Regional Mountain Conference

Modelling debris flows at the basin scale is a formidable challenge that has recently been addressed by an increasing number of studies and tools, such as the TRENT2D–MBRR model, capable of simulating mobile-bed debris-flow dynamics starting from precipitation inputs. Nevertheless, simulating the fundamental physical processes is not only a mathematical and numerical issue, but also requires extensive field data for model calibration and validation. Consequently, the reliability of basin-scale debris-flow models strongly depends on the availability and quality of observational data. This study focuses on analysing available data from the Gadria catchment in South Tyrol (northern Italy) to assess their suitability for the calibration and validation of basin-scale debris-flow models. The Gadria catchment has been monitored for several years. In this area, the Civil Protection Agency of Bolzano, in collaboration with the Free University of Bozen-Bolzano, faced challenges posed by the mountainous environment, including ensuring a reliable power supply and internet connectivity for real-time data transmission. Since 2011, the catchment has been equipped with three rain gauge stations, three hydrometers, four geophone plates, and three webcams, mainly located in the lower part of the basin. This monitoring network has produced long time series of observations, which are potentially valuable for the calibration and validation of catchment-scale hydrological and debris-flow models, such as TRENT2D–MBRR. Unfortunately, as one might expect, the collected data are affected by several sources of uncertainty and inconsistency, as is commonly observed in remote mountain environments. These include measurement errors due to local conditions (e.g., wind effects on rain gauges), data gaps, time shifts among datasets, in the baseline uncertainties, and noise in hydrometer signals. Furthermore, overall data coherence is seldom achieved, and therefore, only a few events are reasonably usable. Finally, additional limitations arise from the lack of monitoring stations across the different initiation areas, preventing the identification of the specific tributaries activated during events and recorded in the main channel. In conclusion, despite the Gadria basin being a well-monitored catchment, the effective use of available data in advanced basin-scale debris-flow modelling remains challenging, and it requires accurate data assessment and consistency analyses before use.