3D reconstruction of cloud fields using cloud resolving modelling

The three-dimensional nature of clouds modifies incoming radiation and the representation of cloud-radiative effects is simplified in climate models. Radiative transfer thus assumes clouds to be plane-parallel and homogeneous, commonly known in the community as the independent column approximation method. Studies have shown that neglecting the 3D radiative transport effect can introduce substantial biases in simulated mean radiation fields. The C3SAR (Cloud Structure & Climate – Closing the 3D Gap) research unit was established, bringing together advanced remote sensing techniques and high-resolution modeling to investigate the 3D radiative effects of clouds. In our subproject, we use hectometer-scale simulations which are essential for resolving clouds, to investigate biases in 3D cloud-radiative effects.

To realize 3D cloud observations and make them available for different cloud scenarios is essential for our studies. We use the hectometer-scale simulations as a virtual testbed to test and evaluate potential 3D reconstruction algorithms, which could then be applied to satellite and ground-based products. While 3D reconstruction from observational data faces many challenges, these simulations provide a consistent framework to evaluate and estimate potential uncertainties.

For our simulations we apply the ICON model centered around Lindenberg, Germany, with a starting resolution of 600 m and a 100 km domain. The resolution is refined through several nests—potentially up to 75 m—forced by operational weather forecasts at 2.2 km resolution. We will show first case study results of the application of 3D cloud reconstruction within our virtual testbed by applying instrument simulators.