New macroseismic intensity predictive models for Turkey


BAYRAK E., Nas M., BAYRAK Y.

Acta Geophysica, cilt.67, sa.6, ss.1483-1513, 2019 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 67 Sayı: 6
  • Basım Tarihi: 2019
  • Doi Numarası: 10.1007/s11600-019-00357-4
  • Dergi Adı: Acta Geophysica
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.1483-1513
  • Anahtar Kelimeler: Macroseismicity, Macroseismic intensity, Earthquake hazard, Earthquake early warning, Attenuation relations, Seismicity of Turkey, MODIFIED MERCALLI INTENSITY, GROUND-MOTION PARAMETERS, HISTORICAL EARTHQUAKES, FELT INTENSITY, ATTENUATION, MAGNITUDE, LOCATION, BUILDINGS, EQUATIONS
  • Atatürk Üniversitesi Adresli: Evet

Özet

In this study, we propose new attenuation relations for use in possible future macroseismicity-based analyses for Turkey. The most significant difference of the new relationships is that they are specifically designed for use in the full extent of Turkey. Besides, this paper supplies an extensive macroseismic database larger than ever collected for Turkey. To this end, a modifiable MATLAB program, which is fully presented in the electronic supplement to this paper, was written to analyze the collected isoseismal maps that were first geo-rectified and then gridded by dividing them into 0.02 degrees arrays. The epicentral distances belonging to each point intersected with the pertinent isoseism were compiled in an event-based log. Then, all subsets were merged into a single dataset, which is presented in the electronic supplement. Required strong ground motion parameters were taken from an improved earthquake catalog recently given by Kadirioglu et al. (B Earthq Eng 1:2, 2016). For modeling the relations, the early relations were mostly selected as patterns for our candidate attenuation models. Of all candidate models, the statistically significant ones were individually tested whether or not they were able to detect the best fitting model to our database. The predicted error margins of the proposed models were compared to those of early models using data-driven statistical tests. In conclusion, considering the usability limits, the estimation capabilities of proposed relationships were found to be useful to some extent than those of the early models developed for Turkey.