Enhancing Microscale Urban Weather Simulations: The Role of High-Resolution Land Cover and Surface Characteristics in Turbulence-Resolving LESs for Murcia, Spain

The influence of surface forcings driven by realistic land cover and surface properties remains underexplored in microscale turbulence-resolving weather simulations, especially within urban settings. Atmospheric models often rely on coarse surface property datasets (typically ranging from 500 m to 10 km), but their low resolution frequently fails to adequately capture critical local surface heterogeneities. Building-resolving large-eddy simulations (LESs) that incorporate detailed land cover and surface characteristics offer a powerful framework for investigating how urban wind speed and turbulence patterns respond to variations in land cover and surface characteristics.

This research examines the significant effects of varying land cover classifications on microscale weather simulations, considering both granularity and resolution, along with other surface characteristics. A series of detailed and comprehensive experiments were carried out for the urban-rural interface of Murcia (Spain) using NCAR-RAL’s GPU-accelerated FastEddy^® LES model coupled to WRF. Three case studies over 4-hour periods with distinctive meteorological conditions were used to test different combinations of land use data, roughness length values, urban parametrizations, and driving mesoscale skin forcings. These resulted in a total of 60 LESs with a grid spacing of 10 m spanning a wide range of surface conditions. These simulations incorporated high-quality land cover and soil data, more realistic roughness length estimates, and the implementation of a local terrain smoothing method and a dynamic thermal roughness length parameterization in FastEddy^®. These non-standard practices were designed to enhance the skill of the simulations.

The results reveal that configuration details, such as land cover, play a critical role in both mesoscale and microscale simulations, significantly influencing surface fluxes and turbulence generation. While simply increasing the resolution of land cover data produces minimal changes, particularly in rural settings, incorporating more realistic surface property values leads to substantial differences even when resolution remains unchanged. These findings highlight the huge importance of detailed surface characteristics in improving microscale weather predictions.

Acknowledgements: The authors acknowledge the ECCE project (PID2020-115693RB-I00) of the Ministerio de Ciencia e Innovación/Agencia Estatal de Investigación (MCIN/AEI/10.13039/501100011033). ERL thanks her predoctoral contract FPU (FPU21/02464) to the Ministerio de Universidades of Spain.