Displacement-based earthquake loss assessment for an earthquake scenario in Istanbul

I. E. Bal, H. Crowley, R. Pinho

Research output: Contribution to journalArticleAcademicpeer-review


DBELA is a Displacement-Based Earthquake Loss Assessment methodology for urban areas which relates the displacement capacity of the building stock to the displacement demand from earthquake scenarios. The building stock is modeled as a random population of building classes with varying geometrical and material properties. The period of vibration of each building in the random population is calculated using a simplified equation based on the height of the building and building type, whilst the displacement capacity at different limit states is predicted using simple equations which are a function of the randomly simulated geometrical and material properties. The displacement capacity of each building is then compared to the displacement demand obtained, from an over-damped displacement spectrum, using its period of vibration; the proportion of buildings which exceed each damage state can thus be estimated. DBELA has been calibrated to the Turkish building stock following the collection of a large database of structural characteristics of buildings from the northern Marmara region. The probabilistic distributions for each of the structural characteristics (e.g. story height, steel properties etc.) have been defined using the aforementioned database. The methodology has then been applied to predict preliminary damage distributions and social losses for the Istanbul Metropolitan Municipality for a Mw 7.5 scenario earthquake.
Original languageEnglish
Pages (from-to)12-22
JournalJournal of Earthquake Engineering
Issue numberS2
Publication statusPublished - 14 May 2008


  • damage prediction
  • displacement-based assessment
  • istanbul
  • social losses
  • urban areas
  • earthquakes


Dive into the research topics of 'Displacement-based earthquake loss assessment for an earthquake scenario in Istanbul'. Together they form a unique fingerprint.

Cite this