Confronting High-Resolution Simulations of Deep Convective Growth with Meteosat Observations

We investigate the characteristics of deep convective clouds during their growth phase based on a combination of high-resolution numerical simulations and Meteosat observations. The simulations are performed with the ICON-LEM model with grid spacings of 150, 300 and 600 meters and within a relatively large domain covering Germany. This unique simulation setup demonstrates the potential capabilities of future operational forecasting systems in which convective-scale processes become more and more resolved. We analyze convective development simulated at different resolutions using the Lagrangian framework which had been applied to study satellite-observed deep convective growth in earlier studies. As a first step, the simulated profiles of thermodynamics variables and hydrometeors are converted into so-called synthetic brightness temperatures using the SynSat satellite forward operator. Then, identification and tracking of simulated cloud growth is performed on synthetic 10.8 micron brightness temperatures. Finally, we compare several growth characteristics, for instance cloud-top cooling rate, anvil expansion speed - and also change in surface precipitation intensity, to its observational counterpart. Furthermore, we discuss the effect of the spatial resolution for the derivation of growth signatures to better understand the limitations of current and future sensors.