ABSTRACT / Administrative boundaries, natural obstacles, railways or major roads may contribute to the segregation of neighborhoods. However, the empirical knowledge about this issue is limited. Here, we present a methodological framework to assess the importance of barriers to urban mobility along their hierarchy, over time and across neighborhoods. Using GPS mobility data, we construct a network of blocks from the sequence of individual visits in a major European city. A community finding algorithm allows us to evaluate the fit of the neighborhoods detected in this mobility network to characteristic urban barriers by adjusting the resolution parameter of the Louvain method to decrease the dominance of mobility hubs. Using the Symmetric Area Difference index, which quantifies the overlap between the polygons separated by urban barriers and the network communities, we find that the match of detected neighborhoods and urban barriers improves as the resolution parameter increases. However, at high values of the resolution parameter, this fit gets better for lower rank administrative or road barriers than for their higher rank pairs. Although the role of urban barriers in hindering mobility became stronger over the COVID-19 restrictions, their fit with detected neighborhoods deteriorated for most analyzed barriers. Exceptions are primary roads, to which the fit of neighborhoods is stable over periods. By increasing the resolution parameter, we find emergent communities of local mobility flows that cross primary roads and a consequent optimal value of the resolution parameter. The Barrier Crossing Ratio, the fraction of barrier crossings that are network community crossings as well, demonstrates that the barriers’ impact on mobility are significantly different across urban areas. These results contribute to the ongoing discourse on urban segregation, emphasizing the importance of barriers to urban mobility in shaping neighborhood interactions and mixing.
BIO / Balázs Lengyel is an economic geographer and works on topics at the intersection of economic geography, innovation studies, and network science. He aims to understand how social interaction facilitates economic and technological progress embedded in geographical space. He is the co-director of the ANETI Lab, an interdisciplinary group co-hosted by the Hungarian Research Network and Corvinus Institute for Advanced Studies focusing on agglomeration, networks, and innovation. He is an Associate Professor at the Corvinus University of Budapest, and is a board member of the Center for Collective Learning. Balázs was a visiting scholar at the MIT Human Mobility and Networks Lab in 2016; he completed his PhD in economics at the Budapest University of Technology and Economics in 2010 and holds a master degree from the University of Szeged.