Abstract

Torrential hazards are omnipresent in the alpine regions, as it frequently causes damage to infrastructures. In some cases, even people's lives are endangered. The classification of these processes takes place according to factors like sediment concentration and flow behaviour and ranges from fluvial process types, including water floods and fluvial sediment transport processes, to fluvial mass movements such as debris flows. Following the hypothesis of this study, a context exists between basic geomorphological disposition parameters and potential dominant flow process types in a steep headwater catchment.Thus, examined catchments were selected based on a historical event documentation of torrential events in the Austrian Alps. In total, 84 catchments could be analysed, and 11 different morphometric parameters were considered. To predict the dominant torrential process type within a catchment, a naive Bayes classifier, a decision tree model, and a multinomial regression model was trained against the compiled geomorphological disposition parameters. All models as well as their combination were compared. Based on bootstrapping and complexity, we present the classification model with the lowest prediction error for our data that might help to identify the most likely torrential process within a considered catchment.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call