Landslide susceptibility assessment in the Izmir (West Anatolia, Turkey) city center and its near vicinity by the logistic regression method


Kincal C., AKGÜN A., Koca M. Y.

ENVIRONMENTAL EARTH SCIENCES, cilt.59, sa.4, ss.745-756, 2009 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 59 Sayı: 4
  • Basım Tarihi: 2009
  • Doi Numarası: 10.1007/s12665-009-0070-0
  • Dergi Adı: ENVIRONMENTAL EARTH SCIENCES
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.745-756
  • Anahtar Kelimeler: Logistic regression, Landslide, GIS, Slope stability, Izmir (Turkey), ARTIFICIAL NEURAL-NETWORKS, REMOTE-SENSING DATA, BLACK-SEA REGION, FREQUENCY RATIO, GIS, AREA, HAZARD, MAPS, PART, VALIDATION
  • Karadeniz Teknik Üniversitesi Adresli: Evet

Özet

A landslide susceptibility assessment for Izmir city (Western Turkey), which is the third biggest city of Turkey, was performed by a logistic regression method. A database of landslide characteristics was prepared using detailed field surveys. The major landslides in the study area are generally observed in the field, dominated by weathered volcanics, and 39.63% of the total landslide area is in this unit. The parameters of lithology, slope gradient, slope aspect, distance to drainage, distance to roads and distance to fault lines were used as variables in the logistic regression analysis. The effect of each parameter on landslide occurrence was assessed from the corresponding coefficients that appear in the logistic regression function. On the basis of the obtained coefficients, lithology plays the most important role in determining landslide occurrence and distribution. Slope gradient has a more significant effect than the other geomorphological parameters, such as slope aspect and distance to drainage. Using a predicted map of probability, the study area was classified into five categories of landslide susceptibility: very low, low, moderate, high and very high. Whereas 49.65% of the total study area has very low susceptibility, very high susceptibility zones make up 11.69% of the area.