Parameterized Horizontal Shear Production in AROME at hectometric scale: evaluation over complex terrain with the TEAMx data and within the UrbanAIR project

For NWP modelling, increasing model resolution at the hectometre scale aims to improve the representation of small-scale variability for high-impact weather events. In complex terrain over orography, the increased resolution is expected to be beneficial for capturing complex, and often coupled, fluid flow processes including slope circulations and mountain waves. In urban areas, high-resolution forecasts including the detailed heterogeneous surface characteristics would allow to describe weather conditions at the district level. In the context of climate change, this would aid in the projection, planning, and adaptation of cities toward key health, environmental, or economic objectives. To these ends, however, one key challenge for current operational models is to adapt physical schemes for operations at high resolution. Indeed, at these resolutions, subgrid parameterizations can have a significant impact on the numerically represented scales, a characteristic feature of the so-called “gray-zone” issue for weather forecast models (Honnert et al. 2020).

In this context, this work focuses on the validation and improvement of AROME NWP hectometric forecasts over complex terrain. Following the work of Rogel et al. (2025), the impact of a so-called "pseudo-3D" approach for the Horizontal Shear Production (HSP) is assessed at high resolution, in complex terrain over the Alps. This parameterization extends approaches from Göger et al (2018) and Goecke et Machulskaya (2021), in order to include horizontal effects in the turbulence scheme. The sensitivity to length scale formulations on the horizontal and vertical directions is studied. Other aspects of the analysis include the surface fluxes coupling with the turbulence scheme and the impact of their parameterization toward the representation of slope circulations. Observation data from the TEAMx campaign is used to assess the performance of the model. In a second part, observations from the PANAME observation campaign (Lemonsu et al 2026) are used to evaluate AROME forecasts at 200m horizontal resolution on a heatwave case over Paris. Sensitivity with respect to surface characteristics is investigated. In particular, following the work of Bazile et al (2025), the sensitivity of the HSP parameterization to the explicit inclusion of buildings inside the orography of the model is analyzed. Results will be discussed within the context of the UrbanAIR project, and with the aim of providing accurate forcing data for Large Eddy simulation (LES) models.