Multilayer networks describing interactions in urban systems: a digital twin of five cities in Spain

Networks specifying who interacts with whom are crucial for mathematical models of epidemic spreading. In the context of emerging diseases, these networks have the potential to encode multiple interaction contexts where non-pharmaceutical interventions can be introduced, allowing for proper comparisons among different intervention strategies in a plethora of contexts. Consequently, a multilayer network describing interactions in a population and detailing their contexts in different layers constitutes an appropriate tool for such descriptions. These approaches however become challenging in large-scale systems such as cities, particularly in a framework where data protection policies are enhanced. In this work, we present a methodology to build such multilayer networks and make those corresponding to five Spanish cities available. Our work uses approaches informed by multiple available datasets to create realistic digital twins of the citizens and their interactions and provides a playground to explore different pandemic scenario in realistic settings for better preparedness.