Forecasting urban shifts post-earthquake: LULC changeanalysis in Elazığ, Turkey using ANN and Markov models

Creative Commons License

Sünbül F., Karadeniz E., Şengün M. T., Kocaoğlu M.

THE GEOGRAPHICAL JOURNAL, vol.57, no.2, pp.1-19, 2025 (SSCI)

  • Publication Type: Article / Article
  • Volume: 57 Issue: 2
  • Publication Date: 2025
  • Doi Number: 10.1111/geoj.70022
  • Journal Name: THE GEOGRAPHICAL JOURNAL
  • Journal Indexes: Social Sciences Citation Index (SSCI), Scopus, Academic Search Premier, IBZ Online, International Bibliography of Social Sciences, Periodicals Index Online, L'Année philologique, Agricultural & Environmental Science Database, American History and Life, Aquatic Science & Fisheries Abstracts (ASFA), Art Source, CAB Abstracts, Environment Index, Geobase, Historical Abstracts, Index Islamicus, Political Science Complete, Pollution Abstracts, Public Affairs Index, Veterinary Science Database, DIALNET
  • Page Numbers: pp.1-19
  • Inonu University Affiliated: Yes

Abstract

Understanding land use and land cover (LULC) dynamics in seismically activeregions is crucial for risk- informed urban planning and sustainable post-disasterrecovery. This study investigates the impact of the Mw 6.8 Elazığ earthquake (24January 2020) on LULC patterns in eastern Turkey by integrating high-resolutionSentinel- 2 satellite imagery with geographic information systems (GIS), remotesensing (RS), artificial neural networks (ANNs), and Markov chain modelling.The methodology comprises four phases: establishing a pre-earthquake baseline(2015–2019), assessing post-earthquake changes (2015–2023), analysing transitionprobabilities to identify key LULC drivers, and forecasting land-use scenarios for2030 and 2050 under seismic and non-seismic conditions. Results reveal thatseismic activity significantly accelerates urban expansion, shifting developmenttowards geologically stable zones. By 2050, artificial surfaces are projected tooccupy 54.70% of the region under seismic influence, compared to 48.87% withoutit. Agricultural land is more preserved in the seismic scenario (26.54%) than inthe non-seismic case (22.68%), while pasture and meadow areas decline sharplyto 6.18%, raising concerns for biodiversity and ecosystem services. These findingsemphasise the importance of integrating ecological considerations and seismicrisk into land-use planning frameworks. By combining multicriteria decision-making with machine learning-based forecasting, the study offers a replicableand scalable model for balancing urban growth, environmental conservation,and resilience. Framed within interdisciplinary insights from disaster resiliencetheory, urban governance, and spatial risk modelling, this research contributes tothe global discourse on sustainable urban transformation in the face of increasingnatural hazards.