Abstract
Summertime convective events and their associated hazards can pose considerable threats to people and property. To-date, forecasting convective events remain a challenge, even for the latest generation of convective-scale numerical weather prediction models. Forecasting these small-scale events is, among others, challenged by the presence of bias and errors in operational analysis data, which are used as initial conditions for subsequent forecasts.
We leverage observations from the "Swabian MOSES" 2023 field campaign (SWM 2023) that took place in summer 2023 in the Black Forest region, Southern Germany, to (i) validate analysis data sets at different scales and (ii) to improve a convective-scale analysis by additionally assimilating campaign observations.
During the campaign, the mobile atmospheric measurement platform, KITcube, was deployed. The measurements include a spatially distributed network of instruments to observe the dynamic and thermodynamic characteristics of the lower troposphere, and in particular a network of several Doppler wind lidars. Here, we will focus on the model representation of mesoscale flow characteristics using 3-months of continuous measurements and show how the assimilation of Doppler wind lidar retrievals using the non-hydrostatic model ICON and the Kilometer Scale Ensemble Data Assimilation system (KENDA) influences the analysis.