Key points are not available for this paper at this time.
In mountainous places, landslides pose severe environmental threats, weakening infrastructure, resulting in death and costing the economy. This article assesses landslide susceptibility and risk in the Baso Liben district of Northwestern Ethiopia using an analytical hierarchy process (AHP) and Geographic Information System (GIS)-based Frequency Ratio (FR). Eleven key environmental and geological components— height, slope, lithology, soil type, and land use—were studied. After field research and Google Earth photos, 342 landslide incidents were collated and separated into validation (30%) and training (70%) datasets. ROC curves provide a technique for analyzing the efficacy of the FR and AHP models. With an Area Under the Curve (AUC) value of 83.4%, the AHP model exhibited superior accuracy than the FR model, with an AUC value of 74.4%. Very low, low, moderate, high, and very high vulnerability are five categories defined as landslide hazard zones. The AHP model assessed 10.5% of the area as very high risk, 19.8% as high danger, 25.6% as moderate risk, 28% as medium risk, and 16.1% as very low risk. The FR model meanwhile assessed 10.16% of the area as very high risk, 21.3% as high risk, 28.9% as moderate risk, 22.5% as low risk, and 17.04% as very low risk. The results reveal that slope angle, lithology, and elevation are key factors impacting landslide vulnerability. These findings equip a practical framework for land-use planning and disaster risk reduction, providing decision-makers with appropriate instruments to help lessen landslide hazards. The research underscores the significance of combining objective data analysis with expert knowledge to enhance the accuracy and reliability of landslide susceptibility models.
Alamrew et al. (Fri,) studied this question.