issues, assimilation increments and probabilistic scores as well as varying the observation error. The predictability of the system is also analyzed. In additional experiments we have a look at the initialization method and the time it takes until storms are assimilated. We also investigate the impact the assimilation has on the unobserved fields and large scale features like the temperature profile. Heiner Lange and George C. Craig, 2014: The Impact of Data Assimilation Length Scales on Analysis and Prediction of Convective Storms. Mon. Wea. Rev., 142, 3781-3808. |