Open dataset of road obstructions due to collapsed buildings from the 2023 Turkey earthquakes

After large-magnitude earthquakes, collapsed buildings can disrupt transportation networks due to debris obstructing streets, sidewalks, and highways. These disruptions can critically compromise rescue activities, the mobilization of seriously injured patients, and aid distribution to the affected population. Accordingly, it is recommended the removal of obstructions during the first 48 hours after a disaster to meet emergency response needs as quickly as possible. However, the literature on seismic vulnerability has yet to deeply address the topic of road obstructions due to the limited availability of data. In this paper, we exploit several open datasets to collect about 1,400 collapsed buildings causing varying levels of road obstructions after the 2023 Turkey, earthquake sequence. The buildings are located in Hatay, Islahiye, Adiyaman, and Kahramanmaras. From visual inspection of high-resolution satellite and aerial images, the buildings were geolocated and classified according to the level of obstruction produced in the road by its debris. In the study, we considered three levels of obstruction: no obstruction, partial obstruction, and complete obstruction of a road. The number of floors was recorded from Google Street View data. Analysis of the data shows that taller buildings that experienced full collapse produced more obstructions. Thus, researchers can utilize the dataset to develop probabilistic models that predict the obstruction level according to the type of construction, the number of floors in a building, and the damage level.