A novel rule-based approach in mapping landslide susceptibility

摘要: Despite recent advances in developing landslide susceptibility mapping (LSM) techniques, resultant maps are often not transparent, and rules barely made explicit. This weakens the proper understanding of conditioning criteria involved shaping events at local scale. Further, a high level subjectivity re-classifying scores into various classes downgrades quality those maps. Here, we apply novel rule-based system as an alternative approach for LSM. Therein, initially assembled relate landslide-conditioning factors within individual rule-sets. is implemented without complication applying logical or relational operators. To achieve this, first, Shannon entropy was employed to assess priority order uncertainty each rule corresponding Next, rule-level uncertainties were mapped used asses reliability map scale (i.e., pixel-level). A set If-Then applied convert values classes, where less guaranteed. In case study Northwest Tasmania Australia, performance proposed method assessed by receiver operating characteristics’ area under curve (AUC). Our demonstrated promising with AUC 0.934. result transparent approach, priorities state/value pixel identified. addition, can be readily accessed, interpreted, replicated. The achieved results demonstrate that beneficial derive insights LSM processes.