Towards a digital twin of the urban atmosphere for decision support

Femke C. Vossepoel, Sam Pickard, Maarten van Reeuwijk, Marion Samler, Natalie Theeuwes, Nele Veldeman

2025

Abstract

Since the atmosphere is omnipresent, it plays a vital role in the complex interactions with other Earth and societal systems. This is especially true in urban areas, where over 4 billion people currently reside, a number expected to grow to 70% of the global population by 2050[1]. Human interventions in the urban environment – including spatial planning, the development of green, blue and grey infrastructure, and mobility choices – interact with climate and meteorological variables to influence the health and well-being of urban dwellers and the liveability of our cities. Such complexity makes it challenging for existing infrastructures to provide robust evidence to support stakeholders who make these decisions. Thus, a digital twin tailored to stakeholder needs that brings together internationally disparate expertise and high-quality research infrastructures would be highly beneficial.UrbanAIR[2], started in January 2025, strives to develop such a digital twin that supports urban decision-makers as they contend with design dilemmas stemming from the impacts of climate change and air quality on citizen health and socio-economic wellbeing. It is a highly interdisciplinary consortium, bringing together computer scientists, environmental modellers, communication specialists, social scientists and software developers. On the technical side, UrbanAIR will include a cascade of atmospheric models, ranging from the global scale, linking via the mesoscale to very high-resolution simulators at the neighbourhood or street level. By starting from the perspective of the decision-maker and fostering co-creation, we will configure the models to generate scenarios that address key dilemmas and support a balanced evaluation of decision criteria. In this presentation, we will present our plans for integrating the different simulation and decision-making components. We will pay specific attention to the integration of observations into the simulator and to uncertainty quantification through emerging data assimilation and machine-learning techniques.The resulting dynamic, user-friendly workflow and tools will be integrated into the Destination Earth infrastructure[3], empowering municipalities and industries to make informed choices on urban planning and design to better prepare for climate change adaptation and hazard exposure. By testing the tools in a variety of real-world settings, the research infrastructure of UrbanAIR will pave the way for effective climate adaptation and hazard mitigation in a more general sense, transforming urban planning and design into a proactive, tool-based, approach for a safer, healthier, and more resilient future. [1] World Bank, https://www.worldbank.org/en/topic/urbandevelopment/overview, accessed 13 January 2025 [2] UrbanAIR is part of the work programme HORIZON-INFRA-2024-TECH-01-03: New digital twins for Destination Earth. [3] https://destination-earth.eu/