Evaluation of bicycle sharing scheme data as a proxy for cycling mobility - How COVID-19 measures influenced cycling in Paris

The use of bicycles can provide myriad benefits to society, especially in crowded urban centres where other modes of transport are at or near capacity. However, integrating cycling into policy and planning requires more comprehensive data about their use in space and time. Current approaches, using sparse networks of counters provide one possible route to more comprehensive data. In this paper we investigate another, using data collected from Paris' bicycle sharing system to explore use during 2020. We chose 2020 as a test year because the use of bicycles was strongly influenced by the COVID-19 pandemic, allowing us to explore how bicycle use adapted to both legal and environmental influences. We used interactive visualization to allow hypothesis generation and data exploration, before analysing bicycle use as a function of weather and COVID-19 restrictions. Our results show that bicycle sharing system data and Paris' counters both capture very similar behaviour patterns, and therefore bicycle sharing system data are a reliable proxy for overall cycling behaviour, providing finer spatial granularity than existing sparse counter networks. Seasonally, precipitation influenced bicycle use more strongly in 2020 than COVID-19 measures.