Makine öğrenme teknikleri ile heyelan duyarlılık haritalarının üretilmesi: Hopa (Artvin) örneği
[ X ]
Tarih
2023
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Artvin Çoruh Üniversitesi
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Türkiye'de heyelanların en fazla görüldüğü bölge Doğu Karadeniz Bölgesidir. Doğu Karadeniz Bölgesindeki Trabzon, Rize, Giresun ve Artvin illerinde meydana gelen heyelanlar Türkiye'de meydana gelen heyelanların yaklaşık %20'sini oluşturmaktadır. Heyelanlar, Artvin gibi engebeli topoğrafyaya sahip yerleşimlerde birincil doğal afet türü olarak ön plana çıkmaktadır. Artvin'in kıyı kesimlerinde, özellikle aşırı yağışların tetiklediği heyelanlar can kayıplarına, altyapı ve üst yapı hasarlarına, doğal çevrede tahribata ve ciddi ekonomik zararlara neden olmaktadır. Bu çalışmada, rastgele orman (Random Forest - RF) ve aşırı gradyan artırma (Extreme Gradient Boosting - XGBoost) algoritmaları kullanılarak Artvin'in Hopa ilçesinin heyelan duyarlılık haritalarının üretilmesi amaçlanmıştır. Çalışmada, heyelanların meydana gelmesinde etkili olduğu belirlenen 10 faktör (litoloji, eğim, arazi örtüsü, yükseklik, bakı, eğrilik, topografik nemlilik indeksi, fay hatlarına, yola ve drenaj ağlarına yakınlık) kullanılmıştır. Çalışmada kullanılan makine öğrenmesi modellerinin performansı, alıcı işlem karakteristik (Receiver Operating Characteristic - ROC) eğrisi ve eğri altında kalan alan (area under the ROC Curve - AUC) yaklaşımı kullanılarak değerlendirilmiştir. Değerlendirme sonucunda XGBoost algoritmasının, RF algoritmasına göre daha iyi performans gösterdiği belirlenmiştir. Ayrıca, çalışma bölgesinde heyelanların meydana gelmesinde en etkili faktörlerin, sırasıyla, yükseklik, fay hatlarına yakınlık, eğim, litoloji ve arazi örtüsü olduğu belirlenmiştir. Eğrilik ise en az etkili ya da en önemsiz faktör olarak belirlenmiştir. Sonuç olarak, XGBoost algoritması ile üretilen heyelan duyarlılık haritasının, bölgede heyelan zararlarını azaltmak için karar vericilere yol gösterebileceği sonucuna varılmıştır.
The region where landslides are most common in Turkey is the Eastern Black Sea Region. The landslides occurring in the provinces of Trabzon, Rize, Giresun and Artvin in the Eastern Black Sea Region constitute approximately 20% of the landslides occurring in Turkey. Landslides stand out as the primary natural disaster type in settlements with rugged topography such as Artvin. In the coastal areas of Artvin, landslides triggered by heavy rainfall cause loss of life, infrastructure and superstructure damage, destruction of the natural environment and serious economic damage. In this study, it is aimed to produce landslide susceptibility maps of Hopa district of Artvin by using Random Forest (RF) and Extreme Gradient Boosting (XGBoost) algorithms. In the study, 10 factors (lithology, slope, land cover, elevation, aspect, curvature, topographic wetness index, proximity to fault lines, road and drainage networks) were used, which were determined to be effective in the occurrence of landslides. The performance of the machine learning models used in the study was evaluated using the Receiver Operating Characteristic (ROC) curve and the area under the ROC Curve (AUC) approach. As a result of the evaluation, it was determined that the XGBoost algorithm performed better than the RF algorithm. In addition, it has been determined that the most effective factors in the occurrence of landslides in the study area are elevation, proximity to fault lines, slope, lithology and land cover, respectively. Curvature was determined as the least effective or least important factor. As a result, it was concluded that the landslide susceptibility map produced by the XGBoost algorithm can guide the decision makers to reduce the landslide damages in the region.
The region where landslides are most common in Turkey is the Eastern Black Sea Region. The landslides occurring in the provinces of Trabzon, Rize, Giresun and Artvin in the Eastern Black Sea Region constitute approximately 20% of the landslides occurring in Turkey. Landslides stand out as the primary natural disaster type in settlements with rugged topography such as Artvin. In the coastal areas of Artvin, landslides triggered by heavy rainfall cause loss of life, infrastructure and superstructure damage, destruction of the natural environment and serious economic damage. In this study, it is aimed to produce landslide susceptibility maps of Hopa district of Artvin by using Random Forest (RF) and Extreme Gradient Boosting (XGBoost) algorithms. In the study, 10 factors (lithology, slope, land cover, elevation, aspect, curvature, topographic wetness index, proximity to fault lines, road and drainage networks) were used, which were determined to be effective in the occurrence of landslides. The performance of the machine learning models used in the study was evaluated using the Receiver Operating Characteristic (ROC) curve and the area under the ROC Curve (AUC) approach. As a result of the evaluation, it was determined that the XGBoost algorithm performed better than the RF algorithm. In addition, it has been determined that the most effective factors in the occurrence of landslides in the study area are elevation, proximity to fault lines, slope, lithology and land cover, respectively. Curvature was determined as the least effective or least important factor. As a result, it was concluded that the landslide susceptibility map produced by the XGBoost algorithm can guide the decision makers to reduce the landslide damages in the region.
Açıklama
Fen Bilimleri Enstitüsü, Harita Mühendisliği Ana Bilim Dalı
Anahtar Kelimeler
Jeodezi ve Fotogrametri, Geodesy and Photogrammetry
