Evaluation of Parameters Affecting Earthquake Damage Using a GIS-based Random Forests Machine Learning Model: The Case of the 6 February 2023 Kahramanmaras Earthquakes in Türkiye

Emre Özşahin; Mikayil Öztürk

Makale detayları

Evaluation of Parameters Affecting Earthquake Damage Using a GIS-based Random Forests Machine Learning Model: The Case of the 6 February 2023 Kahramanmaras Earthquakes in Türkiye

Yazarlar:

Emre Özşahin | Mikayil Öztürk

Yayıncı:

Istanbul University

Ana Başlıklar:

Fiziksel Coğrafya ve Çevre Jeolojisi (Diğer)

Dergi:

Coğrafya Dergisi Issue 49, pp.43-63

Belge Tipi:

Makale

Yayın Tarihi:

12/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001338728

pISSN:

1305-2128

Anahtar kelimeler:

Özet:

 Türkiye is a geographical feature with intense seismic activity due to its tectonic features. Despite such a high earthquake risk, the evaluation of parameters affecting earthquake damage is still very inadequate in Türkiye. The aim of this study was to evaluate the parameters affecting earthquake damage in the 6 February 2023 Kahramanmaras earthquake, which caused the highest number of casualties in the history of the Republic of Türkiye. Therefore, data were produced to understand the differences in the behavior of structures in the case of an earthquake hazard in different parts of Türkiye. The study used sample data from 198,634 buildings with varying types of structural damage in residential areas where the earthquake had been felt. The relationship between these data and key factors causing structural damage was analyzed using a Geographic Information Systems GIS-based Random Forests RF Machine Learning ML model. As a result of this study, it was understood that the 6 February 2023 Kahramanmaras earthquakes caused structural damage as a result of different combinations of building age, local soil conditions, distance to fault lines, distance to the epicenter, ground slip velocity, maximum ground velocity, and soil liquefaction effect factors.

 Türkiye is a geographical feature with intense seismic activity due to its tectonic features. Despite such a high earthquake risk, the evaluation of parameters affecting earthquake damage is still very inadequate in Türkiye. The aim of this study was to evaluate the parameters affecting earthquake damage in the 6 February 2023 Kahramanmaras earthquake, which caused the highest number of casualties in the history of the Republic of Türkiye. Therefore, data were produced to understand the differences in the behavior of structures in the case of an earthquake hazard in different parts of Türkiye. The study used sample data from 198,634 buildings with varying types of structural damage in residential areas where the earthquake had been felt. The relationship between these data and key factors causing structural damage was analyzed using a Geographic Information Systems GIS-based Random Forests RF Machine Learning ML model. As a result of this study, it was understood that the 6 February 2023 Kahramanmaras earthquakes caused structural damage as a result of different combinations of building age, local soil conditions, distance to fault lines, distance to the epicenter, ground slip velocity, maximum ground velocity, and soil liquefaction effect factors.

 Türkiye is a geographical feature with intense seismic activity due to its tectonic features. Despite such a high earthquake risk, the evaluation of parameters affecting earthquake damage is still very inadequate in Türkiye. The aim of this study was to evaluate the parameters affecting earthquake damage in the 6 February 2023 Kahramanmaras earthquake, which caused the highest number of casualties in the history of the Republic of Türkiye. Therefore, data were produced to understand the differences in the behavior of structures in the case of an earthquake hazard in different parts of Türkiye. The study used sample data from 198,634 buildings with varying types of structural damage in residential areas where the earthquake had been felt. The relationship between these data and key factors causing structural damage was analyzed using a Geographic Information Systems GIS-based Random Forests RF Machine Learning ML model. As a result of this study, it was understood that the 6 February 2023 Kahramanmaras earthquakes caused structural damage as a result of different combinations of building age, local soil conditions, distance to fault lines, distance to the epicenter, ground slip velocity, maximum ground velocity, and soil liquefaction effect factors.

 Türkiye, tektonik özellikleri nedeniyle sismik aktivitenin yoğun olduğu bir coğrafyada yer almaktadır. Bu kadar yüksek deprem riskine rağmen, deprem hasarını etkileyen parametrelerin değerlendirilmesi Türkiyede hala çok yetersizdir. Bu çalışmanın amacı, Türkiye Cumhuriyeti tarihinde en fazla can kaybına neden olan 6 Şubat 2023 Kahramanmaraş depremlerinde deprem hasarını etkileyen parametreleri değerlendirmektir. Bu nedenle, Türkiyenin farklı bölgelerinde deprem tehlikesi durumunda yapıların davranışlarındaki farklılıkları anlamak için veri üretilmiştir. Çalışmada, depremin hissedildiği yerleşim bölgelerinde farklı yapısal hasar türlerine sahip 198.634 binadan alınan örnek veriler kullanılmıştır. Bu veriler ile yapısal hasara neden olan temel faktörler arasındaki ilişki Coğrafi Bilgi Sistemleri CBS tabanlı Rastgele Ormanlar RO Makine Öğrenimi MÖ modeli kullanılarak analiz edilmiştir. Çalışma sonucunda, 6 Şubat 2023 Kahramanmaraş depremlerinin bina yaşı, yerel zemin koşulları, fay hatlarına uzaklık, merkez üssüne uzaklık, zemin kayma hızı, maksimum zemin hızı ve zemin sıvılaşma etkisi faktörlerinin farklı kombinasyonları sonucunda yapısal hasara neden olduğu anlaşılmıştır.

Makale detayları

Evaluation of Parameters Affecting Earthquake Damage Using a GIS-based Random Forests Machine Learning Model: The Case of the 6 February 2023 Kahramanmaras Earthquakes in Türkiye

Yazarlar:

Emre Özşahin | Mikayil Öztürk

Yayıncı:

Istanbul University

Ana Başlıklar:

Fiziksel Coğrafya ve Çevre Jeolojisi (Diğer)

Dergi:

Coğrafya Dergisi Issue 49, pp.43-63

Belge Tipi:

Makale

Yayın Tarihi:

12/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001338728

pISSN:

1305-2128

Earthquake

Earthquake damage

GISDeprem

Deprem hasarı

CBS

Sayı

Makale detayları

Evaluation of Parameters Affecting Earthquake Damage Using a GIS-based Random Forests Machine Learning Model: The Case of the 6 February 2023 Kahramanmaras Earthquakes in Türkiye

Yazarlar: Emre Özşahin | Mikayil Öztürk

Yayıncı: Istanbul University

Ana Başlıklar: Fiziksel Coğrafya ve Çevre Jeolojisi (Diğer)

Dergi: Coğrafya Dergisi Issue 49, pp.43-63

Belge Tipi: Makale

Yayın Tarihi: 12/2024

Belge Dili: İngilizce

Document ID (Org No.): 2001338728

pISSN: 1305-2128

Earthquake

Earthquake damage

GISDeprem

Deprem hasarı

CBS

Sayı

Yazarlar:

Emre Özşahin | Mikayil Öztürk

Yayıncı:

Istanbul University

Ana Başlıklar:

Fiziksel Coğrafya ve Çevre Jeolojisi (Diğer)

Dergi:

Coğrafya Dergisi Issue 49, pp.43-63

Belge Tipi:

Makale

Yayın Tarihi:

12/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001338728

pISSN:

1305-2128