作者: Hakan A. Nefeslioglu , Candan Gokceoglu , Harun Sonmez , Tolga Gorum
DOI: 10.1007/S10346-011-0267-7
关键词:
摘要: The main purpose of this study is to develop a new hazard evaluation technique considering the current limitations, particularly for shallow landslides. For purpose, Buyukkoy catchment area, located in East Black Sea Region east Rize province and south Cayeli district, was selected as area. investigations were executed four different stages. These (1) preparation temporal landslide inventory (2) assessment conditioning factors catchment, (3) susceptibility analyses (4) evaluations mapping. A total 251 landslides period 1955–2007 recognised using data sources. ‘Sampling Circle’ approach proposed define initiation mapping units evaluations. To accomplish analyses, method artificial neural networks implemented. According performance conducted training testing datasets, prediction generalisation capacities models found be very high. transform values into rates, with equation developed, taking account behaviour responsible triggering factor over time In equation, threshold value recurrence interval are independent variables. This unique property suggested allows execution more flexible dynamic assessments. Finally, technique, maps area return periods 1, 2, 5, 10, 50 100 years produced.