Population Distribution Modelling at Fine Spatio-temporal Scale Based on Mobile Phone Data

Publikace nespadá pod Filozofickou fakultu, ale pod Přírodovědeckou fakultu. Oficiální stránka publikace je na webu muni.cz.



Druh Článek v odborném periodiku
Časopis / Zdroj International Journal of Digital Earth
Fakulta / Pracoviště MU

Přírodovědecká fakulta

WWW online verze článku
Doi http://dx.doi.org/10.1080/17538947.2018.1548654
Klíčová slova population distribution; modelling; mobile phone data; estimated human presence; emergency management
Popis Population distribution modelling can benefit many different domains, for example, transportation, urban planning, ecology or emergency management. Information about the location and number of people in an affected area is crucial for decision-makers during emergencies and crises. Mobile phone data represents relatively reliable and time accurate information on real-time population distribution, movement and behaviour. In this study, we evaluate the spatio-temporal distribution of population derived from phone data of the selected pilot area (City of Brno, Czech Republic). Analysis is based on the dataset describing the estimated human presence (EHP) with two values – visitors and transiting persons. The temporal change of data is first analysed and further processed using two methodological approaches. First, the dasymetric method is used where the building geometry and technical attributes served as a target layer. Second, the results of building level analysis are transformed into a regular grid zone of both visitors and the general EHP. Resulting spatio-temporal patterns are compared to the census data. Results demonstrate how the proposed building level dasymetric approach can improve the spatial granularity of EHP. Potential use of proposed methodology within selected emergency situations is further discussed.
Související projekty: