A novel activity space approach to discover displacement patterns via mobile phone data: an analysis of the 2023 Türkiye-Syria earthquakes


Creative Commons License

Aydoğdu B., Danış Şenyüz A. D., Bilgili Ö., Yıldızcan C., Yağcıklı S. N., Güneş S., ...Daha Fazla

EPJ DATA SCIENCE, sa.14, ss.1-28, 2025 (SCI-Expanded, SSCI, Scopus)

  • Yayın Türü: Makale / Tam Makale
  • Basım Tarihi: 2025
  • Doi Numarası: 10.1140/epjds/s13688-025-00572-8
  • Dergi Adı: EPJ DATA SCIENCE
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Social Sciences Citation Index (SSCI), Scopus, Agricultural & Environmental Science Database, Compendex, INSPEC, Directory of Open Access Journals
  • Sayfa Sayıları: ss.1-28
  • Açık Arşiv Koleksiyonu: AVESİS Açık Erişim Koleksiyonu
  • Galatasaray Üniversitesi Adresli: Evet

Özet

This study introduces the novel Activity Space Approach (ASA) for measuring disaster-induced displacement patterns using mobile call detail records (CDR), where we explore shifting the focus in displacement detection from home locations to habitual living spaces. We apply our method to analyze the February 2023 Türkiye-Syria earthquakes, which affected over 14 million Turkish citizens and 1.7 million Syrian refugees within Türkiye. Using anonymized and hourly aggregated CDR data from 127,700 individuals, complemented with insights from qualitative fieldwork conducted in the regions affected by the earthquakes, we show that the proposed approach overcomes the main limitations of traditional home location methods and provides more granular spatial insights into displacement patterns. By incorporating measurements of urbanization and infrastructure damage, we illustrate how post-disaster mobility shows variation among locals and refugees, given their pre-existing socioeconomic vulnerabilities and unequal capacities to respond. Our findings demonstrate that, while Turkish citizens were able to evacuate more swiftly and over longer distances, Syrian refugees experienced slower and more spatially constrained displacements, often toward institutional settings such as camps, reflecting legal precarity and constrained mobility options. This comparative perspective underscores the importance of recognizing and mapping variations in displacement experiences across different population segments. Consequently, ASA can inform more targeted short-term policies and support more inclusive long-term recovery planning.

This study introduces the novel Activity Space Approach (ASA) for measuring disaster-induced displacement patterns using mobile call detail records (CDR), where we explore shifting the focus in displacement detection from home locations to habitual living spaces. We apply our method to analyze the February 2023 Türkiye-Syria earthquakes, which affected over 14 million Turkish citizens and 1.7 million Syrian refugees within Türkiye. Using anonymized and hourly aggregated CDR data from 127,700 individuals, complemented with insights from qualitative fieldwork conducted in the regions affected by the earthquakes, we show that the proposed approach overcomes the main limitations of traditional home location methods and provides more granular spatial insights into displacement patterns. By incorporating measurements of urbanization and infrastructure damage, we illustrate how post-disaster mobility shows variation among locals and refugees, given their pre-existing socioeconomic vulnerabilities and unequal capacities to respond. Our findings demonstrate that, while Turkish citizens were able to evacuate more swiftly and over longer distances, Syrian refugees experienced slower and more spatially constrained displacements, often toward institutional settings such as camps, reflecting legal precarity and constrained mobility options. This comparative perspective underscores the importance of recognizing and mapping variations in displacement experiences across different population segments. Consequently, ASA can inform more targeted short-term policies and support more inclusive long-term recovery planning.