1 Application of frequency ratio and logistics regression methods for landslide susceptibility mapping in the Dessie area, Ethiopia Tesfaye Chala Korma 1 , Shambel Beyene Mituku 1 , Azemeraw Wubalem 2 and Alemu Yenehun Beyene 1,3 1 Department of Geology, School of Earth Sciences, Bahir Dar University, P.O. Box 79, Bahir Dar, Ethiopia 2 Department of Geology, College of Natural and Computational Science, University of Gondar, P.O. Box 196, Gondar, Ethiopia 3 Laboratory for Applied Geology and Hydrogeology, Department of Geology, Ghent University, Ghent, Belgium *Corresponding author: E-mail address: tesf.chala2@gmail.com Abstract Dessie area is one of the landslide-prone areas in the Ethiopian highlands where frequent landslide incidents have been documented that disrupt societal well-being and cause infrastructure damage. To fix this reoccurring problems, landslide factor evaluation and landslide susceptibility mapping were performed using the Frequency Ratio (FR) and Logistic Regression (LR) models. For this purpose, 130 landslide points were mapped using detailed fieldwork and Google Earth Imagery analysis, then divided into training (70%) and validation (30%) datasets under the ArcGIS tools of subset. Nine landslide causative factors (aspect, slope, curvature, elevation, lithology, land use, rainfall, distances from road and river) and training landslide datasets were combined using GIS tools to calculate the weight of landslide factor class. After the weight of each factors determined, landslide susceptibility Indexes (LSIs) were generated by sum up all weighted landslide factors using raster calculator in ArcGIS 10.8 tool which were divided into five landslide susceptibility zones of very low, low, moderate, high, and very high. Finally, the resulting maps have been validated by using the Receiver Operating Characteristic (ROC) curve. Even though both models show excellent performance, FR model (93%) showed more accurate compared to the LR model (91.1%). Therefore, it is recommended that landslide susceptibility map produced by FR should be used as an input information in land use planning and various construction projects in the area. Keyword: Dessie area, Landslide, landslide susceptibility, Frequency Ratio, Logistic Regression 1 Introduction In mountainous region, landslides have become one of the most destructive geological hazards, causing losses of life, damages on property and infrastructure (Pan et al. 2008; Mersha and Meten 2020). According to recent studies, landslides inflict an estimated 4 billion US dollars in projected property damage annually and more than 1,000 fatalities worldwide (Lee and Pradhan 2007; Solaimani et al. 2013). Landslide occurrences are common in Ethiopian highlands due to complex geomorphological processes, geological conditions, hydrology, weathering processes and poor land use practices (Woldearegay 2013; Mersha and Meten 2020). Rainfall and anthropogenic activities are the main factors triggering frequent landslide in Dessie and its surrounding areas (Vařilová et al. 2015). Dessie area is known for its frequent landslide occurrences due to a combination of different triggering factors. According to different scientific literatures, the landslides in the study area are triggered by improper land use practices (construction and toe-cutting slope), rainfall, East African Rift System-related earthquakes and active geomorphic processes (Abebe et al. 2010; Mebrahtu et al. 2021). Currently, as the population of Dessie town and its surrounding is rising, constructions and settlements are increasing in the landslide-prone areas. These cause damage to infrastructure (roads, bridges, buildings, and power lines) and other properties (Fig. 4). According to field observation and resident witnesses, from 2015 to 2021, more than 280 houses and roads were displaced specially in newly built areas (e.g., Agere Gizat, Pepsi, Tekaom, Pissa, Bomba, along Kutaber road section, and Dessie to Kombolcha road section due to over-excavation) (Fig. 4 C and D). Therefore, it is important to assess factors that play a significant role in landslide incidence and develop landslide susceptibility maps to minimize the socioeconomic impact on local communities.