Sensitivity analysis of Planetary Boundary Layer height parameterization in the ICON - CHIMERE modeling chain

The Planetary Boundary Layer height (PBLH) is a fundamental parameter governing the dispersion of pollutants and, consequently, the accuracy of Air Quality (AQ) modeling and forecasting. As Arpae-SIMC provides daily AQ forecasts for the Emilia-Romagna region  and over the entire Italian domain, which are also employed as boundary conditions by regional and national stakeholders, a comprehensive sensitivity analysis was performed to evaluate the performance of the PBLH parametrization.

In the current operational workflow, ICON-2I - the implementation over Italy of the ICON model running at 2.2 km - provides meteorological forcing for the CHIMERE AQ model. This study compares the performance of the native CHIMERE PBL scheme against a novel parametrization developed by DWD for ICON. The novel approach refines the description of the boundary layer across different stability regimes, from stable and near-neutral conditions to unstable convective states. These parametrizations are validated against radiosounding data and Automated Lidar Ceilometers (ALC) as ground truth. The analysis accounts for the diverse Italian topography, ranging from complex mountainous terrain to flat areas characterized by stagnant, near-zero wind speed conditions. The ALC data are provided by the CNR-ISAC, thanks to a measurement campaign in Bologna.

Preliminary results for summer 2024 in downtown Bologna, obtained by comparing the various parametrizations against ALC measurements, indicate that the novel ICON scheme significantly improves PBLH estimation during daytime convection by reducing systematic overestimation. Conversely, the native CHIMERE convective scheme exhibits superior performance during the sunrise transition, as it better captures the early-morning growth of the PBL compared to the ICON scheme. Generally, the daytime PBLH values yielded by the ICON configuration are lower than those produced by the native CHIMERE scheme. Analysis of nocturnal regimes reveals that the native CHIMERE method for stable conditions leads to significant overestimation during periods of atmospheric stability coupled with low-level jets. In contrast, the novel approach integrated in the ICON framework provides more reliable results under these conditions.

These findings highlight the capability of the updated PBL scheme to provide a more accurate and consistent characterization of the boundary layer within the AQ modeling framework.