Spatio-temporal analysis of earthquake occurrences using spatial statistics in a Geographic Information System (GIS) environment: A case study of Northwestern region of Iran

ABSTRACT
Spatiotemporal analysis of earthquakes plays a critical role in identifying high-risk areas, prioritizing interventions, and reducing seismic vulnerability. This study aimed to analyze seismic clusters across the northwest region of Iran using data from over 3,500 earthquake events with magnitudes of 4.0 and above, recorded between 1996 and 2024. The data were obtained from the Seismological Center of the Institute of Geophysics, University of Tehran, and analyzed using Geographic Information Systems (GIS) techniques such as Kernel Density Estimation (KDE), Getis-Ord Gi* hot spot analysis, and Moran’s I spatial autocorrelation. Results showed that the highest concentration of earthquakes occurred near major fault lines such as Tabriz, Khoy, Salmas, and Makran, with around 62% of urban settlements located within a 14-kilometer radius of these epicenters. Additionally, 75% of recorded events had focal depths below 15 kilometers, increasing the potential for severe damage in densely built urban areas. Hot spot clusters were also identified in heavily populated cities, aligning with zones of high vulnerability. The study’s novelty lies in integrating advanced spatiotemporal analytics, multidimensional vulnerability indices, and regional-scale modeling using GIS tools to develop a localized framework for seismic risk assessment. Findings highlight the need for strict construction standards, early warning systems, continuous fault monitoring, and up-to-date spatial databases. The application of multi-criteria decision-making (MCDM), Random Forest algorithms, and Artificial Neural Networks (ANN) in disaster planning is also recommended. One limitation is the lack of detailed geotechnical data at the regional scale, which future research can address.
Extended Abstract
Introduction
Earthquake vulnerability in seismically active regions such as northwestern Iran presents a persistent and multifaceted challenge for urban planning and disaster risk management. The region’s complex tectonic structure, high density of active fault systems—including the Tabriz, Salmas, and Maku faults—and rapid urbanization processes make it one of the most hazard-prone areas in the country. Despite a long history of destructive earthquakes, comprehensive spatial assessments of seismic vulnerability at the regional scale remain scarce. Most prior studies have focused on individual cities or applied single-method analyses, limiting the ability to identify macro-scale spatial patterns and formulate coordinated resilience strategies. Addressing this gap, the present study applies a spatiotemporal approach using advanced geospatial technologies to assess seismic exposure, identify high-risk urban centers, and develop a regionally adapted vulnerability model. The study is theoretically grounded in the integration of physical, environmental, and socio-economic indicators of vulnerability, and contributes to the development of spatial decision-support tools for earthquake risk reduction in Iran and similar seismic zones.
 
Methodology
This research adopts a descriptive-analytical methodology based on advanced spatial analysis techniques within a Geographic Information System (GIS) framework. The dataset includes more than 3,500 recorded earthquakes of magnitude 4.0 and above during the period 1996–2024, extracted from the Seismological Center of the Institute of Geophysics, University of Tehran. Earthquake records were geocoded and analyzed using several complementary methods. First, Kernel Density Estimation (KDE) was used to visualize the spatial intensity of seismic events and identify seismic hotspots. Second, spatial clustering analysis using the Local Moran’s I index revealed the degree of spatial autocorrelation and highlighted statistically significant clusters. Third, Getis-Ord Gi* hot spot analysis was applied to pinpoint zones with elevated seismic activity and identify cities situated within them. Buffer analysis determined the proximity of urban settlements to earthquake epicenters and was used to generate proximity-based vulnerability zones. Urban boundary data, population density, and infrastructure layers were overlaid with seismic data to assess spatial vulnerability at the city level. The resulting maps were synthesized to produce a final vulnerability model, facilitating prioritization and crisis planning.
 
Results and discussion
The KDE analysis demonstrated strong spatial clustering of earthquakes along primary fault zones, particularly in northwestern areas near Tabriz, Urmia, Khoy, and Maku. These regions exhibited the highest seismic density throughout the 28-year study period. Hot spot analysis using the Gi* index confirmed that cities like Tabriz, Urmia, and Salmas fall within statistically significant high-risk clusters. Local Moran’s I values indicated non-random distribution of seismic events, highlighting spatial dependence in fault-related earthquake occurrence. Buffer analysis revealed that approximately 62% of the 40 regional cities are located within 14 kilometers of the nearest earthquake epicenter, and thus face substantial seismic exposure. Furthermore, 75% of recorded earthquakes had focal depths less than 15 km, increasing the potential for severe surface-level damage in densely populated urban areas. Risk prioritization analysis identified Tabriz, Urmia, Maragheh, and Sarab as the most vulnerable cities, due to their proximity to fault lines, high population densities, and weak infrastructure resilience.
In comparing these findings with prior research in cities such as Tehran (Afsari et al., 2023), Sanandaj (Yariyan et al., 2020), and Istanbul (Alemdar, 2025), the current study introduces a unique contribution by combining KDE, Gi*, and Local Moran’s I in a multi-method framework. This integration enables a richer understanding of regional-scale patterns of vulnerability. Unlike single-city studies or those based solely on AHP or MCDM methods, the present research offers a holistic, geospatially-driven model that is more scalable and adaptable to large-scale disaster planning. The limitations of the study include the exclusion of earthquakes under magnitude 4.0 and the lack of detailed geotechnical and construction quality data, which should be addressed in future research. Nevertheless, the analysis sets a solid empirical foundation for further expansion.
 
Conclusion
This study conducted a spatiotemporal analysis of earthquake vulnerability in northwestern Iran using advanced GIS-based techniques and seismic data from 1996 to 2024. The results revealed significant seismic clustering near major fault zones, particularly around the Tabriz, Salmas, Khoy, and Makran faults. Kernel Density Estimation (KDE) identified areas with high seismic intensity, while Getis-Ord Gi* hotspot analysis and Moran’s I confirmed strong spatial clustering of events near fault lines and urban centers. Buffer and polygon-based risk assessments showed that approximately 62% of urban areas lie within 14 kilometers of high-magnitude earthquake epicenters. Additionally, more than 75% of recorded earthquakes occurred at depths less than 15 kilometers, increasing potential damage risks in densely populated cities such as Tabriz, Urmia, Sarab, and Maragheh. These findings highlight the urgent need for data-driven mitigation strategies in this seismically active region.
The study underscores the importance of applying multi-criteria decision-making (MCDM) models in GIS environments to prioritize at-risk zones and inform emergency planning. The results also support the use of predictive models such as Random Forest, ANN, and PROMETHEE-VIKOR, especially in data-scarce contexts. Building urban-scale spatial databases from multi-source inputs, as demonstrated in recent European studies, can further support the development of early warning and rapid response systems. Given the population exposure, structural vulnerability, and planning challenges in several cities, implementing stricter seismic codes, monitoring active faults, and improving public education are essential. Future research should integrate geotechnical data, run earthquake scenario simulations, and explore AI-based forecasting models to improve accuracy and usability. This study offers a regionally tailored yet adaptable framework for assessing seismic vulnerability and guiding urban resilience strategies.
 
Funding
There is no funding support.
 
Authors’ Contribution
Authors contributed equally to the conceptualization and writing of the article. All of the authors approved thecontent of the manuscript and agreed on all aspects of the work declaration of competing interest none.
 
Conflict of Interest
Authors declared no conflict of interest.
 
Acknowledgments
We are grateful to all the scientific consultants of this paper.