1 LANDSLIDE SUSCEPTIBILITY ASSESSMENT IN XANTHI AREA (THRACE – GREECE) BY FACTOR ANALYSIS, FUZZY MEMBERSHIP FUNCTIONS AND GIS. Alexandra Gemitzi 1* , George Falalakis 2 , Panagiotis Eskioglou 3 , Christos Petalas 1 , Vassilios Tsihrintzis 1 1, Laboratory of Ecological Engineering and Technology, Department of Environmental Engineering, Democritus University of Thrace, 67100, Xanthi, Greece 2 Thracean Gold Mining SA, e.mail address: georgefalalakis@tgm.gr 3 Aristotle University of Thessaloniki, Dept. of Forestry and Natural Environment, e.mail address: pxeskio@for.auth.gr *Corresponding author: E-mail: agkemitz@env.duth.gr ; Tel & Fax +30-25410-79393 ABSTRACT A methodology for identifying landslide prone areas and producing landslide susceptibility maps is presented, using factor analysis, fuzzy membership functions and Geographical Information Systems (GIS). Five conditioning factors were evaluated: slope angle, slope aspect, land use, geology, and topographical elevation. Fuzzy membership functions were defined for each factor using the landslide frequency data. Factor analysis provided importance weights for each one of the above conditioning factors. To obtain the final landslide susceptibility map, an overlay and index method was adopted. 1. INTRODUCTION Landslides cause dramatic effects on human lives and economy worldwide. The evaluation of landslide susceptibility is perceived as the initial stage for mitigating landslide hazards. There is a growing interest worldwide in developing robust methods for landslide susceptibility and risk mapping, and hence provide planners with tools for selecting suitable areas for development scheme implementation. Much research is focused and published on the preparation of hazard maps based on statistical modeling or on ranking and weighting environmental factors according to their assumed or expected importance in causing slope failures [1-7]. New techniques have been successfully applied in the recent years such as fuzzy logic and neural networks in various other environmental problems and also in order to evaluate landslide susceptibility [8-11]. The purpose of the present study is to develop a methodology for evaluating landslide prone areas and produce a landslide susceptibility map for the northern Xanthi area (Thrace, Greece) (Fig. 1) based on factor analysis, fuzzy logic and GIS. For this purpose, the study includes five main stages: (1) the preparation of a landslide inventory of the study area based on field studies and data from previous works; (2) the determination of controlling parameters and the evaluation of landslide frequency for each of these; (3) the fuzzification