Uncovering temporal changes in Europe’s population density patterns using a data fusion approach
Detailed information on the spatial distribution of population is essential for many applications: the management of risks related to natural, technological, or epidemiological hazards, the planning and development of transport, energy and social infrastructure, but also for a multitude of business applications. It allows governments and companies to improve the services and products they offer to citizens and costumers.
However, most of the available data reflect night-time population distribution only, as they are based on place of residence statistics, ignoring the spatial mobility of population in daily and annual cycles between locations of housing, employment, study or leisure. The emergence of mobile phone technology allowed mobile network operators to track individual users and produce maps of population distribution continuously in time. However, these data are hindered by various technical constraints, and pose privacy concerns too.
In a study by JRC, the authors propose and apply an approach to assess and map population distribution taking into account the variation in the daily and seasonal cycles. The approach is based on the fusion of multiple available and (mostly) free data sources, ranging from statistics on the size of various population groups (e.g. residents, workers, students, tourists, unemployed, etc), and the location of residences, as well as economic, education and leisure activities from very detailed maps. This approach allowed, for the first time, the creation of fine-grained (1 x 1 km), multi-temporal population distribution maps at European scale, comparable across countries, while avoiding data sources that rely directly on the activity of individuals. The newly developed dataset allowed the authors to study the spatial and temporal structure of cities.
This unprecedented effort to map spatio-temporal population in a privacy-concerned manner will not only allow various applications but also set an example for future research.