A GIS based landslide susceptibility mapping using machine learning and alternative forest road routes assessment in protection forests

Forestry activities should be carried out within the purview of sustainable forestry while reaping benefits forestry. Accordingly, construction forest roads through forests carefully planned, especially in protection forests. Forest areas Turkey are generally widespread mountainous and high sloping that susceptible to landslides-landslide susceptibility is one most important criteria for selection protected As such, it evaluate detailed applicable alternatives regarding special private The aim this study determine alternative routes use geographic information systems (GIS), particularly with landslide susceptibility. To end, a map (LSM) was created using logistic regression (LR) random (RF) modeling methods, which widely used machine learning (ML). Two models highest receiver operating characteristic (ROC) area under curve (AUC) values were selected, ten factors (slope, elevation, lithology, distance road, fault, river, curvature, stream power index, topographic position wetness index) used. best LSM method AUC. AUC value 90.6% RF approach 80.3% LR approach. generated LSMs calculated cost path analysis. It hoped landslides road determined approaches techniques will benefit planning as well plan decision makers.