Prof. Dr. Candan GOKCEOGLU graduated from the Department of Geological Engineering, Hacettepe University in 1989. He became a professional engineer and did his Ph.D. at the Institute of Science and Technology, Hacettepe University between 1993-1997. He became associate professor in 2003, and full professor in 2009. Prof. Dr. Candan GOKCEOGLU is currently the Dean of Engineering Faculty of Hacettepe University and the Chairman of the Executive Board Hacettepe Technopolis.
He has written more than 130 scientific papers published in leading academic journals on landslides, natural hazards, rock mass characterization, engineering geology, and slope stability. He is the Associate Editor of Computers & Geosciences, Landslides and Arabian Journal of Geosciences, and the Editorial Board Member of Engineering Geology. Also, he has worked as guest editor of several special issues of leading academic journals. He was awarded by Scientific and Technological Council of Turkey due to his contribution to the international literature about regional landslide susceptibility and hazard assessment methodologies.
Possible contribution of geo-information technologies to prediction of run out distances of landslides
Pressure on nature has been increasing with human population growth and climate change. One of the most typical results of this pressure is the rise in losses sourced from landslides. Although serious success on regional landslide susceptibility assessment has been achieved, the assessment of the risks remains a challenging problem. Two fundamental difficulties must be overcome to obtain a plausible regional landslide risk assessment and its map. One of these difficulties is to obtain the time of occurrence while the other one is the determination of run out distance. Prediction of time of occurrence depends on intrinsic characteristics of slope-forming material, magnitude and intensity of triggering events. Additionally, threshold values of triggering events must be known. Some empirical and analytical approaches for prediction of time of landslide occurrence have been investigated and reasonable results have been obtained. However, prediction of run out is a more complex problem because run out distance depends on not only intrinsic characteristics of slope-forming material but also topography, season, surface water bodies, land use and land cover, etc. In addition, due to large extension of displaced material, it is very difficult to describe the run out by field observations. For this reason, the use of geo-information technologies for determination of run out distance may open a great horizon for landslide risk assessments.
Considering the increasing variety and possibilities in sensor and platform technologies for Earth observation coupled with photogrammetry and geospatial analysis methods, it can be said that the complexity of the run-out distance determination problem can be improved by employing:
- geodata obtained with high spatial resolution and temporal frequency from airborne and spaceborne platforms as well as volunteer contributions;
- fusion methods for multi-temporal and multi-sensor data;
- automated feature extraction methods;
- automatic 3D reconstruction, matching, and spatiotemporal analysis of dense digital terrain and surface models;
- prediction of future risks by utilizing multi-temporal data and machine learning algorithms;
- additional human interpretation through 3D visualization environments.
Accurate determination of amount of displaced material and run out distances are extremely important for empirical and analytical solutions of this problem. Additionally, these solutions may help for producing robust regional risk assessments. Consequently, in this comprehensive review, the difficulties and uncertainties encountered during regional landslide risk assessment studies are given and possible contributions of geo-information technologies to eliminate these problems are discussed.