This work purposes to evaluate the capabilities of the numerical model Trent2D to reproduce and forecast a real event of debris-flow. This aim has been pursued through a back analysis of a well-documented real event, occurred in August 2010 in the village of Campolongo (Province of Trento, Italy) as a result of torrential rains. The back analysis has been carried out starting from the application of a rainfall-runoff model, in order to obtain the liquid discharges of the event. Then the solid discharges have been estimated from measured volume of deposits. With these boundary conditions, a systematic calibration of model parameters has been performed, in order to find the optimal values that minimize the difference between computed and surveyed deposits. Results show that the depositional patterns can be reproduced quite well everywhere, except around a bridge, that is not currently simulated with the model. Moreover, the optimal estimate of parameters gives values comparable to those that can be obtained from a blind approach, in which parameters have been fixed according to an a priori assumption of local equilibrium, i.e. without considering any indication from the back analysis. These results give the model simulations a reliable forecasting capability that can be exploited to produce trustworthy hazard maps for urbanized fans that are commonly situated far from the triggering points of debris-flows, just as the fan of Campolongo.
Evaluation of the Trent2D Model Capabilities to Reproduce and Forecast Debris-Flow Deposition Patterns Through a Back Analysis of a Real Event
Rosatti, Giorgio;Zorzi, Nadia;Begnudelli, Lorenzo;Armanini, Aronne
2015-01-01
Abstract
This work purposes to evaluate the capabilities of the numerical model Trent2D to reproduce and forecast a real event of debris-flow. This aim has been pursued through a back analysis of a well-documented real event, occurred in August 2010 in the village of Campolongo (Province of Trento, Italy) as a result of torrential rains. The back analysis has been carried out starting from the application of a rainfall-runoff model, in order to obtain the liquid discharges of the event. Then the solid discharges have been estimated from measured volume of deposits. With these boundary conditions, a systematic calibration of model parameters has been performed, in order to find the optimal values that minimize the difference between computed and surveyed deposits. Results show that the depositional patterns can be reproduced quite well everywhere, except around a bridge, that is not currently simulated with the model. Moreover, the optimal estimate of parameters gives values comparable to those that can be obtained from a blind approach, in which parameters have been fixed according to an a priori assumption of local equilibrium, i.e. without considering any indication from the back analysis. These results give the model simulations a reliable forecasting capability that can be exploited to produce trustworthy hazard maps for urbanized fans that are commonly situated far from the triggering points of debris-flows, just as the fan of Campolongo.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione