First dynamic national-scale population map based on mobile phone data

Population distribution maps are of vital importance, for instance for assessing the impact of natural disasters, conflicts or epidemics. Up to now, such maps have generally been based on census data, meaning that their contemporariness is limited by the low frequency and low spatial resolution of such censuses.

Researchers from the ULB's Biological Control and Spatial Ecology Lab, part of the ULB Faculty of Science's Interfaculty School of Bioengineering, and from the Institute of Information and Communication Technologies, Electronics and Applied Mathematics (ICTEAM), part of the UCL's Louvain School of Engineering, have succeeded in building dynamic population maps using anonymised mobile phone data.

Published in the 27 October 2014 issue of PNAS, this study shows that the data collected from mobile phone operators can be used to produce precise and detailed population distribution maps on a national scale and for a given period, while at the same time ensuring data privacy.

The method presented could be of particular use to developing countries where census data is rare, out-of-date or not very reliable. It can also be used to study variations in population density over different time-spans, for instance a day, a week or a month, allowing human mobility to be mapped. According to this study, facilitating access to anonymised mobile phone data could allow the rapid and low-cost production of population distribution maps in emergency situations where data is lacking.

YouTube link: http://youtu.be/qsUDH5dUnvY