Hydrological modeling on a regional scale: use of machine learning techniques for the calibration of models in ungauged sites
Reference persons ALBERTO VIGLIONE
Description The conceptual rainfall-runoff model of Viglione and Parajka (2020) is calibrated on a regional scale using the PASS approach explained in Merz et al. (2020, https://doi.org/10.1029/2019WR026008), taking into account the snowpack observed by the satellite (as in Parajka J. and G. Blöschl, 2008, https://doi.org/10.1016/j.jhydrol .2008.06.006).
The work aims to answer the following questions:
- what is the accuracy of the simulated flow rates for the monitored watercourses (of which we have data)?
- what is the accuracy of the simulated flow rates in non-instrumented basins if machine learning techniques such as "random-forest" or "decision trees" are used?
The student's propensity / availability for programming is required. Numerical software R will be used (which includes the hydrological model and machine learning methods).
Deadline 13/05/2023 PROPONI LA TUA CANDIDATURA