Interactions between urban climate and air pollution in high-resolution WRF-Chem simulations over Madrid

Urban climate can be understood as a local perturbation of the surrounding regional climate, modifying the exchange of energy, moisture, and momentum between the surface and the atmosphere. In rural areas, these exchanges are largely controlled by vegetation and soil properties, whereas in cities they are strongly altered by artificial materials, urban morphology, and anthropogenic emissions. These modifications give rise to characteristic urban phenomena, such as the urban heat island, and strongly influence the dispersion, accumulation, and transformation of air pollutants. At the same time, atmospheric pollution can affect radiative processes and feedback on urban meteorology, forming a coupled, bidirectional, and potentially nonlinear system.

In this work, the WRF-Chem model is used to investigate the interactions between urban climate and air quality over the metropolitan area of Madrid, Spain. A set of short-term simulations was performed using a one-way nested configuration, with a parent domain over the Iberian Peninsula at 5-km resolution and an inner domain over Madrid at 1-km resolution. Four main experiment families were considered: rural and urban configurations, each with and without atmospheric chemistry. Additional sensitivity experiments were conducted to assess the impact of emission strength and urban representation. Urban land cover was prescribed using the CGLC-MODIS-LCZ dataset, which provides detailed land-use information at 100-m resolution. For rural simulations, all urban categories were replaced by the dominant surrounding rural land-cover type. Anthropogenic emissions were taken from the CAMS inventory, and aerosol–radiation interactions were included, while aerosol–cloud interactions were neglected.

The results show that air pollution exerts a small but discernible impact on near-surface temperature, whereas the choice of urban parameterization strongly affects the intensity and spatial structure of the urban heat island, particularly at night. Activating urban canopy models such as BEP and BEP+BEM increases simulated urban temperatures by up to 4–5 °C relative to rural configurations, in agreement with the observations, and modifies local circulation patterns. These circulation changes, in turn, affect surface pollutant concentrations, leading to increases of 20–25% in some urban areas. While the magnitude of pollutant concentrations varies with emission strength, the relative impact of urban-induced circulation on pollutant distribution remains similar.

Overall, the results highlight the importance of accurately representing both urban morphology and emissions in coupled atmosphere–chemistry models to properly capture local climate and air-quality interactions in cities.

Acknowledgements: The authors acknowledge the ARUBA project (PID2023-149080OB-I00/MCIN/AEI/10.13039/501100011033, Ministerio de Ciencia e Innovación/Agencia Estatal de Investigación, Spain & FEDER, EU). ERL thanks her predoctoral contract FPU (FPU21/02464) to the Ministerio de Universidades of Spain.